In the Name of Entrepreneurship? The Logic and Effects of Special Regulatory Treatment for Small Business

KAUFFMAN-RAND INSTITUTE FOR ENTREPRENEURSHIP PUBLIC POLICY THE ARTS

This PDF document was made available

CHILD POLICY

from www.rand.org as a public service of

CIVIL JUSTICE EDUCATION ENERGY AND ENVIRONMENT HEALTH AND HEALTH CARE INTERNATIONAL AFFAIRS NATIONAL SECURITY POPULATION AND AGING PUBLIC SAFETY SCIENCE AND TECHNOLOGY SUBSTANCE ABUSE TERRORISM AND HOMELAND SECURITY TRANSPORTATION AND INFRASTRUCTURE WORKFORCE AND WORKPLACE

the RAND Corporation. Jump down to document6

The RAND Corporation is a nonprofit research organization providing objective analysis and effective solutions that address the challenges facing the public and private sectors around the world.

Support RAND Purchase this document Browse Books & Publications Make a charitable contribution

For More Information Visit RAND at www.rand.org Explore Kauffman-RAND Institute

for Entrepreneurship Public Policy

View document details Limited Electronic Distribution Rights This document and trademark(s) contained herein are protected by law as indicated in a notice appearing later in this work. This electronic representation of RAND intellectual property is provided for non-commercial use only. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. Permission is required from RAND to reproduce, or reuse in another form, any of our research documents for commercial use. For information on reprint and linking permissions, please see RAND Permissions.

This product is part of the RAND Corporation monograph series. RAND monographs present major research findings that address the challenges facing the public and private sectors. All RAND monographs undergo rigorous peer review to ensure high standards for research quality and objectivity.

In the Name of

Entrepreneurship?

The Logic and Effects of Special Regulatory Treatment for Small Business Susan M. Gates, Kristin J. Leuschner Editors

Supported by the Ewing Marion Kauffman Foundation

KAUFFMAN-RAND INSTITUTE FOR ENTREPRENEURSHIP PUBLIC POLICY

The research described in this monograph was conducted by the Kauffman-RAND Institute for Entrepreneurship Public Policy (KRI), which is housed within the RAND Institute for Civil Justice (ICJ). KRI’s work is supported by a grant from the Ewing Marion Kauffman Foundation.

Library of Congress Cataloging-in-Publication Data In the name of entrepreneurship? : the logic and effects of special regulatory treatment for small business / Susan M. Gates, Kristin J. Leuschner, editors. p. cm. ISBN 978-0-8330-4204-0 (pbk. : alk. paper) 1. Small business—Government policy—United States. 2. Small business—Law and legislation—United States. I. Gates, Susan M., 1968– II. Leuschner, Kristin. HD2346.U5I546 2007 338.6'420973—dc22 2007030590

The RAND Corporation is a nonprofit research organization providing objective analysis and effective solutions that address the challenges facing the public and private sectors around the world. R AND’s publications do not necessarily reflect the opinions of its research clients and sponsors.

R® is a registered trademark.

Cover design by Eileen Delson La Russo

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from RAND. Published 2007 by the RAND Corporation 1776 Main Street, P.O. Box 2138, Santa Monica, CA 90407-2138 1200 South Hayes Street, Arlington, VA 22202-5050 4570 Fifth Avenue, Suite 600, Pittsburgh, PA 15213-2665 RAND URL: http://www.rand.org To order RAND documents or to obtain additional information, contact Distribution Services: Telephone: (310) 451-7002; Fax: (310) 451-6915; Email: [emailprotected]

Preface

The Kauﬀman-RAND Institute for Entrepreneurship Public Policy (KRI) was established in 2004 (as the Kauﬀman-RAND Center for Regulation and Small Business) to respond to the need to understand the impact of policy and regulation on small businesses. KRI’s initial objective was to evaluate and inform legal and regulatory policymaking related to small businesses and entrepreneurship through objective, rigorous, empirically based research. For the past three years, KRI activities have been devoted to improving understanding of business responses to regulatory policymaking. Our initial research eﬀorts over the past three years have focused on three broad questions: • What insights can be drawn from existing research on the impact of regulation and litigation on small businesses and entrepreneurship? • What data are available to support research on the impact of regulation and policy on small businesses, and what additional data are needed? • What do focused studies in selected policy areas reveal about the diﬀerential eﬀects of regulation and policy on small businesses? KRI has supported a number of research studies on diﬀerent topics. Speciﬁc topics were chosen to leverage existing RAND expertise, utilize readily available data sources, and develop new data sources. This book highlights some of the key ﬁndings from research eﬀorts supported by KRI and describes a road map for future work.

iii

In the Name of Entrepreneurship?

The RAND Institute for Civil Justice The RAND Institute for Civil Justice (ICJ) is an independent research program within the RAND Corporation. The mission of ICJ, a division of the RAND Corporation, is to improve private and public decisionmaking on civil legal issues by supplying policymakers and the public with the results of objective, empirically based, analytic research. ICJ facilitates change in the civil justice system by analyzing trends and outcomes, identifying and evaluating policy options, and bringing together representatives of diﬀerent interests to debate alternative solutions to policy problems. ICJ builds on a long tradition of RAND research characterized by an interdisciplinary, empirical approach to public policy issues and rigorous standards of quality, objectivity, and independence. ICJ research is supported by pooled grants from corporations, trade and professional associations, and individuals; by government grants and contracts; and by private foundations. ICJ disseminates its work widely to the legal, business, and research communities and to the public. In accordance with RAND policy, all ICJ research products are subject to peer review before publication. ICJ publications do not necessarily reﬂect the opinions or policies of the research sponsors or of the ICJ Board of Overseers. The Kauﬀman-RAND Institute for Entrepreneurship Public Policy (KRI), which is housed within ICJ, is dedicated to assessing and improving legal and regulatory policymaking as it relates to small businesses and entrepreneurship in a wide range of settings, including corporate governance, employment law, consumer law, securities regulation, and business ethics. KRI’s work is supported by a grant from the Ewing Marion Kauﬀman Foundation. For additional information on ICJ or KRI, please contact the directors:

Preface

Robert Reville, Director RAND Institute for Civil Justice 1776 Main Street P.O. Box 2138 Santa Monica, CA 90407-2138 310-393-0411 x6786 Fax: 310-451-6979 [emailprotected]

Susan Gates, Director Kauﬀman-RAND Institute for Entrepreneurship Public Policy 1776 Main Street P.O. Box 2138 Santa Monica, CA 90407-2138 310-393-0411 x7452 Fax: 310-451-6979 [emailprotected]

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi CHAPTER ONE

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Managing the Relationship Between Business and Society . . . . . . . . . . . . . . . . . . . 3 The Politics of the Legal and Regulatory Environments . . . . . . . . . . . . . . . . . . . . . . 7 Small Businesses and the Legal and Regulatory Environments . . . . . . . . . . . . . . 8 Improved Understanding of the Impact of Regulation on Small Businesses Is Needed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Overview of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 CHAPTER TWO

The Impact of Regulation and Litigation on Small Businesses and Entrepreneurship: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Lloyd Dixon, Susan M. Gates, Kanika Kapur, Seth A. Seabury, and Eric Talley Corporate and Securities Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Regulatory Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Eﬀects of Corporate and Securities Law on Small Businesses and the Small-Business Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Section Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Environmental Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

vii

viii

In the Name of Entrepreneurship?

Regulatory Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Eﬀects of Environmental Regulations on Small Businesses and the Small-Business Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Section Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Employment Law and Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Regulatory Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Eﬀects of Employment Regulations on Small Businesses and the Small-Business Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Section Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Health-Insurance Regulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Regulatory Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Eﬀects of Health-Insurance Regulations on Small Businesses and the Small-Business Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Section Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Chapter Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 CHAPTER THREE

State Health-Insurance Mandates, Consumer-Directed Health Plans, and Health Savings Accounts: Are They a Panacea for Small Businesses? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Susan M. Gates, Kanika Kapur, and Pinar Karaca-Mandic Small Businesses Typically Face Restricted Health-Insurance Options . . . . 70 State Health-Insurance Mandates Seek to Expand Small-Business Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Rating Reforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Guaranteed-Issue and Guaranteed-Renewal Reforms . . . . . . . . . . . . . . . . . . . . . . 73 Preexisting-Condition Limitation and Portability Reforms . . . . . . . . . . . . . . . 73 State Mandates Have Not Improved Small-Business Access to Health Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 State Health-Insurance Mandates Have Had Unintended Eﬀects . . . . . . . . . . 76 Consumer-Directed Health Plans Could Expand Options for Small Businesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Small Firms Have Not Been Especially Quick to Adopt ConsumerDirected Health Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Contents

Additional Evidence on the Use of Health Reimbursem*nt Arrangements, Health Savings Accounts, and High-Deductible Health Plans by Small Businesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Consumer-Directed Health Plan Utilization and Growth Do Not Vary by Firm Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Persistence in Consumer-Directed Health Plan Oﬀerings . . . . . . . . . . . . . . . . . . . 89 Which Firms Are Likelier to Oﬀer Consumer-Directed Health Plans? . . . 90 Longitudinal Analysis of Consumer-Directed Health Plan Oﬀerings. . . . . 94 Beneﬁt Design of Health Reimbursem*nt Arrangements and Health Savings Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Consumer-Directed Health Plans Are Growing in Popularity but Do Not Appear to Be a Panacea for Small Businesses . . . . . . . . . . . . . . . . . 101 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 CHAPTER FOUR

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 John Mendeloﬀ, Christopher Nelson, Kilkon Ko, and Amelia Haviland The Relationship Between Firm Size and Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Previous Research on Size and Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Fatalities and Other Serious Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Less Serious Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Accident Types and Size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Underreporting and Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Data and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Number of Fatalities (Numerator Data) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Exposure to the Risk of Death (Denominator Data) . . . . . . . . . . . . . . . . . . . . . 116 Regression Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Overall Patterns in the OSHA Fatality Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 The Relationship Between Fatality Rate and Establishment Size . . . . . . . 119 Looking at More Detailed Categories of Small Establishments. . . . . . . . . 121 Fatality Rates by Establishment Size in More Narrowly Deﬁned Industries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

In the Name of Entrepreneurship?

The Relationship Between Fatality Rate and Firm Size . . . . . . . . . . . . . . . . . . 123 The Relationship Between Fatality Rate, Firm Size, and Establishment Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Controls for Other Factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Causes of Fatalities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 OSHA Violations and Establishment Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Size Distribution of Nonfatal Injury and Accident Rates . . . . . . . . . . . . . . . . 133 Tracking the Pattern of Fatality Rates by Establishment Size Over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Implications for Policy and Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Key Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Study Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Policy Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 CHAPTER FIVE

Sarbanes-Oxley’s Eﬀects on Small Firms: What Is the Evidence? . . . . Ehud Kamar, Pinar Karaca-Mandic, and Eric Talley Introduction and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of the Sarbanes-Oxley Act of 2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Internal Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Management Certiﬁcation of Financial Statements . . . . . . . . . . . . . . . . . . . . . . Extended Statute of Limitations for Shareholder Lawsuits . . . . . . . . . . . . . . Executive Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Audit Committees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Separation of Audit and Nonaudit Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Special Case of Small Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evidence on Accounting and Audit Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evidence on Market Reactions and Firm Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evidence on Deregistrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proposals to Mitigate Sarbanes-Oxley’s Eﬀect on Small Firms . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

143 143 145 145 147 147 148 148 149 150 152 156 158 163 164

Contents

CHAPTER SIX

Do the Owners of Small Law Firms Beneﬁt from Limited Liability? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 John A. Romley, Eric Talley, and Bogdan Savych Importance of Liability for Small Professional Firms . . . . . . . . . . . . . . . . . . . . . . . 170 Potential Value of LLP and LLC Forms to Small Professional Firms . . . . 171 Existing Evidence on the LLP and LLC Forms Among Professional Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 Focus of This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Organizational Forms for Multiowner Law Firms . . . . . . . . . . . . . . . . . . . . . . . . . . 177 General Partnership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Professional Corporations and Professional Associations . . . . . . . . . . . . . . . . 177 Limited-Liability Partnerships and Limited-Liability Companies . . . . . . 179 Data and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 A New Data Set on U.S. Law Firms in the 1990s . . . . . . . . . . . . . . . . . . . . . . . . 181 Organizational Structure of the U.S. Legal-Service Industry, 1993 and 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Analysis of Size and Reorganization Among General Partnerships. . . . . 185 Analysis of Size, Reorganization, and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 Size and Reorganization Among General Partnerships . . . . . . . . . . . . . . . . . . 194 Size, Reorganization, and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 CHAPTER SEVEN

Data Resources for Policy Research on Small Businesses. . . . . . . . . . . . . . 205 Amelia Haviland and Bogdan Savych Assessing Data Set Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Available Size Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Source of Data: Employer or Firm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Access to Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Data Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Ability to Link Data Longitudinally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 U.S. Bureau of Labor Statistics Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 Quarterly Census of Employment and Wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Business Employment Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

xii

In the Name of Entrepreneurship?

Current Employment Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 National Compensation Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 Current Population Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 U.S. Census Bureau Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Standard Statistical Establishment Listing or Business Register . . . . . . . . 216 Longitudinal Business Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 County Business Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 Business Information Tracking System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 Economic Census and Company Organization Survey . . . . . . . . . . . . . . . . . . 217 Surveys of Women- and Minority-Owned Business Enterprises, Characteristics of Business Owners, and Survey of Business Owners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Integrated Longitudinal Business Database and Longitudinal Employer-Household Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 Other Government Sources of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 Survey of Small Business Finances. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 National Employer Health Insurance Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Medical Expenditures Panel Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Private and Commercially Available Data Sources. . . . . . . . . . . . . . . . . . . . . . . . . . 221 DUNS Market Identiﬁer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Kauﬀman Firm Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 Research Data Set Derived from the Martindale-Hubbell Law Directory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 The Henry J. Kaiser Family Foundation/Health Research and Educational Trust Employer Health Beneﬁts Surveys. . . . . . . . . . . . . . . 223 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 CHAPTER EIGHT

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Policies Designed Speciﬁcally to Help Small Businesses Do Not Always Have the Intended Eﬀect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 The Regulatory Environment’s Eﬀect on Small-Firm Behavior Diﬀers from Its Eﬀect on Large Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 What Exactly Is a Small Firm? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Further Research Is Needed to Support Entrepreneurship Public Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247

Contents

xiii

Corporate and Securities Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Environmental Regulation and Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Employment Law and Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Health-Insurance Regulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Concluding Thoughts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 APPENDIXES

A. Criteria Used to Deﬁne Small Business in Determining Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Ryan Keefe, Susan M. Gates, and Eric Talley B. Methodology for Analysis of Small Businesses and Workplace Fatality Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 John Mendeloﬀ, Christopher Nelson, Kilkon Ko, and Amelia Haviland C. Regression Analysis for Analysis of Small Businesses and Workplace Fatality Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 John Mendeloﬀ, Christopher Nelson, Kilkon Ko, and Amelia Haviland D. Firms’ Reasons to Go Private or Go Dark After Sarbanes-Oxley . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Ehud Kamar, Pinar Karaca-Mandic, and Eric Talley References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

Figures

1.1. 4.1. 4.2. 4.3. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9.

6.10.

6.11.

Conceptual Model of the Regulatory and Legal Reform Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Fatality Rate, by Establishment Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Fatality Rate, by Establishment Size: Small Establishments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Fatality Rate, by Firm Size, All States, 1992–2001 . . . . . . . . . . . . 124 Total Revenues of Law and Accounting Firms in 1997, by Firm Revenue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Employees of Law and Accounting Firms in 1997, by Firm Revenue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 States Authorizing LLP and LLC Forms for Law Firms, by Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Share of Law Firms, by Organizational Form in 1993 and 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Organizational Forms of 1993 General Partnerships in 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 Organizational Forms of 1999 LLPs in 1993. . . . . . . . . . . . . . . . . . . 187 Organizational Forms of 1999 LLCs in 1993 . . . . . . . . . . . . . . . . . . 188 Share of General Partnerships in 1993, by Size Class . . . . . . . . . 189 Average Rates of Reorganization of General Partnerships Under New Limited-Liability Forms Within States and by Year of First Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Rate of Reorganization of General Partnerships Under the LLP and LLC Forms, as of 1999, by Number of Lawyers in 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Predicted Eﬀect of Increasing Size Class on the Rate of Reorganization, by Number of Lawyers in 1993. . . . . . . . . . . . . . . 198

xvi

In the Name of Entrepreneurship?

6.12. 6.13.

C.1.

Growth, by Reorganization Under New LimitedLiability Form as of 1999 and Size Class in 1993 . . . . . . . . . . . . . . 200 Predicted Growth of New York Firms, by Reorganization Under New Limited-Liability Form as of 1999 and Size Class in 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Poisson Regression Coeﬃcients and Their Conﬁdence Intervals, by Each Size Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

Tables

2.1. 2.2. 2.3. 3.1. 3.2. 3.3. 3.4. 3.5. 3.6. 3.7. 4.1.

4.2. 4.3. 4.4. 4.5.

Options for Business Organizational Structure . . . . . . . . . . . . . . . . . 22 Major Federal Environmental Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Major Federal Regulations Governing Health Insurance . . . . . . 56 State Counts by Reform Level and Year . . . . . . . . . . . . . . . . . . . . . . . . . 77 State Counts by Firm-Size Upper Limit for State SmallGroup Health-Insurance Reform, by Year . . . . . . . . . . . . . . . . . . . . . . . 77 Descriptive Proﬁle of Health Insurance Oﬀerors, by Firm Size and Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Determinants of Consumer-Directed Health Plan Oﬀerings: Estimates from Logit Model (2003–2005) . . . . . . . . . 92 Determinants of Consumer-Directed Health Plan Oﬀerings in 2005: Estimates from a Logit Model . . . . . . . . . . . . . . . 95 Beneﬁt Design of Health Reimbursem*nt Arrangement Plans (OLS Regressions, 2005) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Beneﬁt Design of Health Savings Account Plans (OLS Regressions, 2005) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Factors Inﬂuencing the Predicted Eﬀects of Establishment and Firm Size on Safety: Marginal Beneﬁts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Factors Inﬂuencing the Predicted Eﬀects of Establishment and Firm Size on Safety: Marginal Costs . . . . . 111 OSHA-Investigated Fatalities in Each Establishmentand Firm-Size Category, All States, 1992–2001 . . . . . . . . . . . . . . . 118 Nonconstruction Fatalities Investigated by OSHA, All States, 1992–2001, by Establishment and Firm Size . . . . . . . . . . 119 Fatalities and Fatality Rate, by Establishment and Firm Size, Manufacturing Sector, 1992–2001 . . . . . . . . . . . . . . . . . . . . . . . . 125

xvii

xviii

In the Name of Entrepreneurship?

4.6.

4.7. 4.8.

4.9.

5.1. 5.2. 5.3. 6.1. 6.2. 6.3. 7.1.

Fatalities and Fatality Rate, by Establishment and Firm Size, Manufacturing Sector, Not Including Logging (SIC 241), 1992–2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 The Relative Frequency of Diﬀerent Events in Fatalities in Establishments with 1 to 19 Employees, by Sector . . . . . . . . . 129 Deaths Associated with Citations of Serious Violations, by Establishment-Size Category, Selected Sectors, 1992–2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Reported California Employee Deaths and Hospitalizations at Establishments with 1 to 19 Employees, 1992–2001, by Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Median Audit Fees as a Percentage of Revenues . . . . . . . . . . . . . . . 154 Primary Reasons Cited for Deregistration . . . . . . . . . . . . . . . . . . . . . . 160 Summary of the Studies Reviewed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Important Features of Organizational Forms for Multiowner Law Firms from an Owner’s Perspective . . . . . . . . . 178 Results for Multinomial-Logit Model of Organizational Form in 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Results for OLS Regression of Growth . . . . . . . . . . . . . . . . . . . . . . . . . 201 Summary of the Available Data Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

Acknowledgments

Many people both within and outside RAND contributed to this book. We are grateful for the input of the numerous experts on small business and public policy who participated in the Kauﬀman-RAND symposia in fall 2004 and fall 2005, contributing to the research agenda for KRI. We are particularly indebted to John Graham, who reviewed the entire book manuscript and provided extensive comments and suggestions that have dramatically improved the book. We are also grateful for the comments and feedback received from Robert Litan of the Kauﬀman Foundation and Robert Reville of RAND on earlier versions of the book manuscript. In addition to the review of the entire book, at least two people, both internal and external to RAND, also reviewed each substantive chapter. We wish to thank all of these individuals, whose reviews improved each study. At RAND, these reviewers included Anthony Bower (now at Amgen), Richard Buddin, Debra Knopman, John Russer, Michael Greenberg, Carole Gresenz, Pinar Karaca-Mandic, Jacob Klerman (now at Abt Associates), and Darius Lakdawalla. External reviewers included Scott Baker of the University of North Carolina at Chapel Hill, R. Preston McAfee of Yahoo! Research, Larry Ribstein of the University of Illinois College of Law, and Mark Showalter of Brigham Young University. Of course, we are responsible for the ﬁnal volume. All authors would like to thank our RAND colleagues, especially Donna White, who pulled together the ﬁnal manuscript and managed to keep track of each chapter through the writing and review process.

xix

In the Name of Entrepreneurship?

Nancy Good, Michelle Platt, and Donna White provided administrative support, Lisa Bernard carefully edited the report, and James Torr proofread the ﬁnal copy. This book would not have been possible without generous support from the Ewing Marion Kauﬀman Foundation. We are especially grateful to Carl Schramm and to our project monitor, Robert Litan, for their insightful input to the activities of KRI leading up to this book and for their commitment to research on issues related to entrepreneurship public policy.

Abbreviations

ADA

Americans with Disabilities Act of 1990

ADEA

Age Discrimination in Employment Act of 1967

AHIP

America’s Health Insurance Plans

AHRQ

Agency for Healthcare Research and Quality

BAPCPA

Bankruptcy Abuse Prevention and Consumer Protection Act of 2005

BED

Business Employment Dynamics

BITS

Business Information Tracking System

BLS

U.S. Bureau of Labor Statistics

Business Register

CBO

Characteristics of Business Owners

CBP

county business pattern

CDC

Centers for Disease Control and Prevention

CDHP

consumer-directed health plan

CERCLA

Comprehensive Environmental Response, Compensation, and Liability Act

CES

Current Employment Statistics

CESQG

conditionally exempt small-quantity generators

xxi

xxii

In the Name of Entrepreneurship?

CFN

census ﬁle number

CFOI

National Census of Fatal Occupational Injuries

COBRA

Consolidated Omnibus Budget Reconciliation Act of 1985

COS

Company Organization Survey

CPS

Current Population Survey

CSI

Common Sense Initiative

D&B

Dun and Bradstreet

DMI

DUNS® Market Identiﬁer

DOL

U.S. Department of Labor

Economic Census

EEOC

U.S. Equal Employment Opportunity Commission

EIN

employer identiﬁcation number

EPA

U.S. Environmental Protection Agency

ERISA

Employee Retirement and Income Security Act of 1974

FAT/CAT

fatality/catastrophe

FDA

U.S. Food and Drug Administration

FEI

Financial Executives International

FIFRA

Federal Insecticide, Fungicide, and Rodenticide Act

FSA

ﬂexible spending account

GAO

U.S. Government Accountability Oﬃce

general partnership

household component

Abbreviations

xxiii

HDHP

high-deductible health plan

HDHP/SO

high-deductible health plan with a savings option

health insurance

HIPAA

Health Insurance Portability and Accountability Act

HRA

health reimbursem*nt arrangement

HRET

Health Research and Educational Trust

HSA

health savings account

HSE

health and safety executive

insurance component

ILBD

Integrated Longitudinal Business Database

IMIS

Integrated Management Information System

KFF

Henry J. Kaiser Family Foundation

KFS

Kauﬀman Firm Survey

KRI

Kauﬀman-RAND Institute for Entrepreneurship Public Policy

LBD

Longitudinal Business Database

LEEM

Longitudinal Establishment and Enterprise Microdata

LEHD

Longitudinal Employer-Household Dynamics

MEPS

Medical Expenditures Panel Survey

Martindale-Hubbell

MHPA

Mental Health Parity Act of 1996

MPC

medical-provider component

MSA

medical savings account

xxiv

In the Name of Entrepreneurship?

MSD

musculoskeletal disorder

NAICS

North American Industrial Classiﬁcation System

NCS

National Compensation Survey

NEHIS

National Employer Health Insurance Survey

NEPA

National Environmental Policy Act

NFIB

National Federation of Independent Businesses

NHC

nursing-home component

NLRA

National Labor Relations Act of 1935

NLRB

National Labor Relations Board

NMHPA

Newborns’ and Mothers’ Health Protection Act of 1996

OLS

ordinary least squares

OSH

Occupational Safety and Health

professional association

professional corporation

PCAOB

Public Company Accounting Oversight Board

PCV

piercing the corporate veil

PDA

Pregnancy Discrimination Act of 1978

PLC

professional limited company

PPN

permanent plant number

PRA

Paperwork Reduction Act of 1995

QCEW

Quarterly Census of Employment and Wages

RCRA

Resource Recovery and Conservation Act

RDC

Research Data Center

Abbreviations

RFA

Regulatory Flexibility Act of 1980

S&P

Standard and Poor’s

SARA

Superfund Amendments and Reauthorization Act

SBA

U.S. Small Business Administration

SBIR

Small Business Innovation Research

SBO

Survey of Business Owners

SBPRA

Small Business Paperwork Relief Act of 2002

SBREFA

Small Business Regulatory Enforcement Fairness Act of 1996

SDWA

Safe Drinking Water Act

SEC

U.S. Securities and Exchange Commission

SESA-ID

state employment security agency identiﬁcation number

SIC

standard industrial classiﬁcation

SIPP

Survey of Income and Program Participation

SMOBE

Survey of Minority-Owned Business Enterprises

SOI

Statistics of Income

SOX

Sarbanes-Oxley Act

sole proprietorship

SSBF

Survey of Small Business Finances

SSEL

Standard Statistical Establishment Listing

SWOBE

Survey of Women-Owned Business Enterprises

TAMRA

Technical and Miscellaneous Revenue Act of 1988

TSCA

Toxic Substances Control Act

xxv

xxvi

In the Name of Entrepreneurship?

UCFE

Unemployment Compensation for Federal Employees

unemployment insurance

WARN

Worker Adjustment and Retraining Notiﬁcation

workers’ compensation

WHCRA

Women’s Health and Cancer Rights Act of 1998

WIRS

Workplace Industrial Relations Survey

CHAPTER ONE

Introduction

Small businesses are an important feature of the U.S. political and economic landscape. Small businesses (deﬁned as ﬁrms with fewer than 500 employees) account for almost half of all gross revenues generated by U.S. businesses, employ half of all private-sector workers, and generate between 60 and 80 percent of net new jobs. Entrepreneurship is generally viewed as an engine of technological progress and economic growth. It is also an important element of the American dream: a means for those who have little more than ambition and a good idea to improve their lot in life. Not surprisingly, the interests of small businesses and entrepreneurs are frequently mentioned in policy debates. Every day, policymakers at the federal, state, and local levels make decisions that have important implications for the livelihoods of entrepreneurs and small-business owners. There is ongoing concern that some regulations, rules, and government policies place a disproportionate burden on small businesses. For example, there is empirical evidence that some regulations place higher costs of compliance on small businesses than on other entities, since many of these costs do not vary by ﬁrm size and are incurred on an ongoing (rather than one-time) basis (Bradford, 2004). The tort system can aﬀect small businesses differently as well, though the precise nature of that eﬀect is less clear: In some cases, customers, employees, and government agencies might be less likely to punish or sue small businesses, because the perceived payoﬀ is low; on the other hand, customers or employees might be likelier to sue a small business if they perceive that it cannot aﬀord to mount an eﬀective legal defense.

In the Name of Entrepreneurship?

For both economic and political reasons, policymakers at the local, state, and federal levels have an interest in promoting, or at least not getting in the way of, small businesses and entrepreneurs. As a result, the special concerns of small businesses often receive prominent attention in the laws, regulations, requirements, and programs that result from those decisions. For example, food producers with fewer than 100 full-time–equivalent employees are exempt from the U.S. Food and Drug Administration’s (FDA’s) nutrition-labeling requirements on food products that have U.S. sales of fewer than 100,000 units per year (FDA, 1999). The Massachusetts Health Care Reform Plan requires all employers with more than 10 employees either to provide healthinsurance (HI) coverage or to pay into a statewide fund (Commonwealth of Massachusetts, 2006, section 188[b]). Small public companies, those with a market capitalization of $75 million or less, have been granted four extensions to the compliance deadline, now December 15, 2007, for section 404 of the Sarbanes-Oxley Act (SOX) (P.L. 107-204). This section of the act requires ﬁrms to issue a report assessing the eﬀectiveness of internal controls on ﬁnancial reporting. These are but a few of the countless examples of special regulatory treatment received by small businesses. The desire to support small businesses can come into conﬂict with the desire to address the other social concerns that led to regulation in the ﬁrst place. This can occur, for example, when the cost of compliance places an excessive burden on small businesses or when regulation is ineﬀective in attaining its purpose due to exemptions made for small businesses. Other questions arise with regard to the eﬀect of special regulatory treatment designed to help small businesses. Why and under what circ*mstances does special regulatory treatment for small businesses occur? Why does it take the speciﬁc form it takes? What objectives is it designed to serve, and how eﬀective is it in achieving these aims? The goal of this book is to begin to shed light on the ways in which the legal and regulatory environments aﬀect small businesses— both in terms of any diﬀerential eﬀect of regulation and policy on small businesses compared to large ones and in terms of the impact of the special regulatory treatment aﬀorded to small businesses. We sum-

Introduction

marize ﬁndings from the ﬁrst three years of research eﬀort within the Kauﬀman-RAND Institute for Entrepreneurship Public Policy (KRI). KRI’s research focus to date has been on describing the legal and regulatory environments to examine the eﬀects of speciﬁc policies and regulations on small businesses. This introduction sets the stage for the book by describing the role of regulation in managing the relationship between business and society, illustrating how small businesses ﬁt into that landscape, and emphasizing the limitations in our knowledge of the ways in which regulation aﬀects small businesses.

Managing the Relationship Between Business and Society We begin with a brief discussion of the evolution of business regulation and policy in the United States. Although regulations and the litigation system were typically designed with large businesses in mind, they nevertheless aﬀect small businesses in both direct and indirect ways. To understand their eﬀects on small businesses, it is necessary to understand the general approaches for managing the relationship between business and society and how those approaches have evolved over time. The perceived need to manage or shape the relationship between business and society stems primarily from concerns about the eﬀects of large businesses on society: high or predatory prices charged by monopolists, pollution emitted by large manufacturers, unfair labor practices exercised by large employers, or large manufacturers’ ability to subvert the tort system designed to protect consumers. Government regulation and private litigation provide two important mechanisms for shaping the relationship between businesses and society in the United States. These mechanisms are used to address two overriding public objectives: (1) to promote market competition and control large ﬁrms’ market power over customers and smaller ﬁrms and (2) to mitigate or prevent the adverse eﬀects of business activity (negative externalities) on individuals, organizations, and the environment. The ﬁrst objective is addressed through federal and state antitrust regulation and liti-

In the Name of Entrepreneurship?

gation over anticompetitive practices. The second objective is tackled through an expansive array of environmental, securities, employment, health, and safety regulation and the tort system. Regulation assumed an important role in the U.S. economic landscape during the Progressive Era of the late 1800s and early 1900s with the passage of legislation such as the Sherman Act (26 Stat. 209), the Interstate Commerce Act (P.L. 49-41), and the Pure Food and Drug Act (P.L. 59-384) and the establishment of federal regulatory agencies such as the FDA (P.L. 59-384), the Interstate Commerce Commission (P.L. 49-41), and FTC (38 Stat. 717). Prior to this, private litigation was the primary means of resolving conﬂicts involving businesses, and the government had little capacity to intervene in such conﬂicts. Glaeser and Shleifer (2003) attribute the ascendance of government regulation in the Progressive Era to economic industrialization and the ensuing concentration of economic wealth in the hands of large ﬁrms. They argue that large, wealthy ﬁrms have much to lose in legal contests and that the cost of inﬂuencing the process through bribery and intimidation was relatively low. In other words, large ﬁrms had both a strong incentive and the ability to subvert the legal system for their own beneﬁt. Although, as we discuss below, the regulatory system is not immune to such inﬂuence either, it may be less vulnerable than the legal system is. By this argument, regulation emerged from societal frustration with the legal system’s ability to control the behavior of large businesses and from a sense that regulation could do a better job—at least in some spheres. Regulations targeted large ﬁrms, and little consideration was given to the fact that smaller ﬁrms might be aﬀected. Since the Progressive Era, the regulatory environment facing businesses has grown inﬁnitely more complex. Typically, new regulations and regulatory agencies are added to what already exists, although occasionally regulations are repealed, deregulation occurs, and regulatory agencies close down. At the federal level, new waves of regulation have typically corresponded to heightened public concerns about wideranging social, consumer product, health, HI, safety, and environmental issues. The Securities Act of 1933 (P.L. 73-22) and the Securities Exchange Act of 1934 (P.L. 73-291) established disclosure requirements designed to protect investors in the wake of the stock-market

Introduction

collapse and the Great Depression. The Securities Exchange Act led to the establishment of the U.S. Securities and Exchange Commission (SEC), which has responsibility for enforcing federal securities laws and regulating the industry. The Fair Labor Standards Act of 1938 (P.L. 75-718), which came into eﬀect at the end of the Great Depression, guaranteed a minimum wage and regulated the use of child labor and overtime pay. The Great Society Era saw the introduction of the Civil Rights Act of 1964 (P.L. 88-352) and several other antidiscrimination laws that regulated the employment relationship and established the U.S. Equal Employment Opportunity Commission (EEOC) (42 U.S.C. 2000e). That era also set the stage for a wave of federal environmental and safety regulation, starting with the Air Quality Act of 1967 (P.L. 90-148) and the establishment of the U.S. Environmental Protection Agency (EPA) (Nixon, 1970) and OSHA (P.L. 91-596) in 1970. In the 1980s and 1990s, regulations on employer-sponsored HI emerged in response to rising HI costs and concerns about access to group coverage; these included parts of the Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA) (P.L. 99-272) and the Health Insurance Portability and Accountability Act (HIPAA) (P.L. 104-191). HI regulation, and even the idea of a national, universal, health-care system, remains a subject of consideration well into the new millennium. Finally, in response to the corporate scandals of the late 1990s and early 2000s, SOX created the Public Company Accounting Oversight Board (PCAOB) and introduced additional auditing and documentation requirements for publicly traded ﬁrms. The complexity of the regulatory environment stems not only from the addition of new regulations and regulatory agencies on top of old ones but also from the absence of mechanisms capable of coordinating regulation across substantive areas at the federal level. The situation is further complicated by the layering of federal, state, and local regulations, as well as by a lack of coordination across governmental levels and across regulatory areas. While some federal regulations, such as the Employment Retirement and Income Security Act of 1974 (ERISA) (P.L. 93-406), preempt state regulations in the same regulatory area, in other areas, it is not uncommon for state and even local governments to impose regulations that are more stringent or broader than the fed-

In the Name of Entrepreneurship?

eral regulation. For example, many states and cities have established minimum-wage laws that set a higher wage than the federal minimum wage. In addition, state and local government can impose regulations in areas that the federal government does not regulate. Thus, a single business operating in several states and localities might need to keep abreast of thousands of regulations. A 1996 report by the U.S. General Accounting Oﬃce highlighted the day-to-day implications of the complex regulatory environment and the fragmented nature of information on regulatory responsibilities. In the study, the oﬃce attempted to thoroughly document the federal regulatory burden faced by 15 U.S. companies. Many ﬁrms that were contacted to participate in the study declined because they did not have the type of information for which the oﬃce was looking. Even those that did agree to participate could not provide the oﬃce with a complete list of all federal regulations that applied to them. In particular, ﬁrms had trouble separating federal regulations from state and local ones. Further, the oﬃce found that federal regulatory agencies themselves could often not determine whether a regulation would apply to a particular ﬁrm. The study found great concern among ﬁrms about the lack of coordination across agencies and across government jurisdictions. It is important to recognize that this regulatory complexity is further layered atop the risk of litigation. The risk of private litigation typically remains in spite of government laws. Government can choose to play an active role in enforcing laws that govern business behavior—for example, by establishing an agency with the power to monitor and impose sanctions. However, governments can also rely on the legal system for enforcement. Particularly in the absence of formal regulatory enforcement, government laws can increase a ﬁrm’s legal exposure. This brief summary of the roles of regulation and litigation in managing the relationship between business and society highlights several potentially salient issues for small businesses. First, most of the regulation that applies to small businesses was developed with larger businesses in mind. Secondly, the legal and regulatory environments that have emerged over the past century are extremely complex. The

Introduction

number of laws and the variation in how they are applied make it difﬁcult for small businesses to comply. Finally, the fragmented nature of regulation, with a lack of coordination across substantive areas and across jurisdictions, contributes to the diﬃculty in complying.

The Politics of the Legal and Regulatory Environments The regulatory landscape for businesses is made more complex due to its politicization. Large businesses, in particular, are an important and potentially powerful interest group. Politicization has diﬀerent implications for large businesses and small businesses. The “iron triangle” theory of policymaking and the theory of “regulatory capture” (Stigler, 1971; Peltzman, 1976) suggest that the ﬁrms in a regulated industry will use whatever means available (e.g., lobbying; provision of electoral support, information, and expertise) to inﬂuence the regulatory design for their own beneﬁt.1 Firms use their inﬂuence to make sure that legislators and bureaucrats take their objectives into account in the design of regulation. In certain contexts, ﬁrms might actually prefer a regulated environment, because government regulation enhances ﬁrms’ ability to coordinate with one another to ﬁx prices, deter the entry of new ﬁrms, and reap targeted beneﬁts from government (Kolko, 1970). Inﬂuence can extend beyond the features of the regulation itself to include procedural rules for the regulatory agency in charge of implementing the regulation (McCubbins, Noll, and Weingast, 1987, 1989). These rules are often designed to prevent “regulatory drift” by limiting the discretion that regulatory agencies might otherwise exercise in implementing the regulation. Firms in the regulated industry may have deeper understanding of practical implications of these rules ex ante and their ultimate implications for regulatory outcomes. In the end, regulatory agencies can be substantially aﬀected and constrained by the political inﬂuence 1

Iron triangle is a term used to describe the relationships among interest groups, the U.S. Congress, and the bureaucracy (i.e., federal agencies) in the policymaking process. Regulatory capture refers to a phenomenon in which the targets of regulation can inﬂuence or direct the actions of the government agencies that are responsible for enforcing the regulation.

In the Name of Entrepreneurship?

of interest groups (Wilson, 1989) and by information asymmetries and the knowledge that ﬁrms in the regulated industry possess (Laﬀont and Tirole, 1991). As we mentioned earlier, it has been argued that regulation emerged, in part, out of societal frustration with the legal system’s ability to control the behavior of large businesses and out of a sense that regulation could do a better job—at least in some spheres. This section has illustrated that the regulatory system is open also to inﬂuence from interest groups. A relevant question, then, is whether larger ﬁrms would be in a better position than smaller ﬁrms are to exert that type of inﬂuence over the regulatory environment. We turn to that question next.

Small Businesses and the Legal and Regulatory Environments As described in the preceding section, most regulations were put in place in response to concerns about the eﬀects of businesses— primarily large businesses—on society. Given the importance of small businesses in society, it is crucial to consider how the legal and regulatory environments inﬂuence small businesses and the ways in which those inﬂuences diﬀer from those on large businesses. The public concerns that generate demand for regulation or legal reform are usually signiﬁcant ones, and there are many interest groups representing various sides of these issues. As a practical matter, the speciﬁc concerns of small businesses add rarely more than a footnote to the legislative history of reform. However, that is not to say that the concerns of small businesses are ignored. Small businesses do receive a variety of special considerations— particularly in the regulatory context. This approach is often referred to as tiering. As will be described more fully in Chapter Two of this book, these special considerations vary tremendously across speciﬁc regulations. In some cases, small businesses receive a complete exemption from regulations. In other instances, small businesses receive special consideration in regulatory enforcement, support programs designed

Introduction

to assist them in complying, or delays in the application of new regulations. The deﬁnition of a small business can vary dramatically by regulatory context—typically ranging from two to 500 or more employees. Regulations can also use other measures of ﬁrm size, however, such as gross receipts. Regulatory thresholds inherent in a tiering approach should logically create incentives for ﬁrms that are near the threshold to restrain their growth because, if they go above the threshold, they will suddenly be subject to a regulation from which they used to be exempt. If regulatory compliance is costly and something to be avoided, then such thresholds should have behavioral eﬀects. This is an issue we discuss in greater detail in this book. The theory of collective action (Olson, 1965) suggests that large ﬁrms would be likeliest to exert inﬂuence over the regulatory and legal process. Not only do they have greater ﬁnancial and political resources available than do small businesses, but they also may have more to gain or lose from the regulatory process. On the one hand, small ﬁrms might beneﬁt from the activities of larger ﬁrms. To the extent that the interests of small ﬁrms are well aligned with the interests of larger ﬁrms in an industry, the small ﬁrms can “free ride” on the inﬂuence or activities of the larger ﬁrms. Smaller ﬁrms can also beneﬁt from whatever compromises the larger ﬁrms can win in the political process—for example, in terms of the stringency of the regulation or delays in adoption. However, the interests of small ﬁrms may also diﬀer from those of larger ﬁrms in important ways. For example, a major concern of small businesses is that compliance with regulations can impose substantial ﬁxed costs of operation that create barriers to new entry or increase the minimum eﬃcient scale in an industry. Large ﬁrms are unlikely to be as concerned about such issues, and, indeed, might view such costs or entry barriers as advantageous in the long run, because they reduce potential competitors’ ability to enter the market. The small-business community long ago recognized a need to represent its interests in a coordinated way across industries. The National Federation of Independent Businesses (NFIB) was founded in 1943, and other trade associations subsequently emerged to represent the common interests of small businesses in the political process. Today, the small-business lobby is an important player on the political land-

In the Name of Entrepreneurship?

scape. Initially, this lobby was focused on small-business assistance programs and initiatives to ensure that small businesses received a fair share of government contracts (e.g., the Small Business Act of 1953 [P.L. 83-163]). However, interest in regulatory issues increased significantly with the proliferation of regulation in the 1960s and 1970s, coupled with federal agencies’ tendency to adopt a one-size-ﬁts-all approach to regulation that often involved extensive reporting requirements or arbitrary and piecemeal regulatory tiering. The small-business lobby sought a greater voice for small businesses in the regulatory rulemaking process and a reduction in reporting burdens across the board, as well as assistance with regulatory compliance. As early as 1974, legislation that would ultimately lead to the Regulatory Flexibility Act of 1980 (RFA) (P.L. 96-354) was being introduced in Congress.2 The RFA requires federal agencies to consider carefully whether proposed rules will have a “signiﬁcant economic impact on a substantial number of small entities.” Unless the agency certiﬁes that a rule will not have such an impact, it must conduct a regulatory-ﬂexibility analysis. The regulatory-ﬂexibility analysis focuses on the expected eﬀects of a proposed rule on small entities and requires a public commentary period, a description of alternatives considered, and a justiﬁcation for the ﬁnal rule that was adopted. Tiering is just one of several approaches that agencies can consider for mitigating the eﬀect of a regulation on small businesses. Also passed in 1980, the Paperwork Reduction Act (PRA) (P.L. 96-511), which was designed to reduce reporting requirements and improve the management of information provided to the federal government by centralizing authority over information collection to the Oﬃce of Management and Budget of the Executive Oﬃce of the President of the United States of America (OMB). Although the RFA was considered to be a step in the right direction in terms of providing small businesses with a voice in the regulatory process, its limitations were quickly recognized. For example, it was fairly easy for government agencies to simply ignore the act or to certify that a rule would not substantially inﬂuence small entities (Verkuil, 2

See Verkuil (1982) for a detailed legislative history and description of the RFA.

Introduction

1982; Holman, 2006). The Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA) (P.L. 104-121) amended the RFA to impose speciﬁc requirements on federal agencies in the interest of improving compliance with the act. Executive Order 13272 (signed in August 2002) (Bush, 2002) requires federal regulatory agencies to develop written procedures for implementing the RFA. The NFIB legal foundation remains active in eﬀorts to ensure that federal agencies live up to RFA requirements and in promoting RFA-type legislation at the state (MRP, 2002) and local levels. In response to a growing wave of government regulation at the federal, state, and local levels, small-business interests have organized to represent their interests in the policymaking process. The prevalence of regulatory tiering, as well as legislation such as RFA (P.L. 96-354), suggest that small businesses have had some inﬂuence over regulatory policymaking. Nevertheless, questions remain as to the eﬀectiveness of that inﬂuence.

Improved Understanding of the Impact of Regulation on Small Businesses Is Needed Figure 1.1 provides a conceptual model of the regulatory and legal reform process as discussed in this chapter. As shown in the ﬁgure, regulation or legal reform originates with public concern regarding the impact of some business action on employees, customers, other individuals or organizations, or the physical environment or ecological resources. A new regulatory environment is then created or the existing legal or regulatory environment is modiﬁed to address these concerns. This environment, which includes regulations, laws, and enforcement mechanisms, may aﬀect diﬀerent ﬁrms in diﬀerent ways, whether intentionally or unintentionally. These business responses to the legal and regulatory environments lead to economic and other outcomes (e.g., social, environmental). Once this chain of events has played out,

In the Name of Entrepreneurship?

Figure 1.1 Conceptual Model of the Regulatory and Legal Reform Process

Public concern

Regulatory environment

Outcomes

Business response

RAND MG663-1.1

the outcomes may feed into new public concerns and lead to additional changes in the regulatory environment as part of an ongoing cycle. With RFA (P.L. 96-354), SBREFA (P.L. 104-121), and similar laws at the state level, policymakers have clearly recognized a need to consider the eﬀect of regulation on small businesses. However, the information required to make this assessment and to develop reasonable policy alternatives appropriate to small businesses is sorely lacking. Unfortunately, there is little quantitative evidence to demonstrate the speciﬁc impacts of policies and regulations on small businesses; nor has there been much evidence showing whether rules and exemptions designed to beneﬁt small businesses actually have that eﬀect. Additionally, there is little evidence that small-business exemptions are crafted in a way that appropriately balances the costs and beneﬁts of regulation. We currently know little about exactly when and under what circ*mstances it makes sense for policymakers to institute dif-

Introduction

ferential legal treatment—or wholesale retrenchment from regulatory intervention—based on ﬁrm size. Much of what we know about the interactions among regulation, litigation, and business stems from research that looks at the implications of regulations for large ﬁrms. The greater emphasis given in research to the impact of regulations on larger ﬁrms is understandable. Larger ﬁrms tend to receive the most public exposure, both in the popular press and by word of mouth. Moreover, the policy goal of mitigating the adverse eﬀects of business activity on other individuals or organizations is geared speciﬁcally toward perceived problems generated by large ﬁrms, since the capability of inﬂicting economic and social harms typically increases with ﬁrm size. And ﬁnally, because larger businesses are more frequently subject to reporting and disclosure requirements, they are a much more fertile harvesting ground for empirical data. However, the importance of large businesses does not negate the need to understand the impact of regulation on smaller businesses. Excluding such a signiﬁcant component of economic activity from the landscape of informed policy debate is both risky and imprudent. This book takes an important step in the direction of improving our understanding of how the legal and regulatory environments aﬀect small ﬁrms. It summarizes results from the ﬁrst three years of eﬀort within KRI. KRI’s initial objective was to evaluate and inform legal and regulatory policymaking related to small businesses and entrepreneurship through objective, rigorous, empirically based research. Our initial research eﬀorts over the past three years have focused on three broad questions: • What insights can be drawn from existing research on the impact of regulation and litigation on small businesses and entrepreneurship? • What data are available to support research on the impact of regulation and policy on small businesses, and what additional data are needed? • What do focused studies in selected policy areas reveal about the diﬀerential eﬀects of regulation and policy on small businesses?

In the Name of Entrepreneurship?

KRI has supported a number of research studies on diﬀerent topics designed to address these questions.

Overview of the Book This book highlights some of the key ﬁndings from research eﬀorts supported by KRI and describes a road map for future work. Speciﬁc topics were chosen to leverage existing RAND expertise, utilize readily available data sources, and develop new data sources. Chapter Two describes the regulatory and legal environments in four key areas in which government regulation and private litigation both play a role in controlling the relationship between business and society and in which the regulatory and legal systems might be expected to have a diﬀerent eﬀect on small businesses: corporate and securities law, environmental law, employment law, and HI regulation.3 This chapter provides a review of previous research and summarizes what we know about the relationship between regulation and small businesses. In reviewing the regulatory environment, the chapter describes the exemption thresholds and other approaches to adjusting the regulatory environment for small businesses. The chapter emphasizes the importance of considering whether regulations or programs designed to beneﬁt small businesses are meeting their objectives, whether they are well targeted, and whether they have unintended consequences that interfere with intended aims. The chapter concludes with some general observations on the way in which the regulatory and legal environments are adjusted to address small-business concerns and a summary of what we know about the eﬀectiveness of these approaches. This overview sets the stage for four research studies presented in Chapters Three through Six of this book. These focused analyses

Although these four areas cover a signiﬁcant slice of government regulation, our review does not provide a comprehensive summary of the eﬀect of government policy on small businesses. In particular, we do not consider the eﬀects of taxation and tax policy. While tax policy may not always be viewed as regulatory in nature, government actions and policies related to taxation have an enormous eﬀect on business decisions.

Introduction

contribute to our knowledge about the eﬀect of public policy on small businesses and the role of research in policy assessment. Chapter Three provides an assessment of recent policies designed to improve the HI market that small businesses face. In so doing, it considers the more general issue of the eﬀect of regulatory thresholds on the behavior of small businesses. The study reviews evidence that state-level regulation of HI did not result in the intended beneﬁts for small businesses. Although HI regulations have been designed and implemented by nearly all states with the purpose of improving access to HI among employees of small businesses, research suggests that the regulations have had no impact on the propensity of small ﬁrms to oﬀer HI. Moreover, the study suggests that these regulations had unintended eﬀects. Speciﬁcally, the analysis provides evidence that ﬁrms that were close to the threshold for inclusion in small-group HI reforms actually hired more employees in order to avoid being subject to the regulation. Chapter Four summarizes results from a recent study on the relationship between ﬁrm size and workplace fatalities. The study separates out the eﬀects of ﬁrm size and establishment size on safety risk and discusses the implications for health and safety policy. The distinction between ﬁrm size and establishment size and its implications for public policy arises in many settings. This study found that the smallest establishments within a ﬁrm have the highest fatality rates. Surprisingly, the risk associated with small establishments depends on the size of the ﬁrm of which they are a part. The research found that, for establishments in size categories with fewer than 100 workers, those in the smallest ﬁrms had the lowest fatality rates. Smaller establishments in larger ﬁrms were the riskiest, while small, single-establishment ﬁrms were among the safest. The ﬁndings point to several issues that should be considered in developing policy options to address health and safety problems at small establishments or ﬁrms. Chapter Five provides an overview of the regulatory regime created by SOX (P.L. 107-204) and its implications for small ﬁrms. The authors review the available evidence on SOX’s eﬀects on compliance costs, market reactions, and ﬁrm deregistrations. This chapter points out the challenges involved in comparing the ﬁndings from diﬀerent

In the Name of Entrepreneurship?

research studies due to variation in the deﬁnition used of small business. Nevertheless, the chapter ﬁnds evidence that SOX increased the auditing and accounting costs for public ﬁrms and that the cost burden on small ﬁrms relative to large ﬁrms grew. The review also suggests that SOX adversely aﬀected the market value of small but not large ﬁrms and made small public ﬁrms likelier than large ones to be acquired by private providers. Beginning in the 1990s, states permitted law ﬁrms (and other professional service ﬁrms) to organize as LLPs and LLCs. These organizational forms preserve many of the attractive features of a partnership while shielding each of a ﬁrm’s owners from liability for the malpractice of other owners. Chapter Six examines how the availability of these new business forms aﬀected the organization of law ﬁrms. The authors ﬁnd that smaller ﬁrms were much less likely than larger ﬁrms to reorganize but that small partnerships that reorganized grew faster than those that did not. Limited liability appears to be modestly beneﬁcial to the owners of small law ﬁrms. Chapter Seven turns to a consideration of the availability of data for investigating small-business policy issues. We describe key data sources, highlighting the pros and cons of each source. We also discuss the limitations of existing data. This discussion will be of particular interest to researchers as well as policymakers who would like to commission research to inform the policy debate. The book concludes with an assessment of our eﬀorts to date and some suggestions for future research. This research represents a ﬁrst step in a larger research and policy agenda designed to better understand how government actions aﬀect small ﬁrms and how policies could be better designed to promote entrepreneurship.

CHAPTER TWO

The Impact of Regulation and Litigation on Small Businesses and Entrepreneurship: An Overview Lloyd Dixon, Susan M. Gates, Kanika Kapur, Seth A. Seabury, and Eric Talley

This chapter summarizes key diﬀerences between large and small ﬁrms in the way the legal and regulatory environments aﬀect them. These diﬀerences stem from variations in laws and regulations by ﬁrm size, in implementation, and in business response. The chapter examines the regulatory and policy environments in four key areas: corporate and securities law, environmental law, employment law and regulation, and HI regulation. These four areas cover a signiﬁcant slice of regulatory activity that is important to small business and entrepreneurship. HI is the number-one concern of small businesses today, and HI regulations are designed to address these concerns. Policymakers and other stakeholders have long-standing concerns regarding whether regulations in the environmental and employment areas place an unfair burden on small ﬁrms compared to that placed on larger ones due to the ﬁnancial, personnel, and resource costs required for compliance. Although corporate and securities law has not traditionally garnered much attention from the small-business community, that changed in 2002, when SOX was passed. More generally, it provides an interesting area of focus for understanding issues related to the growth of small ﬁrms and their transformation into large ﬁrms. This discussion builds on the view of the regulatory reform process shown in Figure 1.1 in Chapter One. As noted in Chapter One, most regulation tends to originate from concern about large businesses or about business in general rather than because of concern about small

In the Name of Entrepreneurship?

businesses per se. However, the regulatory environment that is created to address public concerns sometimes focuses speciﬁcally on small businesses by, for example, exempting certain types of businesses from the regulation or applying diﬀerent enforcement mechanisms to diﬀerent types of ﬁrms. The distinction between large and small businesses is reﬂected in the extent to which we ﬁnd diﬀerent responses to the regulatory environment due to ﬁrm size. Such diﬀerences may or may not be evident depending on the regulatory issue in question. For example, the cost of implementing a complicated pollution-abatement technology may be so large as to invoke the same responses from a ﬁrm with 10 employees and from a ﬁrm with 100 employees (i.e., to close down). Meanwhile, the cost associated with a labor-reporting requirement may be a great burden on the ﬁrm with 10 employees but not on the ﬁrm with 100 employees. In this chapter, we use a ﬂexible deﬁnition of small business that is appropriate to the contextual application, the form of legal regulation, and the underlying social policy rationale being discussed. Perhaps the most often-cited deﬁnition of small business is that provided by the Small Business Act: “One that is independently owned and operated and [that] is not dominant in its ﬁeld of operation” (P.L. 83-163). Note that this deﬁnition does not require the size of a business to be viewed through a unique lens, such as employee ranks, gross receipts, ownership structure, or market presence. The remainder of this chapter focuses on the four areas of corporate and securities law, environmental law, employment law and regulation, and HI regulation. For each topic, we attempt to answer two questions: • What are the key features of the regulatory and policy environments, and in what ways does the regulatory environment distinguish between large and small businesses? • How have the regulatory policy and legal environments aﬀected small businesses, and, in particular, in what ways has the smallbusiness response to these environments diﬀered from those of larger ﬁrms?

The Impact of Regulation and Litigation

To tackle these questions, we begin with a review of the major laws and rules governing business behavior as well as any regulatory provisions focusing speciﬁcally on small businesses. Then we consider how the eﬀects of and the small-business response to the regulatory and policy environments diﬀer or might be expected to diﬀer from those for large businesses. In this discussion, we draw on insights from existing research when available.

Corporate and Securities Law We ﬁrst consider the role that business-organization law and securities regulation can play in the formation, growth, and transition of small, entrepreneurial businesses. Corporate and securities law and regulations have real implications for the future growth potential of businesses and are thus relevant to the issue of entrepreneurship. Many small ﬁrms are formed with at least a partial eye toward becoming large ﬁrms, and the road to doing so almost always involves consideration of business form and capital structure. At critical junctures, the role of corporate and securities law is paramount. Consequently, these areas of law are likely to loom large to entrepreneurs even at the very inception of a business plan. We should note that the deﬁnition of small business used in corporate and securities law is not generally based on conventional measures of operational size (such as employees, revenues, or market power), as is the case for most other regulatory spheres. Instead, corporate and securities law typically conceives of size in terms of either the distribution or the value of ownership in the ﬁrm. Privately held ﬁrms are generally considered small for the purposes of securities regulation; they are largely exempt from the mandates of federal securities law as long as they maintain their existing ownership form. While privately held ﬁrms would also usually be classiﬁed as small according to more typical measures (e.g., number of employees, gross receipts), there are exceptions. For example, privately held Cargill does business in the agricultural, food-distribution and -export, and industrial sectors; employs more than 100,000 people in 59 countries; and generates

In the Name of Entrepreneurship?

annual sales of approximately $60 billion, making it among the world’s largest companies. Regulatory Environment

There are three key decisions in the area of corporate and securities law that inﬂuence a business’s size, or how it is aﬀected by the regulatory environment: the decision to incorporate, the choice of organizational form, and the decision to be publicly as opposed to privately held. Incorporation. Incorporation creates a legal distinction between a ﬁrm and its owners. Unincorporated ﬁrms can incur business-liability risks that have the potential to imperil the assets of the ﬁrm’s owners. Incorporation is available to ﬁrms of all sizes—even those owned by sole proprietors. The primary beneﬁt of incorporation is to limit the liability of the ﬁrm’s owners for the ﬁrm’s debts and obligations. Other advantages include an unlimited life span (the corporation can continue even after the owner dies), transferability of shares, and the ability to raise capital. The major disadvantages associated with incorporation stem from the administrative paperwork burden and taxation. Because the debts and obligations of unincorporated businesses are frequently indistinguishable from those of the owners, in the event of a business failure, personal bankruptcy law may aﬀect the owners of unincorporated businesses. The Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 (BAPCPA) (P.L. 109-8) was designed to reform some aspects of U.S. personal bankruptcy law, and (in part) to stem a perceived crisis in consumer credit that had culminated in record numbers of personal bankruptcy ﬁlings (see U.S. House of Representatives, 2005, p. 3). One of the key reforms implemented by the law involves a personal-income threshold for access to Chapter 7 (liquidation) bankruptcy proceedings, the eﬀect of which will likely be to force many would-be bankruptcy petitioners to ﬁle under Chapter 13 (reorganization) instead. As a practical matter, traditional Chapter 7 proceedings involved partial liquidation of a debtor’s assets and legal termination of most debts without lien against a debtor’s future income. By contrast, proceedings under the revised Chapter 13 are more burdensome, and

The Impact of Regulation and Litigation

primarily involve scheduled repayment of debts out of future income rather than a dismissal of debts in return for a liquidation of assets. The result is a new bankruptcy regime that is far less friendly to personal debtors, one that will focus more on restructuring and enforcing debt payments than on dismissing them. Several other changes implemented by the recent legislation likewise serve to make personalbankruptcy laws more favorable to creditors and less protective of debtors. These changes include a longer schedule of mandatory repayments for some debtors under Chapter 13, more limitations on the categories of debt subject to Chapter 13 proceedings, and new statutory provisions designed to prevent debtor forum-shopping for favorable state property exemptions (see, e.g., Walker, 2005; see also Compact Library Publishers, undated). Organizational Structure. Organizational choices are broader than the decision of whether to incorporate, and the options available to ﬁrm owners have grown substantially over the past 20 years. Table 2.1 summarizes these options. Organizational options do sometimes take ﬁrm size into account (e.g., S-corporation status is limited to ﬁrms with fewer than 75 shareholders). The traditional general partnership1 allows for comanagement as well as proﬁt- and loss-sharing and allows governance to be tailored to a ﬁrm’s individual needs. Partnerships receive pass-through treatment for tax purposes, allowing the owners to avoid entity-level taxation of their partnership income. A key disadvantage of partnerships is that all general partners are jointly and severally liable for the professional misconduct of other partners and are jointly liable for all other obligations of the ﬁrm. The state-chartered corporation provides a more formalized structure than a partnership. Corporate status confers limited liability on its owners, so that they risk only the value of the shares they own. On the downside, corporate status is generally perceived as more

This continues to be the default legal relationship for multiperson ﬁrms; however, default does not mean “dominant.” Rather, if a business organization comes into being without complying with statutory formalities for forming, for example, a corporation, it will be considered a general partnership by default.

In the Name of Entrepreneurship?

Table 2.1 Options for Business Organizational Structure Organizational Form

Key Characteristics

General partnership

Allows for comanagement, proﬁt-sharing, and loss-sharing, with governance tailored to the ﬁrm’s individual needs. Receives pass-through treatment for tax purposes, allowing the owners to avoid entity-level taxation of their partnership income. Holds all partners jointly and severally liable for torts of other partners.

State-chartered corporation Provides more formalized structure than partnership. Confers limited liability on its owners. Governance is largely regulated by statutes that are difﬁcult to overturn. S corporation

Allows professional corporations to take advantage of pass-through taxation under subchapter S of the Internal Revenue Code. Limited to companies with fewer than 75 shareholders, who must be U.S. citizens. Allowed to have only one class of stocks.

LLC and LLP

Combines the ﬂexibility and pass-through taxation attributes of partnerships with a form of limited liability akin to that accorded to corporate status. LLCs and LLPs are required to have a limited life span and typically must carry a minimum amount of insurance against claims of third-party creditors.

cumbersome and inﬂexible, as governance procedures are largely regulated by statutes that, while allowing participants to opt out, are nonetheless perceived as diﬃcult to overturn. Moreover, corporate status generally implies double taxation, in which the ﬁrm is taxed at the entity level and distributions to shareholders are once again taxed at the individual level. About 25 years ago, both state business-recognizing bodies and state and federal taxation authorities began to implement signiﬁcant reforms to their company-law statutes in a way that eroded the distinction between corporate and partnership status. During the early 1980s, federal taxation authorities began to allow professional corporations to take advantage of pass-through taxation by electing tax treatment under subchapter S of the Internal Revenue Code. For some ﬁrms, this option has proven extraordinarily beneﬁcial. Nevertheless, Scorporation status still imposes a few important constraints on ﬁrms choosing this form. First, subchapter S applies only to companies with

The Impact of Regulation and Litigation

fewer than 75 shareholders, all of whom must be U.S. citizens. Second, while enjoying pass-through taxation, S corporations often cannot deduct the full expenses of many employee-beneﬁt plans (which C corporations can) and are generally unable to use basic strategies to reduce or avoid tax liabilities on a sale of assets or share redemption, such as a step-up in the tax basis of an asset. In addition, S corporations are allowed to have only one class of stock. Meanwhile, state legislatures have passed statutes authorizing the LLC and LLP forms, which many professional ﬁrms have adopted. LLCs and LLPs combine the ﬂexibility and pass-through taxation attributes of partnerships, while according their owners a form of limited liability akin to corporate status. These novel business forms were adopted either jointly or individually within every state and the District of Columbia between 1977 and 1996. Publicly Held Status. Firms that choose a corporate organizational status may also choose to be publicly held. Generally speaking, a publicly held corporation is one with shares held by a large number of people. The value of the assets and the number of shareholders determine whether the corporation is considered private in the sense that the SEC governs its activities. Although few entrepreneurs choose public status at the inception of a small business, the issue becomes increasingly germane as the ﬁrm grows and requires access to additional capital. Securities regulations are a central consideration for smaller ﬁrms seeking to make the transition to publicly traded status or to sustain and expand that status. Federal securities regulations require ﬁrms to ﬁle an Exchange Act registration statement if they have more than $10 million in assets and a class of equity securities with more than 500 shareholders of record or list securities on an exchange. Firms that do not meet these criteria may, but are not required, to register. Registered ﬁrms face a variety of reporting requirements and restrictions outlined in the Securities Act of 1933 (P.L. 73-222) and the Securities Exchange Act of 1934 (P.L. 73-291). These regulations are designed to protect investors’ interests by requiring the disclosure of pertinent information.2 2

For details on these federal requirements, see SEC (2006b).

In the Name of Entrepreneurship?

Federal securities regulations have simpliﬁed registration procedures for small businesses, allowing them to use streamlined processes either to begin oﬀering securities for sale to the public or to expand their current oﬀerings. In particular, the SEC allows an enterprise to use a special form SB-1 (SEC, 2007a) or SB-2 (SEC, 2007b) to register as a small-business issuer if (1) the business is a U.S. or Canadian issuer that had less than $25 million in revenues in its last ﬁscal year and (2) the business’s outstanding, publicly held stock is worth no more than $25 million. Registration with the SEC using the SB-1 or SB-2 forms still requires the submission of audited ﬁnancial statements. The SEC also allows small businesses (there are some exceptions) to do Regulation A oﬀerings (SEC, 2001), which allow for public oﬀerings of stock not to exceed $5 million in any 12-month period. The Regulation A option was created to allow a small business to “test the waters” for interest in its securities before going through the expense of ﬁling with the SEC. Even though Regulation A oﬀerings still require the submission of ﬁnancial statements, the statements are simpler and do not need to be audited. SOX (P.L. 107-204) imposes additional requirements on publicly traded ﬁrms. These requirements have major implications for the governance, accounting, auditing, and executive-compensation environment for publicly traded ﬁrms. For example, the act requires senior executives to personally certify ﬁnancial statements; requires ﬁrm-auditing committees, nomination committees, and compensation committees to be completely independent; and requires the board of directors to be majority independent.3 The SOX legislative language does not single out small businesses for special or diﬀerent treatment. However, in its rulemaking process, the SEC has delayed the start dates for compliance with key elements of section 404 of the act for nonaccelerated ﬁlers4 and foreign ﬁrms and

Companies that are majority owned by a single shareholder or uniﬁed group, however, are exempt from the some of these requirements.

Accelerated ﬁlers are ﬁrms with a minimum ﬂoat of $75 million and at least one year’s worth of ﬁnancial reporting.

The Impact of Regulation and Litigation

is also debating further guidance regarding small-business compliance with the act. Effects of Corporate and Securities Law on Small Businesses and the Small-Business Response

We now consider the eﬀects of the regulatory and policy environments in corporate and securities law on small businesses and the ways in which small businesses have responded to these environments. Effect on the Decision to Incorporate. In business-organization law, one of the most salient diﬀerences between ﬁrms lies in the degree to which their respective owners bear personal liability for business risks. This diﬀerence in liability exposure can have important implications for ﬁrm behavior. Unincorporated ﬁrms (or those that have not sought refuge in other statutory forms) can incur business-liability risks that could imperil the ﬁrm’s owners’ assets. This is true even if a sole proprietor owns the ﬁrm. Existing research (see, e.g., Ribstein, 2004) indicates that unincorporated business owners5 are less likely to take risks, are often less innovative, and have distinct (often slower) growth trajectories than their corporate counterparts. The diﬀerent legal and regulatory environments facing unincorporated ﬁrms from those facing incorporated ones may also have signiﬁcant implications for the initiation and growth of small businesses. The threat of ﬁnancial liability for a ﬁrm’s obligations might loom large for entrepreneurs and inﬂuence their ability to innovate, grow, or even begin operations in the ﬁrst place. Previous research (Fan and White, 2003) ﬁnds evidence of a “chilling eﬀect” of strict personal-bankruptcy laws on entrepreneurship. Changes to the personal-bankruptcy law following from BAPCPA (P.L. 109-8) may have a signiﬁcant impact on small businesses and entrepreneurs. There is broad concern that BAPCPA, which makes it much harder for individuals to obtain a fresh start, will exacerbate the distinction between incorporated and unincorporated ﬁrms in terms of the level of ﬁnancial risk borne by the owners and further chill entrepreneurship.

These ﬁndings apply to business owners who engage in the business on a full-time basis.

In the Name of Entrepreneurship?

Although incorporation is an organizational choice that is available to ﬁrms of all sizes, small businesses (measured in terms of number of employees or gross receipts) are much likelier than larger businesses to be unincorporated. More than 75 percent of all small businesses in the United States (measured in terms of annual sales receipts) have no payroll employees at all, and a majority of those are unincorporated (U.S. Census Bureau, 2007b). There are important reasons that small ﬁrms may ﬁnd incorporation less attractive than their larger counterparts do. First, the formalities required to incorporate (involving not only the initial paperwork, but also creation and management of governance bodies) involve ﬁxed costs that smaller ﬁrms may be less likely to be in a position to bear. Second, incorporation does not necessarily eliminate the risk of personal liability for shareholders, particularly for closely held companies. In many circ*mstances, courts can (and do) disregard the veil of limited liability that ostensibly protects shareholders of a corporation, using a doctrine known as “piercing the corporate veil” (or PCV). When PCV doctrine is invoked, shareholders of an incorporated entity are held liable for the ﬁrm’s debts and liabilities as if the ﬁrm were an unincorporated business entity. Although PCV doctrine diﬀers slightly across jurisdictions, a factor that is common to all is whether there is unity of ownership and interest between the corporation and its shareholders, an inquiry that generally turns on determining whether there is suﬃcient separation between the ﬁrm and its owners. The only successful PCV cases that have ever, to our knowledge, been asserted have been against closely held corporations and not against publicly traded ﬁrms (although wholly owned subsidiaries of publicly traded ﬁrms, which are technically closely held ﬁrms, are also common targets). Thus, incorporation is unlikely to be a complete liability-risk panacea for ﬁrms whose ownership is closely held. Effect on Choice of Organizational Form. The characteristics of the LLC and LLP forms might make them of particular interest to small entrepreneurial ﬁrms; however, little is currently known about the take-up of these forms among smaller ﬁrms (see Chapter Six). The LLC and LLP organizational forms were intended to allow owners of ﬁrms to share the best attributes of partnerships and corporations in terms of both ﬂexibility and pass-through taxation as well as limited

The Impact of Regulation and Litigation

liability. Most states have permitted ﬁrms to organize as either LLCs or LLPs (sometimes both) without the cumbersome constraints that frequently attend corporate status. However, LLC and LLP forms also come with a few costs that could aﬀect their take-up among small or large ﬁrms. First, unlike corporations (and even partnerships), LLCs and LLPs are required to have a limited life span (frequently in the neighborhood of 35 years). Although ﬁrms are allowed to re-form at the end of this period, the terminal period itself can create both tax and strategic problems for a ﬁrm. In addition, enabling statutes typically require that LLCs and LLPs carry a minimum amount of insurance against claims of thirdparty creditors. Moreover, even within a state, there is frequently some variation in the nature and extent of liability protection that these new business forms aﬀord. For example, the LLP form frequently provides only a partial shield against liability6 and imposes larger ﬁduciary duties on its members than the LLC form does but is also signiﬁcantly more ﬂexible than the LLC form. The Public-Private Divide. As entrepreneurial ﬁrms grow larger and require access to additional capital, they face a choice as to whether the beneﬁts of publicly traded status are worth the costs associated with regulatory requirements. Securities and exchange laws have long imposed reporting and other requirements on ﬁrms that register with the SEC; however, by all accounts, SOX legislation (P.L. 107-204) (as well as related regulations) has changed the landscape of securities law for ﬁrms that are publicly traded. There is no consensus regarding how SOX rules aﬀect the interests, prospects, and growth trajectories of companies that are not listed on a national exchange (and thus not subject to federal securities regulations). On the one hand, some argue that SOX (P.L. 107-204) changes make it easier for closely held businesses to make the transition to pub6

Most notably, a number of states provide partners in an LLP only partial liability shields against third-party creditors (most notably, tort claimants alleging malpractice by other partners). These “partial shield” states still allow for liability as to the LLP’s general debts and include Alaska, Arkansas, District of Columbia, Hawaii, Illinois, Kansas, Kentucky, Louisiana, Maine, Michigan, Nevada, New Hampshire, New Jersey, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Utah, and West Virginia.

In the Name of Entrepreneurship?

licly traded status—because the additional reports and assessments they must produce help convince prospective investors that good internal controls are already in place, thereby making small-business investments safer than they have tended to be in the past.7 Many critics, however, have pointed out that the requirements will simply impose a compliance cost for doing business as a public company. If such costs are high enough, privately held ﬁrms will eschew registration or, if they are already registered, might delist because of the increase in recurring expenses and other eﬀects that the new SOX rules will impose. If the SOX regulatory innovations create a situation in which only large businesses can aﬀord to go or remain public, small businesses may face differential diﬃculties in accessing capital, with potentially far-reaching eﬀects for the markets and economic growth in general.8 Chapter Five explores these issues further. Section Conclusion

This brief review suggests several ways in which the regulatory environment surrounding corporate and securities law might aﬀect small businesses diﬀerently from the way it aﬀects larger ones. However, while there is ample theoretical justiﬁcation for this hypothesis, empirical evidence regarding the diﬀerential eﬀect of corporate and securities law on small businesses is lacking. Key questions remain to be addressed empirically, including the extent to which the new regulatory requirements have aﬀected small ﬁrms’ willingness to go into (or stay within) the public capital markets, how well alternative sources of capital (e.g., private-equity markets) substitute for the beneﬁts of public capital, and

The PCAOB chief auditor, Douglas Carmichael, has expressed the view that “small companies may actually beneﬁt from the new [SOX reporting] requirements, because fraud tends to be more prevalent among small companies, making access to capital markets harder. The new requirements should reduce uncertainty and therefore improve access” (Solomon and Bryan-Low, 2004).

Quantitative estimates have already begun to appear on the costs associated with SOX (P.L. 107-204) compliance. Solomon and Bryan-Low (2004), for example, cite a study released by Financial Executives International estimating that ﬁrms with annual revenues of less than $25 million will incur ﬁrst-year SOX-compliance costs of $0.28 million and 1,996 hours.

The Impact of Regulation and Litigation

whether the viability of these alternative sources diﬀer according to ﬁrm size. Chapters Five and Six in this book will, respectively, examine the regulatory regime surrounding SOX (P.L. 107-204) and its eﬀect on small ﬁrms and consider the beneﬁts of limited-liability organizational forms among small law ﬁrms.

Environmental Protection We now look at the regulatory environment surrounding environmental protection. There are several reasons to expect ﬁrm size to be an important consideration in formulating and evaluating environmental policy. From the regulator’s perspective, it may be more cost-eﬀective to focus regulation and enforcement on large sources, which are usually large ﬁrms. Liability-based approaches may be more eﬀective for ﬁrms that have deep pockets. In addition, regulatory approaches that require ﬁrms to provide information may succeed better with ﬁrms concerned about their public image, which again may tend to be larger. At the same time, it may be easier for larger ﬁrms to comply with environmental regulations; as a result, regulations might increase the minimum eﬃcient scale of production, putting small ﬁrms at a competitive disadvantage. Regulatory Environment

There are three main components to the regulatory environment surrounding environmental protection: the regulations, government mechanisms for enforcing the regulations, and liability or citizen enforcement mechanisms. Regulations. Environmental regulations attempt to reduce the negative eﬀects of manufacturing and other business operations on the environment and people’s health. Firms frequently do not bear the full environmental or public-health costs of their operations. Thus, they do not have appropriate incentives to control emissions. The purpose of environmental regulations is, at least in part, to correct these so-called negative externalities.

In the Name of Entrepreneurship?

In the United States, federal environmental laws initially focused on large sources of pollution and on large ﬁrms. These laws were implicitly designed with large ﬁrms in mind—ﬁrms that could aﬀord in-house environmental-compliance oﬃces and that had engineering expertise. Over time, as large sources increasingly came under control, the EPA, state environmental agencies, and environmentalists gradually turned their attention to midsize sources of pollution and to smaller ﬁrms. While the emissions of any one small ﬁrm might not be large, the large number of small ﬁrms in many industries made the cumulative emissions of all small ﬁrms a source of concern. As attention has shifted toward smaller sources, the question arises whether regulatory approaches that were initially developed with large ﬁrms in mind are appropriate for small ﬁrms. Table 2.2 lists major federal environmental laws.9 Many environmental regulations impose diﬀerent requirements on ﬁrms of diﬀerent sizes. Varying the requirements of environmental regulations by ﬁrm size became common starting in 1985 (Hopkins, 1995, p. 8). This tiering means that small ﬁrms are exempted from certain requirements or are required to meet less stringent emission- or treatment-technology standards. According to the U.S. Small Business Administration (SBA) Oﬃce of Advocacy, the EPA has tiered more than 50 diﬀerent regulations based on either ﬁrm size or the amount of pollution released (SBA, 1995, p. 5). Governmental and Private-Sector Enforcement Mechanisms.

The implementation and enforcement of environmental regulations occurs through a variety of mechanisms that diﬀer according to the regulation, industry, and ﬁrm in question. State agencies frequently play the lead role in this process. Firms are often required to collect substantial data on emissions and the use of hazardous substances and to document the plans and procedures that they have put in place to comply with regulations. These data and plans are then submitted to various local, state, and federal agencies. When ﬁrms are found to be out of compliance, they may be subject to ﬁnes and required to take 9

Several states have also have enacted their own environmental laws and regulations that are stricter than the federal laws.

The Impact of Regulation and Litigation

Table 2.2 Major Federal Environmental Laws Law

Year Enacted or Amended

Focus

Federal Insecticide, Enacted in 1947, but Fungicide, and Rodenticide amended with major Act (FIFRA) (P.L. 80-104) changes in 1972

Regulates pesticides and particular chemicals

Clean Air Act (P.L. 91-604)

Regulates air emissions

National Environmental Policy Act (NEPA) (P.L. 91-190) Federal Water Pollution Control Act Amendments of 1972 (Clean Water Act) (P.L. 92-500) Safe Drinking Water Act (SDWA) (P.L. 93-523) Resource Recovery and Conservation Act (RCRA) (P.L. 94-580)

Enacted in 1967 as the Air Quality Act (P.L. 90-148) and amended in 1970, 1977, and 1990 1969

Addresses general environmental policy and practice

Enacted in 1972 and Regulates discharges into amended in 1977 and 1987 bodies of water

1974

Enacted in 1976 and signiﬁcantly amended in 1984

Sets standards for drinking water and discharges into sources of drinking water Focuses on waste disposal into landﬁlls

Toxic Substances Control Act (TSCA) (P.L. 94-469)

1976

Regulates chemical use and disposal in general

Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) (P.L. 96-510)

1980

Addresses cleanup of abandoned or inactive hazardous-waste sites

Superfund Amendments and Reauthorization Act (SARA) (P.L. 99-499)

1986

Amended CERCLA and established the Toxic Release Inventory

Pollution Prevention Act (P.L. 101-508)

1990

Addresses general environmental policy and practice

action to remedy the problem. The federal statutes listed in Table 2.2 were adopted or substantially amended between the late 1960s and 1990. During the 1990s, eﬀorts were made to integrate and streamline

In the Name of Entrepreneurship?

the fragmented air, water, and waste laws and programs. We brieﬂy review some of these here. Voluntary Agreements. Voluntary agreements can be classiﬁed into three types: public-voluntary, unilateral, and negotiated agreements (OECD, 1998). Public-voluntary agreements are nonmandatory rules developed by EPA or other government regulators. For example, the EPA’s 33/50 program, which concluded in 1995, encouraged manufacturers to voluntarily reduce emissions of 17 target chemicals by 50 percent (EPA, 1999). Unilateral agreements are made by industry for industry. An example is the American Chemistry Council’s Responsible Care® program, which encouraged member companies to adopt environmental management principles (American Chemistry Council, undated). Negotiated agreements are contracts between public authorities and industry. The two most visible examples have been EPA’s Project XL (EPA, 2006a) and its Common Sense Initiative (CSI), both designed in response to complaints from the regulated community regarding the growing complexity of federal environmental laws (OECD, 1998).10 All these types of agreements were largely abandoned when the George W. Bush administration came into oﬃce in 2001. The main exception is EPA’s National Environmental Performance Track program, a voluntary partnership program that recognizes and rewards private and public facilities that demonstrate strong environmental performance beyond current requirements (EPA, 2007). Government Programs. Several government programs were adopted over the past 10 years in an attempt to improve the regulatory system’s performance for small businesses. SBREFA (P.L. 104-121) directs the SBA to establish a regulatory-enforcement ombudsman and regulatory-fairness boards in 10 regional cities. SBREFA allows a small business to ﬁle a grievance in court if it believes that the business has been “adversely aﬀected or aggrieved” by a regulatory ruling. The courts can rule that the regulations should not be enforced against a small ﬁrm (“Small Business Not So Small,” 2002). To monitor agency eﬀorts to reduce regulatory burden, the Small Business Paperwork Relief Act of 2002 (SBPRA) (P.L. 107-198) requires agencies to report to Congress 10

See Coglianese and Allen (2003) for a review of EPA’s Common Sense Initiative.

The Impact of Regulation and Litigation

and the Small Business and Agricultural Regulatory Ombudsman on their enforcement actions against small businesses and the penalty reductions in such actions. The PRA (P.L. 104-13) also requires federal agencies to review the impact of their regulations on small businesses and to consider less costly alternatives for accomplishing public policy goals (“Small Business Not So Small,” 2002). EPA’s Small Business Compliance Policy (EPA, 2000) promotes environmental compliance among businesses with 100 or fewer employees by providing incentives to discover and correct environmental problems. EPA eliminates or signiﬁcantly reduces penalties for small businesses that voluntarily discover violations of environmental law and promptly disclose and correct them (EPA, 2006b). SBREFA (P.L. 104-121) provides new avenues for small businesses to participate in the federal regulatory process. In response, EPA has set up panels to facilitate greater small-business participation in the regulatory process (SBA, undated[b]). This initiative responds to concerns that greater large-ﬁrm participation in the regulatory and political process has resulted in regulations that are tailored to the experiences and capabilities of larger ﬁrms. There do not appear to be studies on the use and eﬀectiveness of this or similar programs. Environmental Management Systems. In recent years, government and industry have been exploring standards and guidelines for the management of a ﬁrm’s activities related to environmental performance. Environmental management systems do not specify particular emission standards, but they provide guidelines for management structures. For example, EPA has adopted management-based regulations aimed at preventing accidents involving hazardous chemicals. These regulations require facilities to conduct risk assessments of their operations, develop procedures to prevent accidents, and seek to make continuous improvement in the management of their operations (Coglianese and Nash, 2006, p. 6). Liability and Citizen-Enforcement Mechanisms. The tort system is an alternative to active regulatory oversight. Firms often face liability for the release of pollutants into the environment. The highest-proﬁle example is probably the federal Superfund program (part of CERCLA implementation, P.L. 96-510), which imposes strict, joint and several,

In the Name of Entrepreneurship?

and retroactive liability for the cleanup of hazardous-waste sites. Small ﬁrms accounted for the majority of businesses potentially liable for cleanup, but for a more moderate share of the total waste sent to the site (Dixon, 2000). Environmental and toxic-tort claims can also cause ﬁrms to incur costs and requirements to change their business practices. The use of class-action lawsuits for environmental and toxic-tort claims has been a topic of ongoing debate. From 1980 to the mid-1990s, the trend was toward more widespread usage of class-action suits for environmental and toxic-tort claims. In the mid-1990s, however, a series of federal appellate court decisions reversed class certiﬁcations in pending classaction tort cases (Schwartz and Sutherland, 1997). A third dimension of environmental liability is the impact of recent court decisions on large ﬁrms’ exposure to liability claims. In the 1990s, the U.S. Supreme Court ﬁrst considered the question of whether large, multinational ﬁrms could shield themselves from CERCLA liability through a parent-subsidiary relationship. In United States v. Bestfoods (1998), the court held that a corporate parent could be held vicariously liable for its subsidiary’s environmental damage if the parent’s right of control over the subsidiary’s business was suﬃciently large to convert the parent into an “operator” under CERCLA (P.L. 96-510). If these events enhanced large ﬁrms’ exposure to vicarious liability, the predicted eﬀect would be to induce them to contract out much of their high-risk work to smaller, less liquid ﬁrms, working substantially independently. Citizen enforcement can provide another form of oversight. The federal Clean Water Act (P.L. 95-217) and several other federal environmental laws allow citizens to bring enforcement actions. Private enforcement of environment-related regulations is also allowed under California’s Safe Drinking Water and Toxic Enforcement Act of 1986 (Proposition 65). Proposition 65 prohibits businesses from knowingly discharging listed chemicals into sources of drinking water and requires warnings before otherwise exposing someone to a listed chemical.11 11

Citizen suits have also been allowed under California’s Unfair Competition Law (California Business and Professions Code, §17200). Citizen suits brought under this law were com-

The Impact of Regulation and Litigation

There is a great deal of controversy about the social value of citizen-suit provisions. Supporters contend that empowering “private attorneys general” is an appropriate and eﬀective way to augment the limited resources of public enforcement agencies. Critics contend that citizen suits are often used to pursue narrow private interests, generate legal fees while focusing on permit violations that cause little environmental harm, and restrict the socially useful discretion of public enforcement agencies. Citizen-suit provisions have now been in place for more than 25 years, but there is little systematic, empirical information about them. There has been a good deal of legal analysis of the various statutes and court cases (see Leonard, 1995; Austin, 1986–1987; James Thompson, 1987), but little data have been collected on the frequency, costs, and outcomes of these cases. Effects of Environmental Regulations on Small Businesses and the Small-Business Response

There has been ongoing debate over whether environmental regulations put small ﬁrms at a disadvantage relative to larger ones. Environmental regulations may more heavily impact small ﬁrms because of variation in statutes, compliance, or enforcement (Dean, Brown, and Stango, 2000, p. 58). We discuss each in turn. Statutory Variation. While the tiering of environmental regulations obviously works to the advantage of small businesses, two factors work to reduce this advantage. First, environmental regulations often contain grandfathering provisions that allow older, and perhaps larger, ﬁrms to postpone compliance with new regulations or to meet less stringent standards. For example, under new federal source-review guidelines, existing ﬁrms are not required to upgrade pollution-control equipment until they modify their existing plants by making nonroutine physical or operational changes that result in a signiﬁcant increase in emissions of a regulated pollutant. Second, as Shaller, McNulty, and Chinander (1998) observed, large ﬁrms are usually much more active in the regulatory process than smaller ﬁrms, with the result that regumonly referred to as Section 17200 suits. The passage of Proposition 64 on the California state ballot in November 2004, however, narrowed these citizen-suit provisions.

In the Name of Entrepreneurship?

lations are tailored to the experiences and capabilities of large ﬁrms. The result may be that the advantages given to small ﬁrms under the regulations may not be as large as they might ﬁrst appear. Compliance Variation. Compliance with environmental regulations can induce responses by ﬁrms along several dimensions. Compliance might involve installation of pollution-control equipment that removes pollution produced in the production process (so-called end-ofthe-pipe treatment) or installation of production equipment that generates less pollution. Compliance can also require ﬁrms to monitor waste streams or releases of pollutants into the environment and to report results to government agencies. Finally, compliance can have internal, organizational implications for ﬁrms by, for example, requiring them to designate points of contact for government agencies or to develop an emergency-response plan for the release of hazardous substances. Complying with environmental regulations can potentially put small ﬁrms at a disadvantage vis-à-vis large ﬁrms. Pollution equipment can increase the minimum eﬃcient scale of production. There can also be economies of scale in discovering and understanding environmental regulations and in completing required paperwork. The result is that environmental regulations may cause costs per unit of output to increase more for small ﬁrms than they do for larger ones. Studies have found evidence of compliance asymmetries. Pittman (1981) found that emission-control technologies required in the pulp and paper industries increased the minimum eﬃcient size of a plant. Not all studies agree, however. Using data on manufacturing ﬁrms from between 1978 and 1981, Evans (1985) did not ﬁnd strong evidence that there were substantial economies of scale in complying with EPA and OSHA regulations. More recent analyses of the impact of environmental regulations also raise questions about the economies-ofscale argument. Dean, Brown, and Stango (2000) believe that a small but growing body of evidence indicates that ﬁrms have found ways to convert environmental regulations into a competitive advantage. For example, there is evidence that, when reducing emissions, some ﬁrms have found ways to save enough inputs so that unit-production cost declines. Weakening the link between environmental compliance and cost might also weaken arguments that regulation gives large ﬁrms

The Impact of Regulation and Litigation

advantages over smaller ﬁrms, although it could still be that large ﬁrms are better at ﬁnding cost savings than smaller ﬁrms are. Even if there are not substantial economies of scale in complying with environmental regulations, environmental regulations may still make it more diﬃcult for new ﬁrms to enter the industry. Existing ﬁrms may have gradually learned the cheapest and most eﬀective way to comply with environmental regulations over time. Thus, compliance costs may be initially higher for potential entrants, discouraging entry. To the extent that entrants tend to be smaller ﬁrms, environmental regulations would disadvantage smaller ﬁrms. Enforcement Variation. Asymmetries in enforcement result when government or private parties enforce regulations more vigorously against ﬁrms in one size range than they do in another. Charles Brown, Hamilton, and Medoﬀ (1990, p. 84) found that government enforcement practices serve to reduce the regulatory burden placed on smaller ﬁrms and that the preferential treatment more than oﬀsets any disadvantages for small ﬁrms created due to economies of scale in complying with environmental regulations. Finto (1990) concluded that limited enforcement budgets cause EPA to focus enforcement eﬀorts on larger ﬁrms. There are contrary views, however. Several studies have concluded that enforcement is less stringent against larger ﬁrms than it is against smaller ones. For example, Bartel and Thomas (1987) suggested that large producers face less stringent EPA and OSHA enforcement. Some argue that larger ﬁrms often escape stringent enforcement because they are more politically inﬂuential than smaller ﬁrms are and can directly or indirectly inﬂuence enforcement priorities. Even if regulations are enforced equally for large and smaller ﬁrms, the ultimate impact of regulations on large ﬁrms may be less if they are more successful in defending themselves. Yeager (1987) found that, because larger ﬁrms have more resources, they are more successful in defending themselves against enforcement actions. Larger ﬁrms must bear the costs of such defenses, but the cost is presumably less than the expected cost of compliance, thus reducing the diﬀerence in the cost of environmental regulations between large and small ﬁrms from what it would otherwise be.

In the Name of Entrepreneurship?

Research has shown that enforcement actions by private parties tend to focus on larger ﬁrms. Greve (1989) found that environmental groups were more like to pursue enforcement actions against larger ﬁrms under the Clean Water Act (P.L. 95-217) even when these ﬁrms were not the largest polluters. Dean, Brown, and Stango (2000, p. 59) argued that private groups are likelier to target large ﬁrms than small ﬁrms because large ﬁrms are more concerned about their reputations and thus more prone to settle. Combined Effects. The net advantage or disadvantage for small ﬁrms created by variations in statutes, compliance, or enforcement policies is diﬃcult to determine and undoubtedly varies by industry as well as environmental regulation. For example, compliance asymmetries that disadvantage small ﬁrms may be oﬀset by statutory and enforcement asymmetries that favor them. Empirical studies that attempt to evaluate the combined eﬀects of environmental regulations on small ﬁrms relative to large ones have come to mixed conclusions. Before environmental tiering became widespread, Pashigian (1984) found that environmental laws placed greater burdens on smaller manufacturing plants, resulting in increased market share for larger ﬁrms. Pashigian (1984) and Bartel and Thomas (1987) concluded that, while regulations can impose signiﬁcant burdens on larger manufacturing ﬁrms, decreased competition from smaller ﬁrms might mean that, on the whole, large ﬁrms are better oﬀ with environmental regulations than without them. Dean, Brown, and Stango (2000) found that higher pollutionabatement costs resulted in fewer small ﬁrms entering into the industries examined, but not in fewer large ﬁrms entering. They concluded that, on the whole, environmental regulations put small ﬁrms at a unitcost disadvantage relative to large ﬁrms. Dean, Brown, and Stango also concluded that the disadvantages faced by small ﬁrms were not a temporary phenomenon that disappeared as ﬁrms learned to cope with regulations or as organizations evolved to aid small ﬁrms in abatement eﬀorts (2000, p. 61). It should be noted, however, that their conclusions are based on data only through 1987 and may not reﬂect conditions today.

The Impact of Regulation and Litigation

Other studies suggest that environmental regulations do not put small ﬁrms at a signiﬁcant disadvantage. Hopkins (1995, p. 61) found that environmental regulations accounted for a smaller share of the overall regulatory burden on small ﬁrms than that for large ﬁrms and that tax- and payroll-related burdens, not environmental regulation, were the main concerns for smaller ﬁrms. A 1994 survey by Arthur Andersen and National Small Business United came to similar conclusions. In addition, the study found that ﬁrms with fewer than 20 employees were more than twice as likely as larger ﬁrms to report that they faced no major regulatory burden of any kind (including environmental regulations) (Arthur Andersen and Company and National Small Business United, 1994, p. 25, quoted in Hopkins, 1995, p. 9). Researchers have found that the presence of environmental regulations can increase the number of small ﬁrms in some circ*mstances. For example, Ringleb and Wiggins (1990) argued that concerns about liability have induced larger ﬁrms to shed operations involving hazardous substances. Becker and Henderson (1997) found a proliferation of small ﬁrms in four high-polluting industries. These ﬁndings are consistent with arguments by Ringleb and Wiggins (1990) that concerns about liability have induced larger ﬁrms to shed operations involving hazardous substances. The 1998 United States v. Bestfoods Supreme Court decision would reinforce such concerns. It is diﬃcult to judge the success of eﬀorts over the past 10 or 15 years to make it easier for small ﬁrms to comply with environmental regulations. EPA’s Oﬃce of Enforcement and Compliance Assurance does not keep records on the number of small businesses participating in the agency’s self-audit program—although data may be available at regional oﬃces. Reed (1999, p. 324) found increasing largecorporation participation in EPA’s self-audit program, and she speculated that small ﬁrms might not want to participate in the program because they feared potentially high costs of correcting violations. Small ﬁrms may also have little incentive to participate in the program if they think the probability of direct government enforcement is low. Some states also have adopted audit programs, although coordination between EPA and the states has not been good (Meason, 1998). Under Illinois’s Clean Break program, businesses agree to come into compli-

In the Name of Entrepreneurship?

ance within a reasonable time in exchange for amnesty for past violations. In spite of Clean Break and EPA’s small-business programs, the audit rate in the Illinois small-business community is almost zero (Meason, 1998, p. 510). Section Conclusion

This review of research to date suggests that there are no easy answers to the question of how environmental regulations have aﬀected small ﬁrms relative to how they have aﬀected larger ones. Moreover, many of the key studies rely on data that are now quite dated. Rapid evolution in environmental regulations and policy may mean that ﬁndings of past studies do not reﬂect today’s regulatory environment. Environmental policymaking must balance competing objectives. Ultimately, it is up to policymakers to determine the balance between the beneﬁts of regulatory compliance and the costs associated with regulation. Better information would allow policymakers to make more informed decisions, particularly as they related to the impact of regulation on small ﬁrms. While the existing body of research on environmental protection in the business context is extensive, further research is needed to better understand how recent trends in environmental regulation, enforcement, and liability are aﬀecting businesses of diﬀerent sizes. Better information is also needed concerning which aspects of environmental regulatory and liability policy cause the greatest problems for small ﬁrms. Information needs to be synthesized on the environmental damage caused by small ﬁrms and the beneﬁts of reducing this damage. Major environmental initiatives have traditionally focused on large ﬁrms, and there is clear evidence that the regulations were formulated with large ﬁrms in mind. There is a need to understand whether a diﬀerent approach to source control, pollution prevention, compliance assistance, and enforcement is needed to deal with the operations of small ﬁrms. There also needs to be a more thorough evaluation of how small ﬁrms have used diﬀerent initiatives, such as the Common Sense Initiative and self-auditing programs, and what types of modiﬁcations to these programs would make them more attractive to small ﬁrms. Large

The Impact of Regulation and Litigation

ﬁrms are motivated to participate in environmental initiatives partly by concerns about their image in the communities in which they operate or their image with consumers. More research is needed to determine the types of concerns that would motivate small businesses to address their operations’ eﬀects on the environment.

Employment Law and Regulation We now examine the ways in which policies regulating the contractual relationships between employers and employees might have a differential impact on small businesses and entrepreneurship.12 A variety of regulations, rules, and policies at the federal, state, and local levels inﬂuence or restrict the ways in which businesses interact with prospective, current, and former employees. Such regulations and policies recognize that various factors can alter the balance of power between employer and employee and are designed to address concerns that one party (usually, but not always, the employer) might intentionally or unintentionally impose harm on the other. Harm may result from an employer’s intentional or unintentional discrimination against certain groups of current or potential employees that denies them access to jobs or fair wages; the establishment of a hostile or unsafe work environment; the exercise of market power to drive down wages; or lost wages due to job loss, workplace injury, or, on the employee’s side, from the theft of intellectual property or a client base from an employer. Employment laws, regulations, and policies can protect or beneﬁt one party (usually employees) from such harms but typically impose some cost on the other party. Thus, in striving to strike a balance between costs and beneﬁts, policymakers often adjust the application of or enforcement of employment-related regulations according to ﬁrm size due to the belief that a given regulation or regulatory policy will impose a greater relative cost on a smaller ﬁrm.

In many cases, the regulations imposed on employers are designed to inﬂuence the health and safety environment for workers.

In the Name of Entrepreneurship?

Regulatory Environment

We can categorize employment-related regulations, rules, or policies under three broad headings: (1) regulations and rules governing employer and employee behavior; (2) workers’ compensation (WC), an administrative compensation program that dictates the remuneration provided to individuals who are injured at or become sick because of their work; and (3) unemployment insurance (UI), a social insurance program that compensates individuals who lose their jobs and are, at least temporarily, unable to ﬁnd new work. We will also discuss mechanisms used to enforce these regulations.13 Regulations. Key regulations and rules limiting employer and employee behavior fall into two broad categories: government regulations and restrictions on contractual form. Government Regulations. Many federal statutes have been put in place over the past 40 years that protect individuals against discrimination or a hostile or unsafe environment in the workplace and that prevent employers from terminating employees in certain protected classes for speciﬁc reasons. Many of these regulations are applied according to size thresholds such that businesses with a small number of employees are not covered. Appendix A provides a summary of such thresholds. Thresholds based on the number of employees in a ﬁrm are particularly common in employment regulation. These size thresholds can be quite low, as in the case of the Fair Labor Standards Act of 1938 (P.L. 75-718), which applies to businesses with two or more employees. The act guarantees a minimum wage and 1.5 times the regular rate of pay for hours worked over 40 hours per week. It also restricts the use of child labor and imposes recordkeeping provisions on employers (DOL, undated[a]). Another regulation with a low employment threshold is the Immigration Reform and Control Act (P.L. 99-603), which restricts employers with four or more employees from discrimi-

It is worth noting one important class of employment regulation that we are explicitly not discussing here: law governing unions and union membership, to which we loosely refer as labor law. There are two main reasons we do not consider it here. The ﬁrst is that fewer workers in the United States are union members than have been in the past, particularly in the private sector. A second, related point is that very few workers in small ﬁrms are unionized.

The Impact of Regulation and Litigation

nating against U.S. citizens, nationals, or authorized aliens on the basis of national origin in hiring, discharge, or referrals. The core federal antidiscrimination acts apply to employers with 15 or more employees. These include Title VII of the Civil Rights Act of 1964 (discrimination in hiring, employment, or termination based on race, color, religion, sex, or national origin; sexual harassment) (P.L. 88-352), the Pregnancy Discrimination Act of 1978 (PDA) (P.L. 95-555), the Equal Pay Act of 1963 (P.L. 88-38), and the Americans with Disabilities Act of 1990 (ADA) (P.L. 101-336). Employers of 20 or more employees must comply with the Age Discrimination in Employment Act of 1967 (ADEA) (P.L. 90-202), which prohibits discrimination against individuals age 40 or over. Employers of this size are also required under COBRA (P.L. 99-272) to provide employees and their families with the opportunity to temporarily extend their healthcare coverage under group heath-care beneﬁt plans (if any) sponsored by the employer in certain cases in which the coverage would otherwise end (e.g., death of the employee, termination, divorce). The Family and Medical Leave Act (P.L. 103-3) requires employers with 50 or more employees to allow employees 12 weeks of unpaid leave for speciﬁc reasons such as the birth or adoption of a child or to care for a seriously ill immediate family member. The federal Occupational Safety and Health Act (OSH Act) (P.L. 91-596), administered by OSHA within the U.S. Department of Labor (DOL), regulates safety and health conditions in most privateindustry workplaces. In general, federal health and safety regulations apply to all ﬁrms, regardless of size, but enforcement practices vary depending on the size of the employer.14 In addition to these federal regulations, 49 of the 50 states plus the District of Columbia and Puerto Rico have their own versions of anti14

Existing research on health and safety issues has suggested that fatality rates are higher at smaller establishments (work sites) across all major industry sectors (Mendeloﬀ and Kagey, 1990; Peek-Asa, Erickson, and Krauss, 1999; Bennett and Passmore, 1985). However, recent research described in this book (speciﬁcally, Mendeloﬀ et al., 2006) reveals that the relationships among ﬁrm size, establishment size, and fatality risk are signiﬁcantly more complicated. That research shows that, after controlling for the size of the establishment (work site), small ﬁrms actually have lower fatality rates than midsized ﬁrms have.

In the Name of Entrepreneurship?

discrimination statutes. In many cases, these statutes are more stringent than the federal regulations, either in terms of the size threshold used to determine the applicability of the antidiscrimination statute or in terms of the type of discrimination covered by the statute. Restrictions on Contractual Form. Employer behavior is also circ*mscribed by restrictions on contractual form that create exceptions to the employment-at-will doctrine. As a general premise, employment in the United States constitutes an at-will contract between an employer and an employee. An employer can terminate an employee for no reason, just as an employee can leave a job for any reason. The employment-at-will doctrine does not (necessarily) apply if an individual works under a contract, including a union contract. In that case, the contract may dictate terms under which an employer may terminate an employment contract. Either party may be subject to a breach of contract claim if the party does not live up to the terms of the contract. Generally, there is no threshold that determines whether a ﬁrm is subject to a lawsuit or claim for violating an employment contract. Businesses can also be inﬂuenced by restrictions on contractual form that address employee behavior. A noncompete agreement is one common type of restriction that may be included in the employment contract. These agreements may prohibit an employee from competing or assisting competitors with the employer by engaging in a related business as an employee, contractor, owner, or investor both while employed and often for some period after employment ends. Similarly, trade-secret rules prevent individuals from making use of information or trade secrets that they acquire on the job to compete with the employer. There is no size threshold of which we are aware that limits an employer’s ability to sue an employee for a similar breach. Workers’ Compensation. Since the early twentieth century, virtually all private-sector employers in the United States have been required to pay WC beneﬁts to employees who are injured at work. WC makes employers liable for medical costs and partial income replacement for workers injured at or because of their jobs. WC functions as a “carveout” of the tort system and acts as more of a social insurance system than as a regulatory program. WC beneﬁts cover some portion of lost wages, as well as direct costs associated with the injury itself. While

The Impact of Regulation and Litigation

state WC programs have frequently been subject to reform over the years, there have been few wholesale changes. Moreover, WC programs make relatively few distinctions among employers with diﬀerent characteristics. Thus, the coverage oﬀered by one ﬁrm will be very similar, if not identical, to that oﬀered by another, regardless of important differences such as industry, the level of risk, or ﬁrm size.15 Fourteen states have established thresholds, ranging from one to ﬁve, that exempt very small ﬁrms from having to provide WC coverage (DOL, undated[b]). Employers are required either to purchase insurance to cover potential WC losses or to demonstrate suﬃcient ﬁnancial resources to selfinsure. For employers who purchase insurance, the market for such insurance is characterized by an experience rating, a tool used by insurers to adjust premiums based on previous claim history and the implementation of speciﬁc safety measures. Unemployment Insurance. UI systems are social insurance programs that provide income-replacement beneﬁts to individuals who are unemployed through no fault of their own. Each state administers a separate program, though federal law has established guidelines. Beneﬁts are paid from taxes on employer payrolls. There are very low size thresholds for employers to be liable for the UI payroll taxes. Generally, employers are required to pay taxes if they pay wages of $1,500 or more in any quarter of a calendar year or if they had at least one employee on any day of a week during 20 weeks in a calendar year. The standards are higher for agriculture employers, with the wage requirement of $20,000 or more or at least 10 employees. While both UI and WC are forms of social insurance and provide partial income-replacement beneﬁts, there are some diﬀerences, particularly in the ﬁnance mechanisms for the programs, which are important for thinking of how they inﬂuence small businesses. Unlike WC, which is structured and ﬁnanced as an insurance program (with ﬁrms paying premiums calculated using experience ratings), UI beneﬁts are

In addition to the size thresholds, some public employees, most notably public-safety employees, receive diﬀerent coverage. The coverage aﬀorded to agricultural employees varies signiﬁcantly across states, with only 14 states covering agricultural employees the same way as other employees (as of 2006).

In the Name of Entrepreneurship?

paid out of a government fund collected as part of a payroll tax. There is no size threshold for participation in the system, and essentially any business that maintains at least one employee is responsible for paying the tax. There are federal and state payroll taxes. The federal tax is equal to 6.2 percent of the ﬁrst $7,000 in wages paid to each employee during the calendar year. The state taxes vary both in the taxation rate and the wage base. In the event that a worker loses his or her job due to lack of work at the place of employment or for some other eligible cause, he or she receives a payment from the government until the worker either ﬁnds a job, violates some aspect of the state eligibility criteria, or exceeds the maximum beneﬁt duration. The payment amount varies by state and typically depends on the worker’s earnings prior to job loss up to some maximum amount. The duration of beneﬁts also varies by state but is typically 26 weeks. The duration of beneﬁts is often extended during times of high unemployment. Enforcement Mechanisms. We will now discuss administrative enforcement and court enforcement mechanisms. Administrative Enforcement of Government Regulations. Several of the regulations or statutes have employer size thresholds that determine the coverage of the regulation. Employers that fall below the threshold cannot be sued, ﬁned, charged, or otherwise reprimanded under the aegis of the law. In addition, these employers are not subject to the recordkeeping requirements associated with these laws. For example, EEOC requires that employers keep all personnel and employment records for one year, and ADEA requires employers to retain payroll records for three years. As part of antidiscrimination enforcement, EEOC requires employers with 100 or more employees (or 50 or more employees and federal contracts totaling more than $50,000) to ﬁle an Employer Information Report EEO-1 that characterizes the workforce by race and gender. EEOC has the legal standing to investigate employee complaints against ﬁrms in the area of antidiscrimination to determine whether there has been a violation of the law. If the government investigation establishes a violation of the law, EEOC may oﬀer mediation to the disputing parties or attempt to settle with the employer. If these attempts

The Impact of Regulation and Litigation

fail, then EEOC may choose to sue the employer. Punitive damages beyond a remedy may also be sought if the violation is deemed to be intentional, malicious, or recklessly indiﬀerent. OSHA regulates safety and health. Safety standards cover hazards such as falls, explosions, ﬁres, and cave-ins, as well as machine and vehicle operation and maintenance. Health standards regulate exposure to a variety of health hazards through engineering controls, the use of personal protective equipment (e.g., respirators, hearing protection), and work practices. Employers covered by the OSH Act (P.L. 91-596) are required to maintain safe and healthful workplaces. These employers must become familiar with job safety and health standards applicable to their establishments, comply with the standards, and eliminate hazardous conditions, to the extent possible. Where OSHA has not set forth a speciﬁc standard, employers are responsible for complying with the OSH Act’s “general duty” clause, which states that each employer “shall furnish . . . a place of employment [that] is free from recognized hazards that are causing or are likely to cause death or serious physical harm to [its] employees.” The act assigns OSHA two regulatory functions: setting standards and conducting inspections to ensure that employers are providing safe and healthful workplaces. OSHA regulations cover such items as recordkeeping, reporting, and posting. OSHA covers every employer with more than 10 employees (with employer deﬁned as an establishment or workplace rather than as a ﬁrm), except for employers in certain low-hazard industries in the retail, ﬁnance, insurance, real estate, and service sectors. These employers must therefore maintain three types of OSHA-speciﬁed records of job-related injuries and illnesses. In addition, every employer, regardless of industry category or the number of its employees, must advise the nearest OSHA oﬃce of any accident that results in one or more fatalities or the hospitalization of three or more employees. Although small ﬁrms are not exempt from health and safety regulations, OSHA has developed a special consultation program that is available to ﬁrms with fewer than 500 employees (DOL, 2007). The program is conﬁdential and separate from the inspection process and is designed to help small ﬁrms identify and correct potential workplace hazards, thereby complying with regulations. OSHA considers the

In the Name of Entrepreneurship?

employing ﬁrm’s size, among other factors, when determining the penalty to be proposed for any violation. Penalties are generally reduced by 60 percent if an employer has 25 or fewer employees, 40 percent if the employer has 26 to 100 employees, and 20 percent if the employer has 101 to 250 employees. While OSHA does engage in regulatory enforcement, it also encourages states to develop and operate their own safety and health programs.16 These plans are subject to OSHA approval. In 2007, 24 states have OSHA-approved state plans.17 OSHA is responsible for enforcement in the remaining states. The state entities enforce their own safety and health standards, which are at least as strict as federal OSHA requirements, but may have diﬀerent or additional requirements. Many states oﬀer additional programs of assistance to small businesses. Litigation and Court Enforcement of Employer and Employee Behavior. An employee who believes that his or her employer has inten-

tionally harmed him or her has the right to ﬁle a civil suit and seek damages from the employer. The threat of such legal action exists for all employers on a wide range of matters. Similarly, an employer of any size may use the court system to enforce an employment agreement if an employee fails to live up to the terms of a contract. Federal regulations, as well as the parallel regulations at the state and local levels, create additional avenues through which ﬁrms may be punished for speciﬁc types of harms. In general terms, this is accomplished through the establishment of a government entity that is given the legal authority to monitor business activities on speciﬁc employment-related matters and to investigate or respond to complaints. The government entity may also have the authority to sue ﬁrms, impose ﬁnes, or promote mediation between employers and employees.

16 17

State plans are authorized under section 18 of the OSH Act (P.L. 91-596).

OSHA (2006). Alaska, Arizona, California, Connecticut, Hawaii, Indiana, Iowa, Kentucky, Maryland, Michigan, Minnesota, Nevada, New Jersey, New Mexico, New York, North Carolina, Oregon, South Carolina, Tennessee, Utah, Vermont, Virginia, Washington, and Wyoming have OSHA-approved state plans, although the Connecticut, New Jersey, and New York plans cover only state and local government employers.

The Impact of Regulation and Litigation

An employer of any size may be subject to a civil suit brought by an employee who claims that the employer has harmed him or her. Firms over speciﬁc size thresholds may also face criminal suits, civil suits brought by a government agency, or ﬁnes for alleged regulatory violations of speciﬁc types. Restrictions on contractual form that limit the behavior of either the employer or the employee (e.g., noncompete agreements, exceptions to the employment-at-will doctrine) are remedied through the court system, which may or may not ﬁnd these agreements to be legal or binding. States vary substantially in terms of their willingness to enforce such contract terms. Effects of Employment Regulations on Small Businesses and the Small-Business Response Enforcement Variation. Two kinds of enforcement are relevant to

the discussion of the small-business response: administrative enforcement of regulations and court enforcement policies. Administrative Enforcement of Regulations. A ﬁrm of any size is potentially at risk of a civil action in response to a claim that the ﬁrm’s actions have harmed an employee. However, federal, state, and local regulations increase the risk by giving a government agency the authority to investigate ﬁrm behavior and take legal action. This implies that very small ﬁrms falling below the employment threshold for a regulation may face a lower risk of legal action in this area. The beneﬁts of staying below a certain size threshold might induce some ﬁrms to limit growth. For example, in the antidiscrimination arena, in which the regulations are fairly vague and therefore the risk associated with legal action may be higher, it is plausible that very small businesses might consider the employment threshold stipulated by the regulation and limit growth to avoid the reporting requirements and the threat of ﬁnes or legal action. In addition to the legal risk, administrative enforcement imposes information-gathering burdens (such as the requirement to ﬁle an EEO-1 report) on ﬁrms larger than a certain size threshold. Some ﬁrms might also try to remain smaller than the reporting threshold to avoid the costs of gathering, reporting, and maintaining required information.

In the Name of Entrepreneurship?

Existing research suggests that, in some industries, large, unionized ﬁrms are more eﬃcient in implementing health and safety programs, in the sense that the per-worker cost of such programs is lower (Bartel and Thomas, 1987). This research shows that, as a result of regulation, large, unionized ﬁrms are more proﬁtable at the expense of small, nonunionized ﬁrms. This is consistent with a hypothesis that there are substantial ﬁxed costs involved in complying with health and safety regulations. Court Enforcement Policies and Regulations. Although court enforcement does not explicitly consider ﬁrm size, one of the key diﬀerences between small and large ﬁrms is the level of resources available to spend on litigation, either as plaintiﬀs or defendants. There are several means through which these resources might inﬂuence the prospects of larger ﬁrms in the tort system. For actions that would be initiated by employees (e.g., discrimination, wrongful discharge, violation of the employment contract), employees (or, more realistically, lawyers) might be likelier to go after large ﬁrms with deep pockets. On the other hand, large ﬁrms with deep pockets might have a stronger incentive to spend substantial resources aggressively defending any one suit to deter future suits. This threat of deterrence might make employees less likely to go after large ﬁrms because they perceive a lower chance of winning. For legal actions that would be initiated by ﬁrms (e.g., violation of a noncompete agreement, trade-secret suit) and for which the ﬁrm seeks redress from employees, large ﬁrms can spend more resources litigating against employees who violate these agreements and may have a stronger incentive to do so to deter other employees from violating these contract clauses in the future. Small ﬁrms are, by nature of their size, likelier to face bankruptcy due to a costly legal action. Of course, small ﬁrms may be more vulnerable to breach of a noncompete agreement or violation of trade-secret rules as the entire business may depend on that trade secret. As a result, they may be likelier to prosecute, in spite of the costs and the risks of bankruptcy. It is currently diﬃcult to assess whether restrictions on the contractual form that employers use to restrict employee behavior aﬀect small businesses diﬀerently from how they aﬀect large ones. Nor is there any

The Impact of Regulation and Litigation

indication that small businesses are prevented from using noncompete agreements. However, because the agreements can be enforced only through litigation, a small business may face a greater burden in enforcing such a clause. On the other hand, these businesses may also have more to lose in the event that such an agreement is violated. Moreover, noncompete agreements may impact labor supply in a way that has a particularly strong impact on entrepreneurship or small businesses. For example, the natural labor pool for start-ups may include individuals who recently worked for a larger company in the same industry. Noncompete agreements might prevent current employees from leaving a company to start their own businesses but might also hinder employers’ ability to hire individuals who had worked for a competitor. Overall, economic theory points to conﬂicting forces regarding the question of whether the threat of or use of lawsuits places a greater burden on small businesses than on large businesses. In the end, it is diﬃcult to determine who bears a heavier burden from legal costs. A recent study by Pendell and Hinton (2007) suggests that the legal costs per dollar of revenue are substantially greater for small businesses than they are for large businesses. Small businesses, which Pendell and Hinton deﬁne as those with less than $10 million in annual revenue and at least one employee in addition to the owner, account for 19 percent of business revenue but 69 percent of tort-liability costs. Very small businesses, deﬁned as those with less than $1 million in annual revenue, account for 6 percent of business revenue and 21 percent of tort-liability costs. The authors ﬁnd that the tort-liability costs per $1,000 in revenue decline steadily as the revenue category increases. For ﬁrms with revenue less than $1 million, the cost per $1,000 in revenue is $20.84; for ﬁrms with revenue greater than $50 million, that ﬁgure is $1.33. Impact of the Workers’ Compensation System. Despite the relative uniformity of coverage, there are good reasons to believe that WC might have a diﬀerential impact on diﬀerent-sized ﬁrms. Employers are required either to purchase insurance to cover potential WC losses or to demonstrate suﬃcient ﬁnancial resources to self-insure. Large ﬁrms typically have a greater ability to self-insure their beneﬁt payments because they are better able to bear the risks involved in a WC claim,

In the Name of Entrepreneurship?

which could require a large payout in one period. An ability to selfinsure can reduce the expected costs of WC by allowing ﬁrms to bypass an insurance system with two possible sources of ineﬃciency. First, many argue that the WC-insurance market is not perfectly competitive and therefore that premiums exceed the expected value of insurance payouts. Second, experience rating, a tool used by insurers to adjust premiums based on previous claim history and the implementation of speciﬁc safety measures, is imperfectly applied. This imperfect application of experience rating tends to work to the relative disadvantage of smaller ﬁrms, resulting in less reliable experience on which to base premiums. Experience rating is a potentially important tool because it allows insured ﬁrms to reap some beneﬁts from investments in safety. Absent any bias in the application of experience rating, smaller ﬁrms might still ﬁnd it costlier to promote safety measures if the implementation of those measures involves substantial ﬁxed costs because there will be fewer workers over whom to spread the ﬁxed costs. If smaller ﬁrms are imperfectly experience rated, this further reduces the incentive to promote workplace safety. All of this suggests that WC insurance will be costlier for smaller employers than for larger ones. In addition to the higher costs, we might also expect that the outcomes for the injured workers are worse at smaller ﬁrms. If, as suggested, smaller ﬁrms have fewer incentives or less ability to implement eﬀective safety measures, then we might expect workers to suﬀer injuries with greater frequency and perhaps greater severity at small ﬁrms. Large ﬁrms are also likelier to be able to accommodate injured workers with modiﬁed work, and return to work is a critically important predictor of the long-term impact of disability (as shown in Peterson et al., 1998, and Reville et al., 2001). All of this suggests that WC in particular, and occupational health and safety programs more generally, might lead to signiﬁcantly diﬀerent costs for employers and workers depending on employer size. Speciﬁcally, there are reasons to believe that employers will face higher WC costs, while workers will face worse potential outcomes from jobrelated injures at smaller ﬁrms. Given that WC premiums represent approximately 2 percent of all payroll costs nationwide, this could have substantial implications for the operating costs of small businesses.

The Impact of Regulation and Litigation

Impact of the Unemployment Insurance System. Since the UI system covers all employers and there is no opportunity to self-insure, we do not expect the business response to the UI system to vary dramatically by ﬁrm size. In principle, the taxes that an employer pays should be proportional to the number of people it employs, reducing any disparities between ﬁrms of diﬀerent sizes. In reality, an employer is experience rated, in the sense that the tax rate it pays depends on its past experience with unemployment. New employers are assigned a ﬂat rate, and, over time, their rates will change based on the stability of their labor force and the number of layoﬀs they experience. In California in 2002, for example, the base rate for new employers was 3.4 percent, while the minimum rate was 0.7 percent and the maximum rate 5.4 percent. This has the potential to beneﬁt large ﬁrms more than it might small ones, because large ﬁrms are probably more ﬂexible in response to changing economic conditions and may be able to avoid some layoﬀs that would increase their payroll taxes. Smaller ﬁrms might not be able to absorb the cost of a worker when faced with lower demand or higher costs and will bear the full brunt of unemployment taxes. Just how strong this eﬀect would be is an empirical matter. Section Conclusion

This section has discussed a number of ways that policy instruments designed to regulate aspects of the employee-employer relationship might have intended and unintended consequences that impose higher costs on small businesses than it does on larger businesses. Overall, there is little empirical research that documents the eﬀects of workplace regulations in general and the eﬀects on small ﬁrms in particular. As noted, there are many federal statutes that protect individuals against discrimination or a hostile or unsafe environment in the workplace and that prevent employers from terminating employees in certain protected classes for speciﬁc reasons. Many of these federal rules are applied using size thresholds such that the regulation does not cover businesses with a small number of employees. Similar laws that exist at the state and local levels supplement these federal statutes. Because these regulations increase the risk of legal action by establishing a gov-

In the Name of Entrepreneurship?

ernment agency with the authority to investigate ﬁrm behavior and take legal action, the very small ﬁrms that fall below the employment threshold for a regulation may face a lower risk of legal action in this area. There is little empirical research examining the eﬀectiveness of the exemption thresholds that do exist. In general, more careful consideration of unintended consequences and of the beneﬁts of policy and regulation in the workplace setting is warranted.

Health-Insurance Regulations Firms face myriad regulations governing the HI coverage that they oﬀer to their employees. However, HI regulations are often targeted to companies that sell group HI products to ﬁrms rather than toward the ﬁrms that oﬀer HI to their employees. Although HI regulation has emerged in response to a general concern about the cost of access to HI, there is particular concern about access for individuals who are employed by small businesses. Nearly three-fourths of employed Americans obtain HI through an employer. However, while 79 percent of workers in large ﬁrms have employerprovided HI, only 36 percent of workers in small ﬁrms have such coverage. This diﬀerence stems mostly from the fact that small ﬁrms are substantially less likely to oﬀer HI coverage than large ﬁrms are. In particular, only 40 percent of ﬁrms with fewer than 10 workers oﬀer HI, compared to 99 percent of ﬁrms with more than 500 workers (authors’ calculations from AHRQ, 2000). The diﬃculties that small ﬁrms face in obtaining and maintaining HI for their employees have been widely documented (Charles Brown, Hamilton, and Medoﬀ, 1990; McLaughlin, 1993; Fronstin and Helman, 2000). Among small ﬁrms that oﬀer coverage, HI is routinely cited as the most salient area of concern (NFIB, undated). The low proportion of small ﬁrms oﬀering HI coverage has been attributed, in part, to the high cost of HI for small ﬁrms, the low demand for HI among workers in these ﬁrms, and insurers’ unwillingness to take on small-ﬁrm risks (McLaughlin, 1993; Fronstin and Helman, 2000; Monheit and Vistnes, 1999).

The Impact of Regulation and Litigation

Regulatory Environment Regulation. The intention of small-ﬁrm HI regulation has consis-

tently been to make HI more accessible and aﬀordable for small-ﬁrm employees. Several goals have dominated the policy landscape (Blumberg and Nichols, 1995). Because small ﬁrms may be disadvantaged relative to large ﬁrms, reforms have aimed to extend economies of size to small ﬁrms. Small ﬁrms are disadvantaged relative to large ﬁrms in two ways: First, HI companies face substantially higher administrative costs in insuring small ﬁrms than they do in insuring large ones, which has led to higher premium levels for small ﬁrms; second, small ﬁrms have limited opportunities to share health-care risks with other individuals or groups, which can threaten access to HI for some small ﬁrms, especially those with employees with potentially higher health expenses. Policy reforms have also aimed to promote competition in the private HI market. A competitive HI market is likely to lead to more eﬃcient delivery of health services, higher service quality, and decreases in average premium levels. The more informed the purchasers of services are, the more that providers of services will compete on quality and cost. Therefore, an important goal of policy is to promote various avenues for small ﬁrms to obtain HI. Policymakers have had to balance insurers’ concerns against small ﬁrms’ needs. Insurers have resisted stringent regulations on premiums and underwriting and have stopped doing business in some states that have had excessively restrictive regulations (Epstein, 1996). As a result, in many cases, HI regulation has been weak and ineﬀective. In other cases, as discussed later in this section, regulations have had unintended consequences on labor-market outcomes and possibly on business size. Table 2.3 lists major federal regulations related to HI. We discuss these regulations in relation to three topics: HI access, small-group HI, and HI beneﬁts. Health Insurance Access. Many federal and state regulations aim to increase access to HI for ﬁrms and individuals.18

For another source of information on HI, see U.S. General Accounting Oﬃce (2003a).

In the Name of Entrepreneurship?

ERISA preempts state regulation of self-insured health plans. Essentially, this implies that ﬁrms can avoid potentially burdensome state HI regulations by choosing to self-insure rather than purchasing coverage from an HI company. COBRA (P.L. 99-272) provides continuation of group health coverage that otherwise would be terminated when an employee leaves a job. COBRA applies to ﬁrms that employ more than 20 workers. COBRA contains provisions giving certain former employees, retirees, spouses, and dependent children the right to temporary continuation of health coverage at group rates. Job leavers are entitled to continue purchasing group coverage from their former employers for up to 18 Table 2.3 Major Federal Regulations Governing Health Insurance Law

Main Purpose or Relevance for HI

ERISA (P.L. 93-406)

Preempts state regulation of self-insured health plans, thus implying that ﬁrms can choose to self-insure rather than purchasing coverage from an HI company. Sets uniform, minimum standards to ensure that employee-beneﬁt plans are established and maintained in a fair and ﬁnancially sound manner.

COBRA (P.L. 99-272)

Provides continuation of group health coverage that would otherwise be terminated when an employee leaves a job.

HIPAA (P.L. 104-191) Adds provisions to ERISA that are designed to provide participants and beneﬁciaries of group health plans with improved portability and continuity of HI coverage. Limits scope and length of exclusion periods for people with preexisting conditions in group health plans and prohibits cancellation of health coverage due to illness. ADA (P.L. 101-336)

Prohibits employers from discriminating against individuals with disabilities in the provision of HI but allows medical underwriting.

Mental Health Parity Stipulates that, if mental-health coverage is offered, dollar Act of 1996 (MHPA) limits on mental-health beneﬁts have to be equal to dollar (P.L. 104-204) limits on medical beneﬁts. PDA (P.L. 95-555)

Requires businesses with 15 or more employees to cover expenses for pregnancy and medical conditions related to pregnancy on the same basis as it does for other medical conditions.

The Impact of Regulation and Litigation

months after a separation at a maximum of 102 percent of the employer’s average group rate. Many states have continuation-coverage laws. These laws specify that employees and their families may continue coverage for a speciﬁed length of time after job termination. Unlike COBRA, state continuation-coverage laws apply to all ﬁrms that are not exempt from state regulation by ERISA. The majority of states (39) extended the federal COBRA requirements to individuals covered by group HI provided by businesses with fewer than 20 employees (U.S. General Accounting Oﬃce, 2003a). HIPAA (P.L. 104-191) added several provisions to ERISA that are designed to provide participants and beneﬁciaries of group health plans with improved portability and continuity of HI coverage. The HIPAA portability provisions relating to group health plans and HI coverage oﬀered in connection with group health plans are designed to improve access to HI and protect against discrimination on the basis of health status.19 HIPAA also prohibits a person’s health coverage from being canceled because of sickness. Moreover, HIPAA requires that HI coverage be guaranteed issue and renewable for small employers (2 to 50 employees). HIPAA also prohibits any employer-sponsored health coverage from charging employees a higher premium based on healthrelated factors. However, it is important to note that these requirements regulate only the premium charged to employees, not the premium faced by employers. ADA (P.L. 101-336) is a federal law focusing on employment and other rights for the disabled, including several provisions that apply to employer-provided HI. ADA prohibits employers from discriminating against individuals with disabilities in the provision of HI; however, ADA explicitly allows medical underwriting.20 An employer who treats individuals with disabilities diﬀerently from others under an HI or beneﬁt plan because the people who are disabled represent increased risks or costs is not in violation of the ADA if the employer treats the disabilities in the same manner as other conditions of the same risks or 19

For further details on HIPAA, see HHS (2007).

See DOJ (2005) for details on ADA.

In the Name of Entrepreneurship?

costs. While an employer must provide people with disabilities equal access to the HI coverage provided to all employees, the employer may oﬀer a policy that limits the number of treatments or excludes certain conditions from coverage that are not disability based. ADA applies only to businesses with more than 15 employees. Small Group HI. During the 1990s, most states passed laws regulating the terms and conditions of HI provided to small ﬁrms (Monheit and Cantor, 2004). States have tended to pass these reforms in packages that generally contain the following provisions: • Guaranteed-issue and -renewal laws. Every state (except Georgia) that has passed small-group HI reform has included guaranteedrenewal reform in its package. This reform requires insurers to renew coverage for all groups, except in cases of nonpayment of premium or fraud. Guaranteed-issue legislation, on the other hand, is excluded from the reform packages of eight states that have passed guaranteed renewal laws. These laws have now been preempted by the federal HIPAA. • Preexisting-condition exclusion laws. Health plans often impose waiting periods for coverage generally or coverage for preexisting health conditions. In some instances, health plans permanently exclude coverage for speciﬁc health conditions. State reforms limit the length of time for which preexisting health conditions can be excluded from coverage. HIPAA reinforces preexisting-condition exclusion limitations. • Portability reforms. Portability reforms ensure that an individual who is covered by HI on a previous job does not face any new preexisting-condition exclusions or waiting periods as a result of changing jobs. However, portability reforms do not place any restrictions on either premiums that HI companies charge small ﬁrms or premium contributions that ﬁrms charge workers. • Premium rating reforms. State reforms have placed restrictions on the factors that can be used to set HI premiums and have limited the rate variations to speciﬁed ranges. The reforms restrict the variation in premiums that an insurer can charge to ﬁrms within each of a set number of classes and restrict the variation

The Impact of Regulation and Litigation

allowed between business classes. Most states allow nine business classes, with about 15 to 30 percent premium variation within and between classes. The use of factors such as group size, family type, age, and other demographic variables to set premiums is generally allowable. • Reinsurance provisions. An administered reinsurance mechanism allows individual insurers to reinsure any ﬁrms that are expected to generate costs exceeding the prices of HI. Reinsurance allows insurers to pass their highest-risk clients over to an industry-funded reinsurance pool. This outlet encourages insurers to accept all clients, making it less risky for insurers with small market shares to remain in a market with guaranteed issue. HI Beneﬁts. Another set of federal regulations place restrictions on the package of beneﬁts oﬀered by health insurers. MHPA (P.L. 104-204), which took eﬀect in January 1998, stipulates that, if mental-health coverage is oﬀered, the dollar limits on mental-health beneﬁts have to be equal to dollar limits on medical beneﬁts. The parity legislation exempts businesses with fewer than 50 employees. PDA (P.L. 95-555) requires businesses with 15 or more employees to cover expenses for pregnancy and medical conditions related to pregnancy on the same basis as coverage for other medical conditions.21 State-mandated HI beneﬁt laws regulate the services that insurers must cover to sell HI in a state. While all states have mandated that certain beneﬁts be covered by HI policies, the number, type, and scope of the states’ requirements vary substantially. According to a survey published in 2002, the total number of mandated beneﬁts varied among states from fewer than 10 in ﬁve states to more than 30 in seven states. The two most commonly mandated beneﬁts, required by 43 or more states, were mammography screening and diabetic supplies. States also have provider mandates that specify that insurers must 21

The Newborns’ and Mothers’ Health Protection Act of 1996 (NMHPA) (P.L. 104-204) mandates a minimum length of stay after childbirth; however, this legislation applies equally to businesses of all sizes that oﬀer HI. Similarly, the Women’s Health and Cancer Rights Act of 1998 (WHCRA) (P.L. 105-277) requires that employer-sponsored HI cover reconstructive surgery after mastectomies—this regulation also applies to ﬁrms of all sizes.

In the Name of Entrepreneurship?

cover the services of certain providers, such as chiropractors, psychologists, and optometrists. Most states allow the sale of bare-bones policies that do not need to comply with beneﬁt mandates (U.S. General Accounting Oﬃce, 2003a). Policies to Assist Businesses in Providing Insurance. In addition to these regulations, several policies have been developed to aid businesses, particularly small businesses, in providing HI. Premium Assistance Programs for Small Businesses. Tax incentives are often used as a tool for encouraging access to HI for small businesses. A common strategy is for small businesses to receive transitional tax credits when they insure for the ﬁrst time. More than 10 states oﬀer some form of tax incentives. Another form of premium assistance relies on direct subsidies to small businesses. Several states have implemented programs to subsidize premiums using public dollars under HIPAA. In some states, the employee is subsidized directly, while, in others, the subsidy goes through the employer (U.S. General Accounting Oﬃce, 2003a; Williams, 2003). Purchasing Pools. Some state small-group HI reforms have included provisions to establish a publicly sponsored purchasing pool for small employers. In a purchasing cooperative, ﬁrms join together to purchase HI in larger volumes at more aﬀordable prices, thereby aiming to diversify risk and reduce administrative costs. Most purchasing cooperatives are private, but a few states have public cooperatives. By one estimate, almost one-third of small ﬁrms purchase their HI through some form of cooperative purchasing arrangement (Long and Marquis, 1999). Health and Medical Savings Accounts. Health savings accounts (HSAs) or medical savings accounts (MSAs) are tax-free savings accounts for medical expenses.22 In 2007, taxpayers with highdeductible health plans (HDHPs) can contribute up to $2,850 per year ($5,650 for families) into an HSA. Both employers and employees may contribute to these accounts. While HSAs are not explicitly targeted to

The MSA was a precursor of the HSA, was authorized as a demonstration project in 1996, and was not reauthorized when the demonstration period ended.

The Impact of Regulation and Litigation

small businesses, it is yet to be seen whether small businesses are more or less attracted than large businesses are to HDHPs. Patient and Provider Protections. Federal and state laws have also established a number of laws and standards to protect the interests of patients and providers. ERISA (P.L. 93-406) lays out administrators’ ﬁduciary standards (to administer the plan in the best interests of beneﬁciaries) and provides for internal review of denied claims, requirements for plan descriptions to be given to enrollees, and reporting to the federal government. ERISA sets uniform, minimum standards to ensure that employeebeneﬁt plans are established and maintained in a fair and ﬁnancially sound manner. ERISA also sets out requirements for managing and administering private pension and welfare plans. Most states have enacted laws to regulate the nature of the provider panels created by managed-care ﬁrms and the administration of managed-care beneﬁts. The extent to which ERISA (P.L. 93-406) preempts these laws is still unknown. These laws can generally be grouped into three categories: • Any willing provider. These laws require managed-care plans to allow any provider to be included in the network if he or she is willing to abide by the network contract’s terms and conditions. • Freedom of choice. These laws require that a managed-care subscriber be allowed to obtain services outside the network from any licensed provider as long as the subscriber pays more out of pocket than would be necessary to pay for an in-network provider. • Review of denied claims. States have stipulated a process for the internal review of denied claims, although considerable variation exists among these processes, including determining which claims are eligible for review. Most states have also mandated an external review process that requires an independent, external review of denials by managed-care companies. Many states have enacted patient-protection laws to protect consumers through means such as the availability of a point-of-service option. Most states also have patient protections that allow patients

In the Name of Entrepreneurship?

direct access (without prior approval) to certain health-care providers and services, such as emergency services and obstetricians and gynecologists. Most states also prohibit gag clauses in insurers’ contracts with health-care providers. These laws allow a provider to inform a patient about treatments that are not covered by his or her HI policy. States have also regulated utilization review by requiring registration and accreditation. In addition, states have instituted solvency requirements and reporting requirements. Many of these regulations are more stringent than ERISA. Most states require health insurers to maintain a certain minimum level of reserves (Charette, 1995). Effects of Health-Insurance Regulations on Small Businesses and the Small-Business Response

The regulations reviewed in this section might lead small businesses to respond diﬀerently from larger businesses to maximize the beneﬁt and minimize the negative impact of regulation. We consider the eﬀect of the regulatory environment on small businesses in four areas: HI oﬀering, premiums, labor-market outcomes, and business size. Effect on HI Coverage and Premiums. HI regulations that aﬀect small ﬁrms diﬀerently from how they aﬀect large ﬁrms might be expected to impact the likelihood that small ﬁrms will oﬀer HI coverage or lead to changes in HI premiums. Furthermore, the cost of state regulation might inﬂuence a small ﬁrm’s decision on whether to self-insure. Self-insurance is an attractive option for larger ﬁrms that are able to diversify variations in worker medical costs internally. However, small ﬁrms are less able to do this and therefore are usually unable to self-insure. A recent survey found that 10 percent of covered workers in all small ﬁrms (3–199 workers) are in self-insured plans, compared to 50 percent of workers in midsized ﬁrms (200–999 workers) and 79 percent of workers in jumbo ﬁrms (5,000 or more workers) (KFF/ HRET, 2003). As a result, small ﬁrms are likelier than large ﬁrms are to face the burden of state HI regulations even if there are no explicit size thresholds in the state legislation. Research examining the eﬀect of state insurance mandates on HI coverage, ﬁrms’ propensity to oﬀer coverage, and HI premiums gener-

The Impact of Regulation and Litigation

ally has shown a small eﬀect or no eﬀect on small ﬁrms’ propensity to oﬀer HI or on employees’ insurance coverage (Sloan and Conover, 1998; Jensen and Morrissey, 1999; Zuckerman and Rajan, 1999; Monheit and Schone, 2004; Buchmueller and DiNardo, 2002; Marquis and Long, 2001). A few studies do ﬁnd modest eﬀects of the reforms on insurance oﬀer rates and insurance coverage; however, the direction of the eﬀects varies among the studies (Uccello, 1996; Hing and Jensen, 1999; Simon, 2005; Buchmueller and Jensen, 1997). The Health Insurance Association of America estimates that guaranteed-issue provisions have a small impact on premiums, equal to about 2 to 4 percent. Another line of research has examined the eﬀects of mandated beneﬁts on HI costs and oﬀerings. Two studies have estimated that the additional costs associated with state-mandated beneﬁts represented about 3 to 5 percent of total premium costs (Gruber, 1994a, 1994b). Another study ﬁnds little eﬀect of mandated beneﬁts on HI coverage among employees in small ﬁrms, primarily because most ﬁrms oﬀer comprehensive beneﬁts even in the absence of the mandates (U.S. General Accounting Oﬃce, 2003a). In general, it appears that HI regulations have had little eﬀect on premiums or HI coverage. Proponents of small-group HI reforms are likely to ﬁnd these results disappointing. The lack of realizable eﬀect on HI may be due to the lack of eﬀective price controls. In most states, limitations on premium increases and premium rating factors are weak. As a result, guaranteed-issue provisions mandating coverage for small groups or continuation-coverage laws mandating all employees to be oﬀered coverage have a muted eﬀect, since the available coverage is too expensive for small ﬁrms. Without stronger and more eﬀective premium regulation, we are unlikely to see the HI changes that policymakers had envisioned. Nonprice factors such as the relative ineﬀectiveness of small-ﬁrm purchasing alliances and the lack of information on HI alternatives may also be partly responsible. Effect on Workforce Composition. HI can be thought of as a ﬁxed cost associated with hiring someone. HI regulation can aﬀect this cost, and businesses might respond to regulation by changing their hiring practices. For small ﬁrms, regulation might aﬀect an employer’s choice of hiring full-time versus part-time workers. If a ﬁrm oﬀers

In the Name of Entrepreneurship?

HI, it must be oﬀered to full-time employees, but the oﬀering is not required for part-time employees. In addition, small ﬁrms may prefer to hire personnel with demographic characteristics associated with low and stable HI premiums. Empirical evidence suggests that the workforce composition in small ﬁrms might have shifted as a result of HI regulations and costs (Kapur, 2004). Similarly, workforce composition might be aﬀected by the possibility of health and safety violations in small ﬁrms. A business might also respond to the regulatory environment by attempting to reduce its compensation costs by encouraging employees with working spouses to take family coverage from the other employer. The use of practices of this sort might vary by ﬁrm size. For example, earlier research found that about 10 percent of employees and dependents with coverage from a large ﬁrm have a working family member in a small ﬁrm, suggesting that large employers may “subsidize” small employers (Monheit and Vistnes, 1994). Some empirical support has been presented for the idea that employers adopt strategies, such as raising contribution rates, to encourage employees to use HI coverage provided by the employer of a spouse or another family member. However, there is further scope for research that examines whether small ﬁrms systematically shift their workforce composition to reduce HI costs. Even though the primary goal of HI regulation has been to improve HI access and aﬀordability, the policy community has also claimed that such regulations have reduced labor-market distortions by, for example, providing HI protections to certain groups of people. While there is evidence that small-group policy reform may have reduced distortions for some groups (such as workers with certain preexisting medical conditions), others who are not explicitly protected by the reforms might now face higher costs. In addition, with rising HI costs, we are likely to see a growing trend toward hiring part-time and other types of employees who do not have to be oﬀered HI. Effect on Worker Turnover. HI regulations might also potentially aﬀect worker turnover. For instance, small-group HI legislation and HIPAA include portability and preexisting-condition exclusion provisions that might make it easier for individuals to accept jobs at small ﬁrms. However, two existing studies that examine the labor-market

The Impact of Regulation and Litigation

eﬀects of small-group HI reform, including job mobility, ﬁnd little or no eﬀect on mobility; however, no research has examined the labormarket implications of HIPAA (Kaestner and Simon, 2002; Kapur, 2003). Other research has examined the eﬀect of continuationcoverage mandates and has found that these mandates increase mobility and therefore reduce labor-market ineﬃciency (Gruber and Madrian, 1994). Further research on worker turnover may beneﬁt from focusing more closely on the variation in state continuation-coverage laws and their eﬀect on small businesses. The lack of controls on premiums is likely to be the most important factor preventing HIPAA from increasing transitions and HI coverage. Policymakers will need to balance pricing policies with the potential distortions that might arise from instituting price regulations. Effect on Business Size. Virtually no prior research has examined the eﬀect of HI regulations on business size. The explicit size thresholds in many HI regulations suggest that ﬁrms considering changing their workforce size might be inﬂuenced by HI regulations. For instance, in the case of small-group HI regulations, small ﬁrms that can obtain HI that is protected by small-group regulations might choose not to expand beyond the upper size threshold. On the other hand, if the regulations result in higher premiums and lower availability, small ﬁrms might prefer to expand to a size that is beyond the reach of small-ﬁrm regulations. Other regulations such as state-mandated beneﬁts may also aﬀect business size, since larger ﬁrms can self-insure and avoid state regulation. Chapter Three provides evidence that state HI mandates have led small ﬁrms to adjust their size to escape the more highly regulated market for HI. Section Conclusion

Businesses face a vast array of HI regulations either directly or indirectly via their contracts with health insurers. These regulations might be expected to aﬀect HI choices, workforce composition, turnover, and size. Even though there is some research that examines the diﬀerential eﬀect of these regulations on small and large ﬁrms, there is substantial scope to expand this research agenda. An important policy issue is whether policymakers should consider pricing regulation to accom-

In the Name of Entrepreneurship?

pany HI access regulations. Premium regulations are likely to lead to an increase in the number of small ﬁrms oﬀering HI. However, the magnitude of the increase depends on small ﬁrms’ price sensitivity. In general, small ﬁrms have not been very responsive to price incentives. Excessively stringent premium regulations may also drive insurers away from unproﬁtable markets, resulting in worse HI availability. Furthermore, premium regulations may change the landscape of labor-market distortions. Policymakers will need to carefully balance these considerations in revising existing HI regulation.

Chapter Conclusion In this chapter, we have considered the impact of regulation and litigation on small businesses and entrepreneurship in the context of four key regulatory areas: corporate securities, environmental protection, employment, and HI. Across these four areas, we ﬁnd that regulation tends to originate from concerns about the behavior of large businesses and that the strategies for addressing these concerns tend to focus on actions that would most eﬀectively inﬂuence the behavior of larger businesses. This is true even for HI regulations, which, although intended to beneﬁt small businesses and to improve their ability to oﬀer aﬀordable HI to their employees, are directed at the HI companies—which are large businesses. However, the attention given to large businesses does not mean only that regulators are unconcerned about the potential impact of small ﬁrms’ behavior on employees, customers, or society as a whole, but perhaps more that the policy debate is dominated by the magnitude of large-business interests. There is broad recognition that regulation impacts small businesses diﬀerently from the way in which it aﬀects large businesses. For example, compliance with environmental regulations can be costlier for small ﬁrms than for large ones because of the ﬁnancial and other costs of compliance. To cite another example from this chapter, there are also reasons to believe that small-business owners will face higher WC costs than will large-business owners. To mitigate the potential

The Impact of Regulation and Litigation

negative impacts of regulation, small businesses often receive special consideration in the policymaking process. Such special consideration takes many forms, including opportunities to voice speciﬁc concerns or raise issues about proposed regulations before they go into eﬀect or alterations in the regulatory environment that small ﬁrms face (e.g., exemptions from regulation, modiﬁed compliance procedures, reduced penalties for violation of regulation, special programs to assist small ﬁrms in complying with regulation). The existence of a diﬀerent regulatory environment for small ﬁrms is referred to as regulatory tiering and is a common approach to addressing the concerns of small businesses. As we have described in this chapter and further detail in Appendix A, the size thresholds that are used to determine whether a ﬁrm is eligible for a diﬀerent regulatory environment vary dramatically across regulatory contexts. Although these thresholds create clear incentives for ﬁrms to limit growth to avoid them, in our review of the literature, we uncovered no research on whether regulatory thresholds aﬀect ﬁrm growth in the United States or on the magnitude of that eﬀect. Across the board, there is little empirical evidence to demonstrate that the special consideration oﬀered to small businesses in the regulatory context makes sense from a cost-beneﬁt perspective. Further, it is not always clear that regulations designed to beneﬁt small businesses always achieve their intended aims, that programs designed to assist small businesses in complying with regulations are well targeted or well utilized, or that thresholds that deﬁne exemptions from regulations are based on a careful consideration of the relative costs and beneﬁts of regulation. For example, although much of the regulatory regime surrounding HI was designed explicitly to increase access to HI among small businesses, it appears that these regulations have had little eﬀect on premiums or HI coverage. In the environmental realm, consensus is lacking over whether environmental regulation put small ﬁrms at a disadvantage relative to larger ones, and it has been diﬃcult to judge the success of recent eﬀorts to make it easier for small ﬁrms to comply. In the area of securities and exchange law, data have been lacking to address key questions, such as whether new regulatory requirements such as SOX have aﬀected the willingness of small ﬁrms to enter public capital

In the Name of Entrepreneurship?

markets. Finally, the discussion of employment law shows that, despite the existence of size thresholds, many policy instruments designed to regulate aspects of the employer-employee relationship might have had unintended consequences that impose higher costs on small businesses than on larger ones. The remainder of this book attempts to expand our understanding of the inﬂuence of public policy on small businesses. Although each of the studies is focused on a speciﬁc policy area, many of the lessons learned are potentially generalizable. Regulations or programs designed to beneﬁt small businesses are rarely criticized or questioned. However, as has been described in this chapter, even programs that are well targeted may fail to achieve their intended aims and also have unintended consequences. Chapter Three examines the issue of the unintended eﬀects of policies in the context of regulation of HI. This chapter also explores eﬀect of regulatory thresholds on business growth. Research on small businesses rarely makes a distinction between small establishments (workplaces) and small ﬁrms (business operations). Indeed, data on establishment size is often used to draw conclusions about diﬀerences by ﬁrm size. The importance of this distinction and its implications for understanding the eﬀects of policy are illustrated in the study on workplace fatalities and ﬁrm size described in Chapter Four. This study illustrates how important it is for the research unit of analysis to be congruent with the policymaking unit of analysis. A major challenge facing research on the eﬀect of regulation on small businesses is isolating the eﬀect of the regulation’s introduction or reform from other factors. This issue is a key element of the assessment, provided in Chapter Five, of the impact of SOX on small businesses. The choices that entrepreneurs make regarding incorporation, organizational status, and publicly held status can have important implications for ﬁrm behavior, growth, and success. Chapter Six considers whether organizational forms that were designed, at least in part, with small businesses in mind have actually beneﬁted small law ﬁrms.

CHAPTER THREE

State Health-Insurance Mandates, ConsumerDirected Health Plans, and Health Savings Accounts: Are They a Panacea for Small Businesses? Susan M. Gates, Kanika Kapur, and Pinar Karaca-Mandic

Small ﬁrms in the United States that seek to oﬀer HI to their employees have historically reported problems with the availability and aﬀordability of their options. The cost of HI has been the primary concern of small-business owners for several decades. In 2004, two-thirds of small-business owners listed health-care costs as a critical problem—a proportion that increased by 18 percentage points between 2000 and 2004 (NFIB, 2004). Small businesses are likelier to report problems with their health-care availability and costs than larger businesses (Charles Brown, Hamilton, and Medoﬀ, 1990; McLaughlin, 1993; Fronstin and Helman, 2000). Extending HI to workers and the families of workers in small ﬁrms continues to be a pressing issue. HI plans oﬀered to small businesses tend to suﬀer from limitations that are widely acknowledged. First, small-group HI premiums have varied dramatically depending on the expected cost of the group (Cutler, 1994). In addition, the HI policies oﬀered to small ﬁrms often contain preexisting-condition clauses that exclude expensive conditions from coverage (U.S. Congress, 1988). Some insurers simply do not oﬀer policies to small ﬁrms, resulting in limited choices for small ﬁrms. These limitations, along with double-digit increases in HI costs and consumer dissatisfaction with managed care, have led to both employers and government policymakers seeking new ways to contain health-care costs.

In the Name of Entrepreneurship?

Policymakers have pursued various avenues to address the problems that small businesses face in the market for HI. In this book, we provide a summary of the success of two diﬀerent approaches: one that is regulatory in nature and the other that is market based. The ﬁrst is state HI mandates. To try to address problems with the small-group market, most states passed small-group HI reforms in the 1990s. These reforms have three key characteristics. First, they restrict insurers’ ability to deny coverage to small ﬁrms. Second, they restrict premium variability, and, ﬁnally, they encourage portability when employees move from job to job. In this chapter, we summarize the evidence of the inﬂuence these mandates had not only on HI premiums and HI availability, but also on business size. An alternative solution to the HI crisis that has been advocated by the Bush administration and by some policy analysts is the development of consumer-directed health plans (CDHPs). These plans aim to control costs by increasing consumers’ ﬁnancial responsibility and involvement in their health-care choices. Since CDHPs are potentially less costly than traditional health plans and may appeal to younger workers with low health-care demands, these plans may be well suited to workers in small businesses (Laing, undated). In this chapter, we examine the eﬀect to date of two types of policy initiatives that could have substantial beneﬁts for small businesses: state HI mandates and key components of CDHPs—HSAs, health reimbursem*nt arrangements (HRAs), and HDHPs. We summarize the key policy issues, review existing research evidence, including our own research, on the eﬀect of these initiatives on small businesses, and oﬀer some conclusions for policymakers.

Small Businesses Typically Face Restricted HealthInsurance Options The diﬃculties that small ﬁrms face in obtaining and maintaining HI for their employees have been widely documented (Charles Brown, Hamilton, and Medoﬀ, 1990; McLaughlin, 1993; Fronstin and Helman, 2000). Only 43 percent of ﬁrms with fewer than 50 employ-

State HI Mandates, CDHPs, and HSAs

ees oﬀer HI, compared to 95 percent of ﬁrms with 50 or more employees (AHRQ, 2003). This low proportion has been attributed, in part, to the high administrative cost of HI for small ﬁrms, the low demand for HI among workers in these ﬁrms, and the unwillingness of insurers to take on small-ﬁrm risks (McLaughlin, 1993; Fronstin and Helman, 2000; Monheit and Vistnes, 1999). According to surveys conducted by the National Federation of Independent Business (NFIB, 2004), the cost of providing HI has been the number-one concern of small-business owners since 1986. In 2004, nearly two-thirds of small-business owners cited it as a critical issue. While the cost of HI is a concern for all employers irrespective of size, it is well documented that the administrative cost of HI is substantially higher for small employers—20 to 25 percent of employee premiums in small ﬁrms compared to 10 percent of premiums in large ﬁrms—and is one possible reason that so few small businesses oﬀer HI to their employees (U.S. General Accounting Oﬃce, 2001).1 Several studies have shown that small-ﬁrm employees who do not have HI are relatively young and healthy and are likelier to have higher job turnover and therefore have a lower demand for employment-based HI (Monheit and Vistnes, 1994, 1999, 2006). Even though the demographic characteristics of small-ﬁrm employees as a whole (insured and uninsured combined) appear to be quite similar to those of other employees, small ﬁrms employ a slightly larger share of workers under age 25 and a much larger share of workers over age 65 (Headd, 2000). This suggests that small ﬁrms are likelier to employ individuals with a relatively low demand for employer-sponsored HI: the youngest and healthiest workers but also the oldest workers who are eligible for HI coverage under Medicare.

The lower administrative costs in large ﬁrms may be due to the fact that large ﬁrms tend to have a beneﬁts manager to coordinate health claims and complete paperwork. The beneﬁts oﬃce in large ﬁrms acts as an intermediary between employees and insurers, reducing administrative burden for large-ﬁrm insurers. Large ﬁrms are also less likely to drop insurance, resulting in lower transition costs for insurance companies.

In the Name of Entrepreneurship?

State Health-Insurance Mandates Seek to Expand SmallBusiness Options To address the aforementioned problems with the small-group market for HI, virtually all states passed some form of small-group HI reform in the 1990s. Although the extent of and approach to the reforms vary from state to state, they contain broadly similar elements. Rating Reforms

State reforms have placed restrictions on the factors that can be used to set HI premiums or limited the rate variations to speciﬁed ranges. Most states’ premium rating reforms follow the rate-banding approach, which limits insurers to a set number of classes for which they can charge separate rates. Age, geographic location, family size, and group size are often allowable factors that can be used to set classes. The reform restricts the variation in premiums that the insurer can charge to ﬁrms within each of these classes and restricts the variation allowed between business classes. Most states allow nine business classes and about 15 to 30 percent premium variation within and between classes, although these numbers vary somewhat from state to state. Rating reforms do not regulate the dollar value of the premium; however, they do often restrict the percentage increase in premiums from year to year. About 10 states have implemented modiﬁed community rating, in which the use of claims experience and employee health status in setting premiums has been restricted, and premiums can be set only on the basis of demographic factors such as family size and age. Community rating, the strongest form of rating reforms, has been implemented only by a few states and disallows variation in premiums due to demographic and health factors. It is plausible that these restrictions on premiums may have limited premium variability for a small ﬁrm. In addition, these reforms may have succeeded in reducing premiums for small ﬁrms that employ individuals with high health costs. The rate-banding approach is the most common premium rating reform, and this form of reform often allows claims experience to be used to set premiums. Therefore, in practice, in most states, premiums still do vary substantially due to claims

State HI Mandates, CDHPs, and HSAs

experience and the health characteristics of the insured (U.S. General Accounting Oﬃce, 1995; Hall, 2000). Guaranteed-Issue and Guaranteed-Renewal Reforms

Every state (except Georgia) that has passed small-group HI reform has included guaranteed-renewal reform in its package. This reform requires insurers to renew coverage for all groups, except in cases of nonpayment of premium or fraud. Guaranteed-issue legislation, on the other hand, is excluded from the reform packages of eight states that have passed guaranteed-renewal laws. Guaranteed-issue legislation requires HI companies to oﬀer HI coverage to any small employer in the state. Some guaranteed-issue legislation requires HI companies to oﬀer only one or two speciﬁc beneﬁt plans, while some requires insurers to oﬀer every small-group health plan they sell to each small employer. Guaranteed-issue legislation limits the ability of insurers to circumvent rating reform by insuring only low-cost, small ﬁrms. Preexisting-Condition Limitation and Portability Reforms

Health plans often impose waiting periods for coverage. These waiting periods may pertain to all coverage or only to coverage for preexisting health conditions. In some instances, health plans permanently exclude coverage for speciﬁc health conditions. State preexistingcondition reforms limit the length of time for which preexisting health conditions can be excluded from coverage. Most states limit the waiting period for coverage for preexisting conditions to a maximum of 12 months and allow only conditions present in the past six months to be deﬁned as preexisting. Portability reforms ensure that an individual who is covered by HI on a previous job does not face any new preexisting-condition exclusions or waiting periods as a result of changing jobs. Note that portability reforms do not place any restrictions on either premiums that HI companies charge small ﬁrms or premium contributions that ﬁrms charge workers. Portability and preexisting-condition limitation laws have been enacted at the same time in most states. HIPAA reinforces preexisting-condition exclusion limitations. In essence, these laws virtually remove small-group insurers’ ability to

In the Name of Entrepreneurship?

exclude coverage for certain conditions or to deny individuals coverage in small-ﬁrm policies. Therefore, after the passage of these laws, charging higher premiums, subject to the state’s premium rating reforms, may be small-group insurers’ only available underwriting option. Small-group HI reforms regulate the type of HI that HI companies can sell to small ﬁrms. They have no direct eﬀect on the HI oﬀered to other ﬁrms, although they may have an indirect eﬀect if insurers adjust policies in the large-group market to make it easier to comply with the regulations in the small-group market.

State Mandates Have Not Improved Small-Business Access to Health Insurance Research examining the eﬀect of state HI mandates on HI coverage, ﬁrms’ propensity to oﬀer coverage, and HI premiums generally has shown a small eﬀect or no eﬀect on small ﬁrms’ propensity to oﬀer HI or on employees’ HI coverage (Sloan and Conover, 1998; Gail Jensen and Morrisey, 1999; Zuckerman and Rajan, 1999; Monheit and Schone, 2004; Buchmueller and DiNardo, 2002; Marquis and Long, 2001). A few studies do ﬁnd modest eﬀects of the reforms on HI oﬀer rates and HI coverage; however, the direction of the eﬀects found varies among the studies (Uccello, 1996; Hing and Jensen, 1999; Simon, 2005; Buchmueller and Jensen, 1997). In addition, some work has demonstrated that stronger reforms increased HI coverage for high-risk workers relative to low-risk workers (Monheit and Schone, 2004; Davidoﬀ, Blumberg, and Nichols, 2005). Most of these studies exploit cross-sectional and time-series variations in the implementation of state reforms to identify the reforms’ eﬀects on HI coverage and do not focus on analyzing employment and employment ﬂows in small and large ﬁrms as a result of the reforms. The overall eﬀect of reforms is likely to depend on the characteristics of those reforms. The Health Insurance Association of America estimates that guaranteed-issue provisions have only a small impact on premiums—2 to 4 percent (Roger Thompson, 1992). Gail Jensen, Michael Morrisey, and R. J. Morlock (1995) found no evidence that

State HI Mandates, CDHPs, and HSAs

guaranteed-issue laws, preexisting-condition limits, or laws limiting exclusions on the basis of condition or occupation resulted in premium increases. Premiums in New York, which enacted very stringent rating reforms in the small-group market, rose about 5 percent during the ﬁrst year that community rating was in eﬀect (Chollet, 1994). Minnesota, which adopted restrictions on premium rate variations, also experienced premium rate increases of less than 5 percent in the year after it enacted these rating reforms in combination with a number of other small group reforms (Blumberg and Nichols, 1996). Two studies examined the labor-market eﬀects of small-group HI reform and ﬁnd small or no eﬀect on mobility among workers with high expected health costs and no eﬀect on wages or hours worked (Kapur, 2003; Kapur et al., 2005; Kaestner and Simon, 2002). Because of the way in which these HI mandates were implemented—applying only the HI products oﬀered to ﬁrms below a certain size threshold—we were also curious as to whether the mandates had any unintended eﬀect on the size of ﬁrms. While there is no prior research on the eﬀect of small-group HI reforms on the size of small ﬁrms, a few studies have examined the eﬀect of other regulations on business size. Schivardi and Torrini (2004) examine the eﬀect of employment-protection legislation on business size in Italy. Employment-protection legislation, which imposes higher unfairdismissal costs on ﬁrms that employ more than 15 employees, was found to reduce business size and growth for ﬁrms that were just below the size threshold. Using the same data source, Garibaldi, Pacelli, and Borgarello (2004) ﬁnd results that are consistent with Schivardi and Torrini’s. Germany’s Protection Against Dismissal Act allows ﬁrms above a certain size threshold to sue for wrongful termination. The threshold size has varied over time. Verick (2004) examined the eﬀect of this size threshold on ﬁrm size and found mixed eﬀects. We undertook a study to examine whether there was a size eﬀect (Kapur et al., 2006). We summarize our ﬁndings in the next section of this chapter.

In the Name of Entrepreneurship?

State Health-Insurance Mandates Have Had Unintended Effects In our study, we used data from a nationally representative, employerbased survey conducted by KMPG Consulting (now BearingPoint). This data source contains information on HI oﬀering, number of workers employed in the ﬁrm, and the industry to which the ﬁrm belongs.2 Because most states adopted small-business HI reforms during the early 1990s, we used the surveys from 1993, 1996, and 1998. These were the only years during the 1990s in which the survey included smaller ﬁrms with fewer than 200 employees. We also used a data set that characterizes the presence of a smallbusiness HI reform for any given state and year, as well as the detailed characteristics of the reform, if one exists. These data come from the state small-group reform survey conducted by Simon (2005) and Marquis and Long (2001). Our analysis used the upper and lower limits of the ﬁrm-size thresholds for the reform to be applicable. The HI reform data and the ﬁrm-level survey data were merged using the survey year and the state of the ﬁrm. Small-group HI reform was coded using a binary indicator of having a reform or not. Table 3.1 provides a data summary of the state health reforms. As the table indicates, in 1993, 14 states had no reform; by 1997, all states except for one (Michigan) had adopted some type of small-business health-care reform. Most states have a moderate reform that includes restrictions on premiums using a rate-band approach rather than by imposing community rating or modiﬁed community rating. Table 3.2 presents the upper size limit for small-group HI reforms. During these years, most states with reforms had either 25 or 50 employees as the upper size threshold. In our data, 81 percent of state-year observations had thresholds at either 25 or 50 employees. Over time, states tended to raise their thresholds, and the number of states with 2

We used data on the number of workers employed in the entire ﬁrm rather than in a single location because HI decisions tend to be made at the ﬁrm level rather than the plant level. However, as a sensitivity check, we reestimated our models for the sample of single-location ﬁrms. We found results that were qualitatively similar but far less precise, primarily because we lost about half the sample while conducting this check.

State HI Mandates, CDHPs, and HSAs

Table 3.1 State Counts by Reform Level and Year Reform Level

1992

1993

1995

1996

1997

1998

No reform

Reform

NOTE: Does not include Hawaii but does include Washington, D.C.

Table 3.2 State Counts by Firm-Size Upper Limit for State Small-Group HealthInsurance Reform, by Year Firm-Size Upper Limit

1992

1993

1995

1996

1997

1998

26–50

51–100

No reform

NOTE: Does not include Hawaii but does include Washington, D.C.

upper size thresholds of 25 employees decreases. By 1997, no state had 25 employees as the upper size threshold.3 As mentioned, the deﬁnition of small ﬁrm varies among states and, in some cases, over time within the same state. The upper size threshold for a small ﬁrm varies between 25 and 100 employees, depending on the state and year. The lower threshold varies between one and ﬁve employees. Small-group HI reform may aﬀect the scope, price, and availability of HI for small ﬁrms. For the sake of exposition, let us assume that there are two types of small ﬁrms: low-cost ﬁrms, which employ 3

The lower size limit for the reforms was one, two, or three employees, depending on the state and year. However, California, in 1993, had a lower threshold of ﬁve employees. Our data set includes only ﬁrms that had three or more employees. We have reestimated our models, excluding Californian ﬁrms with fewer than ﬁve employees in 1993 (N = 8), and ﬁnd virtually identical results.

In the Name of Entrepreneurship?

a high proportion of young and healthy workers, and high-cost ﬁrms, which employ workers with high expected health-care costs (either older workers or workers at risk for injury or illness). Small-group HI reforms prevent insurers from excluding preexisting conditions from HI coverage, implying more complete HI for all small ﬁrms. However, in states that impose tight premium rating restrictions and guaranteed issue, the combination of the two types of reforms may drive insurers to set premiums in a way that increases premiums for low-cost ﬁrms and reduces premiums for high-cost ﬁrms. Alternatively, the regulations might aﬀect the completeness of the plans oﬀered if insurers ﬁnd it impossible to oﬀer comprehensive plans to all small ﬁrms at a reasonable price. In states with weak premium rating restrictions, premiums may be aﬀected relatively little. Guaranteed-issue and -renewal laws directly aﬀect the availability of HI. In particular, in states with guaranteed issue, high-cost ﬁrms that may have had problems obtaining access to HI should ﬁnd obtaining a policy much easier. However, the overall burden of complying with the state small-group HI regulations may be a disincentive for oﬀering HI in the small-group market for some insurers, and insurers may consider exiting the market in highly regulated states or consider reducing their marketing eﬀorts in those states. As a result, the reforms may have an adverse eﬀect on availability for low-cost ﬁrms. Therefore, the reforms may have heterogeneous eﬀects on price and availability, depending on the strength of their component provisions and the market composition of low-cost versus high-cost small ﬁrms. In our empirical estimation (see Kapur et al., 2006), we focus on ﬁrms that oﬀer HI right around the legislative threshold—since the reform is likeliest to aﬀect their decisions. We estimate whether reform states are likelier than nonreform states are to have a higher or lower proportion of ﬁrms oﬀering HI below the threshold. If ﬁrms value the reforms, the proportion of ﬁrms below the threshold to those above it should be higher in reform states as ﬁrms attempt to manipulate their size to remain below the reform threshold. If ﬁrms do not value the reforms, they will do just the opposite: They will expand so that they are no longer subject to the reforms, and then the proportion of ﬁrms

State HI Mandates, CDHPs, and HSAs

under the threshold relative to ﬁrms over the threshold will be lower in reform states.4 Our analysis provides evidence that, in states that implemented these reforms, ﬁrms oﬀering HI are signiﬁcantly likelier to be just above the threshold than just below the threshold. In states that implement a 25-employee threshold, we estimate that 31 percent of ﬁrms with 20 to 30 employees would fall below the 25-employee threshold, compared with 75 percent in states that did not have a reform. In states that implement a 50-employee threshold, we estimate that 65 percent of ﬁrms with 45 to 55 employees would fall below the threshold, compared with 82 percent of ﬁrms with 45 to 55 employees in states that did not have a reform. The magnitudes of these predicted changes in ﬁrm-size distribution are large; however, they apply to a relatively small segment of the ﬁrm distribution that is clustered around the regulatory threshold. These ﬁndings suggest that small employers near the threshold that oﬀered HI found the state HI mandates to be onerous and increased their size to avoid the regulated market. As expected, our analyses suggest that ﬁrms’ ability to make such an adjustment was greater for ﬁrms that were closest to the regulatory threshold. The magnitude and statistical signiﬁcance of the eﬀect declined as we expanded the size of the band around the threshold under consideration. Our study shows that the small-group HI reform implemented by states in the mid-1990s likely had unintended consequences. The reforms appear to have led ﬁrms to distort their ﬁrm-size decisions to avoid the more regulated market. What happened to the HI market in reform states to lead to these outcomes? There is evidence from previous research to suggest that the implementation of reforms increased 4

To capture proximity of ﬁrm size to the reform threshold, we restrict our analysis to states that implemented a reform with an upper size threshold of either 25 or 50 and use separate models to examine the eﬀect of each threshold. Since the inherent distribution of ﬁrms around the 25-employee size threshold diﬀers from the distribution of ﬁrms around the 50employee size threshold, we cannot estimate a model that compares distributional changes across diﬀerent thresholds. Our empirical strategy is to focus on a narrow set of ﬁrms around the threshold and study whether the proportion of ﬁrms under the threshold diﬀers across reform and nonreform states.

In the Name of Entrepreneurship?

the breadth of HI policies but also led to an increase in premiums as insurers that found the small-group regulations burdensome exited the market. For example, in New York, premiums were estimated to have risen for about 30 percent of the insured, and 500,000 New Yorkers were estimated to have cancelled their individual or smallgroup policies after the implementation of reforms (NCPA, 1994). In Oregon, insurers were reported to have exited the small-group market in response to the reforms (Brock, 1998). However, Buchmueller and DiNardo (2002) compared the New York market that had community rating (strong reform) to the markets in Pennsylvania and Connecticut (states that did not have strong reform) and found no evidence that HI oﬀering levels had fallen in New York. These reports suggest that the reforms may have resulted in changes in the small-group market that some, but not other, small ﬁrms valued. High-cost ﬁrms (that is, ﬁrms that employ workers with high expected health-care costs, as deﬁned in the conceptual framework section) that previously could not obtain HI are able to access coverage after the reform. Some of these ﬁrms may value the access to HI and the broad coverage oﬀered under reform, even if it means higher premiums. Low-cost ﬁrms (that is, ﬁrms that employ workers with low expected health-care costs), on the other hand, may place little value on the breadth of coverage oﬀered under reform. For example, many small ﬁrms hire a younger, healthier workforce and have higher worker turnover than larger ﬁrms have (Kapur, 2004), and these workers may not value the more complete policies and higher premiums associated with small-group HI reforms. If it is at all feasible, these ﬁrms may increase their ﬁrm size to avoid the reform and purchase HI in the unregulated market.

Consumer-Directed Health Plans Could Expand Options for Small Businesses Our review of the eﬀects of state HI mandates suggests that this regulatory approach was not terribly successful in terms of expanding access to HI for small businesses. Recently, policymakers have advocated an

State HI Mandates, CDHPs, and HSAs

alternative approach to achieving this aim. Since high and increasing costs of health services and HI are perceived as the primary barrier to access, new innovations in the HI market, CHDPs, are designed to encourage individual responsibility in health-care choices in the hope of increasing price sensitivity, controlling cost escalation, and ultimately improving access. This approach would yield beneﬁts for all businesses, but particularly for small businesses that are often shut out of the traditional HI market because of high costs of coverage. The basic logic behind this argument is that CDHPs change individual incentives by making consumers ﬁnancially responsible when they choose costly health-care options (Robinson, 2003). Ultimately, this change in individual incentives should reduce the cost of HI and possibly the cost of health care as well. Increases in consumer costsharing, especially deductibles, are part of this new strategy (Gabel et al., 2002). Despite the popular notion that encouraging the provision of CDHPs could improve the health-care market, economic theory can also support the opposite conclusion. In a market in which there is a trade-oﬀ between making consumers ﬁnancially responsible for their health care and providing consumers with complete HI, reducing HI to increase ﬁnancial responsibility can lead to a suboptimal outcome (Zeckhauser, 1970). This possible consequence of CDHPs has received little attention in a policy debate that is focused primarily on the potential role of CDHPs in reducing overall medical expenditures. HDHPs are an important feature of CDHPs. Often, these HDHPs are combined with a personal health-care spending account that provides individuals with favored tax treatment for money spent to pay for deductibles and copayments. Federal legislation has facilitated the formation of HRAs and HSAs. HRAs and HSAs potentially make HDHPs more palatable to individuals by providing them with a means to avoid taxes on money used to pay health expenses not covered by HDHPs. The accounts can compensate individuals somewhat for the risk associated with HDHPs. The employee can use money in these accounts to pay for unreimbursed, qualiﬁed medical expenditures. Unused funds in the account

In the Name of Entrepreneurship?

may be carried over from year to year.5 This carryover provision of HRAs and HSAs is intended to beneﬁt employees who use fewer and less costly services and encourage them to do so. The legal foundation for HSAs was established in 2003 under the Medicare Prescription Drug, Improvement, and Modernization Act (P.L. 108-173), creating the newest form of personal savings accounts. HSAs are available to all individuals and employer groups. To operate an HSA, employers or enrollees make deposits into a specially designated account that is then used to purchase health services. If an enrollee spends all of the funds allocated to his or her account in a given year and if this amount is less than the plan deductible, the enrollee must then pay for additional health services out of pocket until the plan deductible is met. (The expenditure amount between the annual account contribution and the deductible is often referred to as a doughnut hole). Above the deductible, enrollees’ health plans cover most costs. The earlier generation of personal health accounts, ﬂexible spending accounts (FSAs), did not permit enrollees to roll funds over from year to year. An HSA must be combined with an HI plan with a deductible of at least $1,100 for an individual and $2,200 for a family. The maximum account contribution is the lesser of 100 percent of the deductible or $2,850 for an individual or $5,650 for a family.6 While contributions can be made by the employee, the employer, or by both parties, the employee owns the account, and, thus, the account is fully portable across jobs. Unused funds are rolled over from year to year. Moreover, accounts can earn investment income that is not taxed as earned. In addition, funds in HSAs can be withdrawn to pay for nonmedical expenses, though they are then subject to taxes and to a penalty if the accountholder is under age 65. HRAs, available since 2002, diﬀer from HSAs in several important respects: They need not be paired with HDHPs with federally mandated characteristics; only the employer contributions to the 5

In the case of HRAs, the employer chooses whether the accounts have this carry-over provision.

The amounts reﬂect the requirements for 2007 and are indexed to inﬂation.

State HI Mandates, CDHPs, and HSAs

account receive favorable tax treatment; portability across employers and annual carry-over is permitted but not required; accounts are funded by employers only; and third-party administration of the accounts is required. Some observers have argued that HSAs are particularly well situated to help small ﬁrms without medical plans to oﬀer some form of HI to their employees (Laing, undated). HDHPs typically have lower premiums and are more accessible than other types of health-care oﬀerings are to small businesses. The Small Business and Entrepreneurship Council supported the implementation of HDHPs, and HSAs in particular, as a way to provide small-business owners and their employees greater access to aﬀordable choices in HI. Early evidence suggests that CDHPs are associated with both lower costs and lower cost increases (Buntin et al., 2006). However, CDHPs continue to be controversial as a mechanism for controlling costs and shifting responsibility to consumers (Ginsburg, 2006; Lee and Hoo, 2006). Among other things, there is some evidence that healthier individuals are likelier than less healthy people to opt for these plans (Buntin et al., 2006). Early evidence also raises questions as to whether CDHPs really are a panacea for small businesses.

Small Firms Have Not Been Especially Quick to Adopt Consumer-Directed Health Plans Insurers’ interest in HRAs and HSAs is widespread and growing rapidly. According to a recent survey by America’s Health Insurance Plans (AHIP), the number of individuals covered by an HSA or other HDHP reached 3.2 million in January 2006—having tripled in less than one year (AHIP, 2006). Approximately 30 percent of HSA purchasers did not previously have HI, according to the AHIP survey, with 16 percent of new small-business HI-plan purchasers previously not oﬀering HI (AHIP, 2006). This growth in coverage was due to increases in both the group and the individual market. Today, at least 75 insurers oﬀer accountcompatible plans nationwide (KFF, 2004; AHIP, 2005). Fifty-eight

In the Name of Entrepreneurship?

oﬀer high-deductible, account-compatible plans to large employers, 56 to small employers, and 47 to individuals. Most large insurers will also have full integration of HSAs and high-deductible plans by 2006, meaning that the carrier has an established relationship with a bank and can provide information about the account along with information about total claims (CDMR, 2005). There is some evidence that, while HSA products were more popular among small businesses and individuals than among larger groups initially, their use is growing most rapidly among large employers. Large employers are generally introducing these products in a gradual way. Few large employers have chosen the full-replacement route of abandoning traditional plans in favor of CDHPs (Schieber, 2004). HIindustry oﬃcials report that employee take-up is low when CDHPs are oﬀered alongside traditional plans. Insurers and employers also report that employers’ success in enrolling employees in these new plans to date depends on comprehensive education and communication eﬀorts rather than waiting for employees to respond to premium diﬀerences. An AHIP survey of member companies found that only 3 percent of HSA enrollees in 2004 were in large-group plans (see AHIP, 2005). However, by January 2006, that ﬁgure had grown to 33 percent. Small-group plans represented 18 percent of enrollees in 2004 and 25 percent in January 2006 (AHIP, 2006). Some smaller businesses that might not otherwise oﬀer HI see HSAs as a way to provide low-cost coverage. According to the 2006 survey of AHIP member companies, 33 percent of small-group HSA policies were sold to businesses that previously did not oﬀer HI. This suggests that HSAs have the potential at least to serve as a meaningful tool for expanding health-care coverage to small-business employees, a ﬁnding that a simulation study conducted by Goldman, Buchanan, and Keeler (2000) supports. Goldman, Buchanan, and Keeler found that similar plans could increase the proportion of small businesses oﬀering HI. The 2006 AHIP survey of insurance companies provides information on the characteristics of the HSAs and other HDHPs provided to individuals, small groups, and large groups. This information suggests that small businesses are on a more level playing ﬁeld with large

State HI Mandates, CDHPs, and HSAs

businesses in this market. A comparison of the HSA and HDHP policies oﬀered in the small- versus large-group markets7 reveals that average annual deductibles are somewhat higher in the small-group market but that other characteristics are remarkably similar (AHIP, 2006). In the small-group market, the average annual deductible is $2,143 for individual coverage and $4,311 for family coverage, compared with $1,754 and $3,494, respectively, in the large-group market. The average annual premium is $2,772 for individual coverage and $6,955 for family coverage in the small-group market, compared with $2,745 and $6,715, respectively, in the large-group market. A survey of employers by the Henry J. Kaiser Family Foundation (KFF) and the Health Research and Educational Trust (HRET) (2006) provides information on the availability, enrollment, and characteristics of HDHPs that are either oﬀered with HRAs or are HSAcompatible (Claxton et al., 2005). This type of plan is referred to as an HDHP with a savings option (HDHP/SO). The data reﬂect the situation as of 2006. The survey ﬁnds that 7 percent of employers oﬀer one of these arrangements, with 1 percent oﬀering HRA HDHPs and 6 percent oﬀering HSA HDHPs. The fraction of employers oﬀering HSAs was up signiﬁcantly from the previous year. Large ﬁrms are likelier than small ﬁrms to oﬀer an HSAqualiﬁed HDHP. Twelve percent of ﬁrms with more than 1,000 employees oﬀered such a plan in 2006, up from 4 percent in 2005. Firms with 3 to 999 employees were half as likely to oﬀer an HSAqualiﬁed HDHP—6 percent of such employers oﬀered one. Conﬁrming the ﬁndings of other studies that suggest that individuals who have a choice among several plans are not likely to choose an HDHP/SO, the KFF survey reveals that 40 percent of workers covered by an HDHP/SO are in ﬁrms in which 100 percent of covered workers are enrolled in an HDHP/SO. In ﬁrms that provide other options in addition to HDHP/SOs, on average, 19 percent of those employees enroll in an HDHP/SO (KFF/HRET, 2006).

The AHIP survey deﬁnes the small-group market as one covering groups of 50 or fewer employees.

In the Name of Entrepreneurship?

Employer contributions to the savings accounts also vary tremendously. Thirty-seven percent of employers oﬀering HSA-qualiﬁed HDHP/SOs do not contribute to an HSA. Twenty-seven percent contribute $1,200 or more (KFF/HRET, 2006). This information is not broken down by ﬁrm size in the report.

Additional Evidence on the Use of Health Reimbursem*nt Arrangements, Health Savings Accounts, and HighDeductible Health Plans by Small Businesses In this section of the chapter, we expand on existing descriptive analyses of HSA and HDHP oﬀerings, focusing in particular on small ﬁrms. We compare the proﬁle of small-ﬁrm oﬀerings to that of largerﬁrm oﬀerings. We also perform a multivariate analysis of HSA and HDHP oﬀerings. Our goal in this analysis is primarily to describe the CDHP oﬀerings in small businesses and to assess whether the popularity of these plans varies by ﬁrm size. Our analysis does not test whether the advent of CDHPs has increased the propensity of small businesses to oﬀer HI.8 However, the descriptive proﬁle in this report provides a useful backdrop for understanding the role of CDHPs in smallbusiness HI. Following most of the literature, we use the term CDHP to refer to any HDHP; typically, high-deductible refers to a plan with a deductible of $1,000 or more. HDHPs may be coupled with HSAs or HRAs (Buntin et al., 2006). Our work on HSAs, HRAs, and CDHPs uses data from the 2003, 2004, and 2005 KFF/HRET annual employer health-beneﬁts surveys (KFF/HRET 2003, 2004, 2005). This is an annual, national telephone survey of about 5,000 randomly selected public and private employers. Firms range in size from small enterprises with a minimum

Identiﬁcation of the eﬀect of CDHP availability on HI oﬀering would require exogenous cross-sectional or time-series variation in the availability of CDHPs. Given that our data have only a limited time-series variation, we do not undertake to test the eﬀect of CDHP availability on HI oﬀering.

State HI Mandates, CDHPs, and HSAs

of three workers to corporations with more than 300,000 employees (see Claxton et al., 2005, for a detailed description of the survey). The data contain detailed information about the health beneﬁts oﬀered by the ﬁrm and about other ﬁrm characteristics. In particular, the survey asks about the types of health plans oﬀered (PPO, HMO, fee for service), enrollment in each type of plan, and whether the ﬁrm oﬀers an HSA or HDHP or both in conjunction. Moreover, the survey asks ﬁrms that do not oﬀer these plans about the likelihood of oﬀering HSA plans combined with an HDHP. The survey also asks additional details about the features of these plans such as the deductible, premiums, and plan enrollment. Additional information about whether the ﬁrm is considering CDHPs in the future and whether the ﬁrm is aware of these products is also available. The survey does not ask this full set of questions every year—for example, it requested information on oﬀering and characteristics of HRA plans in 2005 only. Other ﬁrm data include the composition of the workforce (such as percentage that is low wage), the unionization of workers, and the number of workers in the ﬁrm, industry, rural versus urban, employee turnover, whether the ﬁrm laid oﬀ any workers in the previous year, and percent of the workforce that is part time. There are also measures of the cost and quality of health-beneﬁt oﬀerings such as whether the ﬁrm oﬀers retiree beneﬁts, wait periods, and employer contribution to each plan oﬀered. A subsample of ﬁrms is interviewed for two consecutive years, allowing us to construct a two-year longitudinal sample as well as a cross-sectional sample. Our analyses using these data are weighted using ﬁrm-level weights.

Consumer-Directed Health Plan Utilization and Growth Do Not Vary by Firm Size As is well known, the smallest ﬁrms (3 to 49 employees) are less likely than larger ﬁrms to oﬀer HI, as shown in Table 3.3. About 58 percent of small ﬁrms oﬀer HI, compared to 60 percent of ﬁrms with 50 to 199

In the Name of Entrepreneurship?

Table 3.3 Descriptive Proﬁle of Health Insurance Offerors, by Firm Size and Year Characteristic

All Firms with 3–49 Firms with 3–199 Firms Employees Employees

2003–2005 Offering HI (%)

5,794

1,611

2,415

5,288

1,157

1,925

719

137

235

Offering HI (%)

HI offerors offering HDHPs (%)

Offering HI (%)

HI offerors offering HDHPs (%)

HDHP offerors offering HSAs (%)

HDHP offerors offering HRAs or HSAs (%)a

Sample size HI offerors offering HDHPs (%) Sample size HDHP offerors offering HSAs (%) Sample size 2003 Offering HI (%) HI offerors offering HDHPs (%) HDHP offerors offering HSAs (%) 2004

HDHP offerors offering HSAs (%) 2005

a Information on HRAs is available only in the 2005 data.

State HI Mandates, CDHPs, and HSAs

employees and 61 percent of all ﬁrms regardless of size.9 Despite the notion that CDHPs may be especially attractive to small ﬁrms, there is no evidence that oﬀering HDHPs, conditional on oﬀering HI, or oﬀering HSA plans conditional on oﬀering HDHPs is higher in small ﬁrms. Twelve percent of small and large ﬁrms that oﬀer HI also oﬀer HDHPs. Conditional on oﬀering HDHPs, 8 percent of small ﬁrms and 9 percent of all ﬁrms oﬀer HSAs; however, this diﬀerence is not statistically signiﬁcant. CDHPs have grown in popularity between 2003 and 2005. In 2003, only 5 percent of ﬁrms that oﬀered HI also oﬀered HDHPs, and 13 percent of ﬁrms that oﬀered HDHPs also oﬀered HSAs. By 2005, these percentages had grown to 20 percent oﬀering HDHPs and 20 percent oﬀering HRAs or HSAs, conditional on oﬀering HDHPs. However, there was no diﬀerence in the growth rate between small and large ﬁrms. Even though we observe little diﬀerence between small and large ﬁrms in HRA and HSA oﬀerings, simply examining the propensity to oﬀer these plans provides a partial picture. Firms may diﬀer in the generosity of their HSA and HRA plans—some may provide generous contributions and use these plans as a mechanism for subsidizing health-care expenditures, and others may have very high deductibles and large doughnut holes to shift costs to employees. In a later section, we examine beneﬁt-generosity variations in plans to develop a full picture of the diﬀerences in CDHP oﬀerings between small and large businesses.

Persistence in Consumer-Directed Health Plan Offerings Given that CDHPs are new products, we may expect a moderate degree of churning in the oﬀering of these plans. Firms may choose to oﬀer CDHPs in one year and drop them the following year if take-up

While these diﬀerences may not seem large, HI oﬀer rates do drop precipitously as ﬁrm size falls—only 48 percent of the smallest ﬁrms (three to nine employees) oﬀer HI, compared to 98 percent of the largest ﬁrms (200 or more employees) (KFF/HRET, 2005).

In the Name of Entrepreneurship?

was poor or if they proved to be onerous to administer. To examine this issue, we analyze ﬁrms that are surveyed both in 2004 and 2005 to develop a longitudinal descriptive proﬁle of CDHP oﬀering. We ﬁnd that 75 percent of ﬁrms that oﬀered HDHPs in 2004 continued to oﬀer them in 2005. Small ﬁrms (3 to 49 employees) appear to be slightly less likely to oﬀer HDHPs in 2005, conditional on having oﬀered them in 2004. Sixty-six percent of ﬁrms with 3 to 49 employees oﬀer HDHPs in 2005, conditional on having oﬀered them in 2004; however, this statistic is based on a very small sample. In addition, 99 percent of ﬁrms that oﬀered HSAs in 2004 oﬀered either HRA or HSA plans in 2005; however, again, this statistic is based on a very small sample (N = 21). Conversely, 25 percent of ﬁrms oﬀering HDHPs in 2005 also oﬀered them in 2004. Twenty-six percent of ﬁrms oﬀering HRAs or HSAs in 2005 also oﬀered HDHPs in 2004. Among ﬁrms that oﬀered HRAs or HSAs in 2005, 16 percent oﬀered HSAs in 2004; however, this number is based on a very small sample size. Smaller ﬁrms appear to be slightly likelier than larger ﬁrms to be new adopters of CDHPs—22 percent of small ﬁrms oﬀered HDHPs in 2004, conditional on oﬀering HDHPs in 2005, compared to 25 percent for ﬁrms of all sizes. In summary, it appears that there is some evidence of higher churning in small-ﬁrm CDHP oﬀerings—small ﬁrms are likelier to be new adopters and less likely to retain their CDHP oﬀerings from year to year.

Which Firms Are Likelier to Offer Consumer-Directed Health Plans? We estimate logit models of CDHP-oﬀering behavior to parse out the ﬁrm and worker characteristics that are associated with a ﬁrm’s propensity to oﬀer CDHPs. We estimate a three-equation model to develop a complete picture of the CDHP-oﬀering decision. First, we estimate a model of the propensity to oﬀer HI. Second, we estimate a model of the propensity to oﬀer HDHPs, conditional on oﬀering HI. Lastly, we estimate a model of HSA oﬀering, conditional on oﬀering HDHPs.

State HI Mandates, CDHPs, and HSAs

The explanatory variables used in the models are ﬁrmcomposition variables, industry indicators, ﬁrm location, and survey year. Firm-composition variables include ﬁrm size (3 to 49 workers, 50 to 199 workers, 200-plus workers), and variables to capture workforce composition and worker demand for HI such as the percentage of the workforce earning $20,000 or less, percentage of the workforce working part time, percentage of covered employees, and union coverage. We also include a full set of industry-indicator variables to measure variations in insurance practices, insurance availability, and industrylevel worker demand. The ﬁrm’s geographical location is measured by indicators for region and an indicator for location in an urban area. HI premiums and safety-net availability vary by region and population density; therefore, location variables are useful proxies to capture this variation. Year indicators are included in the model to capture annual trends in CDHP availability and demand as well as annual variations in survey administration. We use ﬁrm-level data from 2003 to 2005 and apply ﬁrm-level sample weights to the models. The estimates from the three logit models are reported in Table 3.4. The ﬁrst and second columns show that small ﬁrms are no diﬀerent from larger ﬁrms in their propensity to oﬀer HSAs and HDHPs. This result is in keeping with the descriptive data reported earlier. The third column shows that small ﬁrms are substantially less likely to oﬀer HI, consistent with the rest of the literature on small ﬁrms and HI. The results also show that ﬁrms with a higher proportion of lowincome workers are both less likely to oﬀer HSA plans and less likely to oﬀer HI. The model predicts that the percentage of ﬁrms oﬀering HSA plans increases from 0.3 percent to 9 percent as the fraction of the workforce that is low income falls from 1 to 0.2 (75th to 25th percentile of the income distribution). This result is consistent with media reports that HSA plans may have more appeal for educated, higherincome workers. Firms that have union workers are signiﬁcantly likelier to oﬀer HDHPs as well (20 percent for union ﬁrms compared to 11 percent for nonunion ﬁrms), suggesting that unions have been lobbying for the introduction of more choice in HI oﬀerings. CDHP oﬀering also varies by industry and location. Construction and health-care industries are likelier than manufacturing

In the Name of Entrepreneurship?

Table 3.4 Determinants of Consumer-Directed Health Plan Offerings: Estimates from Logit Model (2003–2005) HDHP Offerors Offering HSA Determinant

Estimate

Standard Error

HI Offerors Offering HDHP Estimate

Standard Error

Offering HI Estimate

Standard Error

Firm composition Size: 3–49 workers

–0.43

0.52

0.09

0.23

–2.14 a

0.52

Size: 50–199 workers

–0.25

0.55

0.21

0.20

–1.95a

0.62

Workforce earning $20,000 or less (%)

–4.73a

1.44

0.60

0.59

–1.89b

0.88

Workforce working part time (%)

–1.12

1.50

–1.03

0.81

–3.67a

1.03

0.13

1.15

–0.58

0.69

–0.41

0.74

0.77b

0.37

1.84 b

0.85

–1.17

1.31

0.30

0.78

0.60

0.78

Covered employees (%) Union Industry Mining Construction

1.75c

0.98

–0.33

0.48

–0.48

0.73

Transportation, utilities, communication

1.65

1.09

–0.45

0.48

–0.55

0.85

Wholesale

1.68

1.06

0.48

0.56

0.63

0.70

Retail

0.76

0.93

1.16b

0.51

–1.05

0.73

Financial

–0.98

0.98

0.25

0.47

0.38

0.74

Service

–0.48

0.82

1.16a

0.40

0.07

0.62

Government

0.56

0.91

–0.69

0.44

2.51a

0.69

Health care

1.65b

0.80

0.18

0.47

–0.27

0.78

State HI Mandates, CDHPs, and HSAs

Table 3.4—Continued HDHP Offerors Offering HSA Determinant

Estimate

Standard Error

HI Offerors Offering HDHP Estimate

Standard Error

Offering HI Estimate

Standard Error

Location Midwest

2.22b

0.93

0.96b

0.44

–0.10

0.50

South

1.99b

0.97

0.61

0.47

0.21

0.45

West

0.94

1.02

0.23

0.56

0.01

0.62

Urban

–1.18 c

0.64

0.13

0.31

0.20

0.42

2003

0.19

0.73

–1.64 a

0.39

0.26

0.40

2004

–1.46 c

0.82

–0.76b

0.36

0.55

0.40

Observations

719

Year

5,288

5,794

a Signiﬁcant at 1 percent. b Signiﬁcant at 5 percent. c Signiﬁcant at 10 percent.

(the omitted category) to oﬀer HSAs, and retail and service industries are likelier to oﬀer HDHPs than manufacturing industries. Furthermore, CDHPs (particularly HSAs) appear to be most popular in the Midwest, followed by the South, and less popular in urban areas than in rural areas. There also appears to be an increase in CDHP oﬀerings in 2005 from earlier years, suggesting that these plans are growing in popularity. The models presented in Table 3.4 focus on HDHP and HSA oﬀerings. Recently, HRAs have become an important part of the CDHP landscape. Given that HRAs are a recent development, our data set contains information on these plans for 2005 only. We have reestimated the logit models on 2005 data including a separate logit model that includes ﬁrms that oﬀer either an HSA or HRA, conditional on oﬀering HDHPs. Consistent with the results reported for HSA plans, we found no diﬀerences in oﬀering by ﬁrm size. We have

In the Name of Entrepreneurship?

not reported these results in Table 3.4; however, they are available on request.

Longitudinal Analysis of Consumer-Directed Health Plan Offerings Our data allow us to follow ﬁrms for a two-year period. We use our two-year analytic database for 2004–2005 to analyze the eﬀect of HI status in 2004 on CDHP oﬀering in 2005. This analysis provides us with a picture of the dynamics of plan determination and the importance of persistence in health-plan oﬀerings. We estimate two logit models for 2005. First, we model a ﬁrm’s propensity to oﬀer HDHPs, conditional on oﬀering HI. Next, we model a ﬁrm’s propensity to oﬀer an HRA or HSA, conditional on oﬀering an HDHP. The explanatory variables in these models are, for the most part, the same as those in the logit models presented earlier. We include a set of ﬁrm-composition variables (ﬁrm size and workforce characteristics), industry indicators, and location indicators. We also include a set of current HI oﬀering variables—these are whether the ﬁrm oﬀers only one plan, two to four plans, or ﬁve or more plans. We expect that ﬁrms that oﬀer many plans may be likelier to choose a CDHP as one of the options. We also include a set of lagged (2004) HI variables. These are whether the ﬁrm oﬀered an HDHP in 2004 and whether the ﬁrm oﬀers only one plan, two to four plans, or ﬁve or more plans in 2004. Our results in Table 3.5 show that, while there continues to be no diﬀerence among small and large ﬁrms in HDHP oﬀering, we do ﬁnd that HRA and HSA plan oﬀering is 26 percent lower in the smallest ﬁrms (3 to 49 employees) than in ﬁrms that employ 200 or more workers and that this result is statistically signiﬁcant. This result controls for all other variables that may be associated with HRA and HSA oﬀering, including lagged and current HI oﬀerings. We also ﬁnd that ﬁrms that oﬀered only one plan or two to four plans were signiﬁcantly less likely to oﬀer HRAs or HSAs than were ﬁrms that oﬀered ﬁve or more plans. In addition, ﬁrms that oﬀered two to four plans in 2004 were signiﬁcantly less likely to oﬀer HDHPs in 2005 than were ﬁrms that oﬀered

State HI Mandates, CDHPs, and HSAs

Table 3.5 Determinants of Consumer-Directed Health Plan Offerings in 2005: Estimates from a Logit Model HI Offerors Offering HDHP Determinant

Estimate

Standard Error

HDHP Offerors Offering HRA or HSA Estimate

Standard Error

Firm composition Size: 3–49 workers

0.12

0.41

–2.12a

0.78

Size: 50–199 workers

0.54

0.36

–0.89

0.76

–0.43

0.71

–7.53a

1.72

0.33

0.52

0.50

0.96

Offer only one plan

0.10

0.85

–1.98b

1.05

Offer 2–4 plans

0.38

0.86

–4.90a

1.18

3.20a

0.54

0.46

0.82

Workforce earning $20,000 or less (%) Union HI

Lagged HI (2004) Offered HDHP Offered only one plan

–0.93

0.93

–1.42

1.14

Offered 2–4 plans

–1.70 b

0.92

–1.09

1.13

0.41

1.24

1.42

1.38

1.34

1.08

Industry Mining Construction

–0.1

0.84

Transportation, utilities, communication

–0.96

0.72

Wholesale

2.28a

0.87

Retail

0.84

0.78

In the Name of Entrepreneurship?

Table 3.5—Continued HI Offerors Offering HDHP Determinant Financial Service Government

Estimate 0.71

Standard Error 0.77

HDHP Offerors Offering HRA or HSA Estimate –1.29

0.07

0.67

2.52b

–0.49

Standard Error 1.74 1.00

0.74

2.93

1.83

1.69b

0.87

4.38a

1.29

Midwest

1.80a

0.51

1.22

1.00

South

0.94 b

0.51

1.95c

0.94

West

1.32c

0.58

Health care Location

Observations

1,169

–1.27

1.05

268

NOTE: Sample consisted of all ﬁrms surveyed both in 2004 and 2005. a Signiﬁcant at 1 percent. b Signiﬁcant at 10 percent. c Signiﬁcant at 5 percent.

ﬁve or more plans. These results are consistent with the notion that ﬁrms that oﬀer more choice are likely to also oﬀer CDHP options.

Beneﬁt Design of Health Reimbursem*nt Arrangements and Health Savings Accounts The 2005 survey data provide detailed information on the beneﬁt design of HRA and HSA plans. We analyze the existence and magnitude of diﬀerences in the beneﬁt design of HRA and HSA plans by ﬁrm size. In general, small ﬁrms are thought to provide HI policies that are less generous than larger ﬁrms, though recent evidence suggests that small-ﬁrm and larger-ﬁrm policies are similar along many dimen-

State HI Mandates, CDHPs, and HSAs

sions (KFF/HRET, 2005).10 We revisit this issue, focusing on beneﬁt generosity in HRA and HSA plans. We estimate ordinary least squares (OLS) models for the monthly premium for a single individual, monthly worker contribution to the single premium, annual deductible for a single worker, annual ﬁrm contribution to a single worker, and the maximum out-of-pocket liability for a single worker. We estimate one set of models for HRA plans and another set of models for HSA plans. An important caveat with these models is that they have a relatively small number of observations—50 to 60 depending on the model. However, the key ﬁndings from the models remain the same after reestimating a parsimonious speciﬁcation that excludes detailed ﬁrm and industry characteristics. Table 3.6 reports the OLS regression estimates for HRA plans in 2005. We ﬁnd that small ﬁrms with 3 to 49 employees have signiﬁcantly lower premiums ($86.46) and signiﬁcantly higher deductibles ($912.35) than large ﬁrms, suggesting that they have somewhat lowerquality policies. However, ﬁrms with 50 to 199 workers have signiﬁcantly lower worker contributions ($52.15) and higher ﬁrm contributions ($730.94) and lower maximum out-of-pocket liabilities ($1,490) than large ﬁrms, suggesting that these ﬁrms are more generous than large ﬁrms. Firms with a higher proportion of low-income employees appear to oﬀer HRAs that have higher deductibles, however, we do not observe that these ﬁrms have signiﬁcantly lower premiums to account for the higher deductibles. Firms with a high proportion of part-time workers also have a lower ﬁrm contribution to HRAs, suggesting that worker demand for HI inﬂuences the ﬁrm’s contribution decision. We also observe diﬀerences in plan beneﬁts by industry and region. 10

The KFF/HRET employer surveys on HI beneﬁts have showed that there are no statistically signiﬁcant diﬀerences among small-ﬁrm plans and large-ﬁrm plans in their oﬀerings of prescription drugs, adult physicals, outpatient mental, inpatient mental, annual OB-GYN visit, oral contraceptives, and well-baby care. Only the propensity to oﬀer prenatal and chiropractic care diﬀered signiﬁcantly. Small-ﬁrm policies were likelier to have no policy limit (60 percent in small ﬁrms and 45 percent in large ﬁrms) and likelier than large-ﬁrm policies to have a limit on out-of-pocket spending (87 percent in small ﬁrms and 77 percent in large ﬁrms), and likelier to have higher deductibles ($559 in large ﬁrms and $280 in small ﬁrms for individual coverage) (KFF/HRET, 2004).

Monthly Total Premium (individual) Determinant

Monthly Worker Contribution to Premium (individual)

Standard Error

Estimate

Standard Error

–86.46a

47.93

–4.51

22.22

44.54

41.74

–52.15b

–24.83

67.96

–9.94

Annual Deductible (individual)

Annual Firm Contribution (individual)

Maximum Out-ofPocket Liability (individual)

Standard Error

Estimate

Standard Error

912.35b

391.74

209.48

209.01

19.35

181.36

323.58

730.94 c

172.64

20.02

31.51

1,171.64 b

481.99

–362.04

257.16

161.47

1,020.61

85.08

–52.37

39.44

–501.39

734.24

–783.82a

391.74

523.07

1,277.63

137.12a

74.97

–11.18

34.76

–255.82

673.78

–686.43a

359.48

1,241.56

1,125.88

37.49

32.14

–5.68

14.90

–274.56

271.12

–48.14

144.65

–605.66

482.61

Estimate

Standard Error

738.26

719.72

Firm composition Size: 3–49 workers Size: 50–199 workers Workforce earning $20,000 or less (%) Workforce working part time (%) Covered employees (%) Union

–1,490.23b

626.85

In the Name of Entrepreneurship?

Table 3.6 Beneﬁt Design of Health Reimbursem*nt Arrangement Plans (OLS Regressions, 2005)

Table 3.6—Continued Monthly Total Premium (individual)

Monthly Worker Contribution to Premium (individual)

Annual Deductible (individual)

Annual Firm Contribution (individual)

Estimate

Standard Error

Estimate

Standard Error

Estimate

Standard Error

Estimate

Standard Error

Mining

46.52

64.27

20.1

29.80

572.46

578.38

–83.24

Construction

32.65

53.64

37.99

24.87

722.41

480.98

–17.64

84.55

–7.01

39.20

241.47

12.99

94.36

10.17

43.75

129.09b

54.91

31.14

Financial

89.93b

42.29

Service

51.33

Government

79.9

Determinant

Maximum Out-ofPocket Liability (individual) Estimate

Standard Error

308.58

–150.35

865.10

–58.88

256.62

670.87

805.46

765.44

50.4

408.39

–1,523.54

1,269.77

–315.18

454.19

–647.98 b

242.33

–93.25

1,417.04

25.46

668.17

475.13

253.50

–102.84

824.56

17.71

19.61

85.4

357.83

–389.10 b

190.91

76.7

635.04

40.10

–3.78

18.59

328.67

–131.39

175.35

–490.03

602.16

70.17

–32.95

32.53

614.93

–226.92

328.08

–2,120.78a

388.04

–412.73a

207.03

–399.78

Industry

Wholesale Retail

Health care

52.05

48.33

–7.24

22.41

123 –431.77 –321.71

4.92

1,053.81 725.85

State HI Mandates, CDHPs, and HSAs

Transportation, utilities, communication

100

Monthly Total Premium (individual)

Monthly Worker Contribution to Premium (individual)

Estimate

Standard Error

Midwest

–1.92

36.03

–5.49

16.70

226.91

313.47

South

10.43

33.46

9.83

15.51

141.4

288.86

Determinant

Estimate

Standard Error

Annual Deductible (individual) Estimate

Standard Error

Annual Firm Contribution (individual)

Maximum Out-ofPocket Liability (individual)

Standard Error

Estimate

84.19

167.25

149.1

541.04

10.55

154.12

1,124.76b

502.54 737.77 633.14

Estimate

Standard Error

Location

West

51.01

49.13

15.38

22.78

273.21

398.62

245.36

212.68

1,660.42b

Urban

25.39

42.16

–30.09

19.55

–355.69

325.46

97.66

173.64

–130.11

Observations

R-squared

0.41

a Signiﬁcant at 10 percent. b Signiﬁcant at 5 percent. c Signiﬁcant at 1 percent.

54 0.38

66 0.45

66 0.53

54 0.41

In the Name of Entrepreneurship?

Table 3.6—Continued

State HI Mandates, CDHPs, and HSAs

101

Table 3.7 reports OLS regression estimates for HSA plans. Unlike the models for HRA plans, we observe almost no diﬀerence in beneﬁt design by ﬁrm size. The only exception is that it does appear that ﬁrms with 3 to 49 workers have signiﬁcantly higher individual premiums ($128.49 per month); however, we do not observe a statistically signiﬁcant diﬀerence in any other feature of the plan beneﬁt. Firms with a higher proportion of low-income workers have plans with a higher worker contribution to the premium, but we do not observe a statistically signiﬁcant diﬀerence on any other measure of plan beneﬁt. In summary, it appears that the evidence on plan generosity for small ﬁrms is mixed—HSA plans in small ﬁrms appear to have lower premiums and lower quality, but HRA plans appear to have somewhat higher premiums and do not appear to diﬀer in other dimensions. In general, it does not appear that small businesses are oﬀering plans that are systematically diﬀerent from larger businesses in generosity, along the full spectrum of beneﬁt features.

Consumer-Directed Health Plans Are Growing in Popularity but Do Not Appear to Be a Panacea for Small Businesses Our analysis of the KFF/HRET survey shows that, in general, small ﬁrms are no likelier than larger ones to oﬀer CDHPs and that their uptake of CDHPs has not grown any more rapidly than has that of larger ﬁrms. Small ﬁrms appear to have slightly higher rates of churning in their CDHP oﬀerings and seem to be somewhat likelier to adopt and drop CDHP policies. However, we ﬁnd no consistent evidence that CDHP oﬀerings vary systematically in premiums or generosity between small and large ﬁrms.

102

Monthly Total Premium (single) Characteristic

Estimate

Standard Error

Monthly Worker Contribution to Premium (single) Estimate

Standard Error

Annual Deductible (single) Estimate

Standard Error

Annual Firm Contribution (single)

Maximum Out-ofPocket Liability (single)

Estimate

Standard Error

Estimate

Standard Error

Firm composition 128.49a

62.44

26.57

33.90

–311.52

442.60

–16.29

278.44

–418.41

752.52

12.59

65.23

30.2

35.42

–219.53

429.96

9.24

270.49

–367.71

786.20

Workforce earning $20,000 or less (%)

107.53

105.74

117.19b

57.42

–865.69

706.08

–332.01

444.19

805.45

1,274.43

Workforce working part time (%)

237.63b

137.28

57.89

74.54

–1,581.18

999.19

133.51

628.59

–2,789.86

1,654.56

Covered employees (%)

256.44 b

126.18

121.49b

68.52

–1,355.90

869.10

432.36

546.75

–989.43

1,520.78

85.30 b

44.06

66.14 c

23.92

81.83

324.33

121.05

204.03

–86.37

530.96

Size: 3–49 workers Size: 50–199 workers

Union

In the Name of Entrepreneurship?

Table 3.7 Beneﬁt Design of Health Savings Account Plans (OLS Regressions, 2005)

Table 3.7—Continued Monthly Total Premium (single)

Monthly Worker Contribution to Premium (single)

Annual Deductible (single)

Annual Firm Contribution (single)

Estimate

Standard Error

Estimate

Standard Error

Estimate

Standard Error

Estimate

Mining

–357.72a

160.16

–36.18

86.97

817.11

1,165.77

Construction

–176.81b

102.23

37.44

55.51

596.17

–4.83

86.17

–9.2

46.79

–230.27b

134.86

–26.73

73.23

Characteristic

Maximum Out-ofPocket Liability (single)

Standard Error

Estimate

Standard Error

339.5

733.39

987.01

1,930.29

738.76

257.73

464.75

–1,265.21

1,232.10

–125.43

530.33

119.64

333.63

–1,010.49

1,038.50

1,311.86

929.79

655.8

584.93

–199.55

1,625.31

392.93

1,161.10

1,223.08

Industry

Wholesale Retail

62.19

101.48

94.49b

Financial

50.14

86.80

7.04

55.10

–17.28

624.60

880.02a

47.13

4.69

567.55

64.24

357.05

–899.46

1,046.14

441.05

264.97

277.46

–128.06

780.61

–27.32

64.77

21.29

35.17

797.04b

Government

37.33

89.10

3.47

48.38

868.98

611.92

–53.69

384.96

–638.74

1,073.85

Health care

16.69

74.30

57.54

40.34

330.16

517.07

190.13

325.29

33.26

895.46

Service

State HI Mandates, CDHPs, and HSAs

Transportation, utilities, communication

103

104

Monthly Total Premium (single) Characteristic

Estimate

Standard Error

Monthly Worker Contribution to Premium (single) Estimate

Standard Error

Annual Deductible (single) Estimate

Standard Error

Annual Firm Contribution (single) Estimate

Standard Error

Maximum Out-ofPocket Liability (single) Estimate

Standard Error

Location Midwest

–61.13

61.34

5.17

33.31

146.94

434.17

–92.04

273.13

–188.73

739.33

South

–38.12

58.13

–10.45

31.56

419.86

432.06

–49.34

271.81

–545.28

700.59

West

137.29b

78.32

25.09

42.53

289.32

543.06

–240.93

341.64

311.87

943.91

Urban

–70.23

51.46

–6.35

27.94

–482.92

352.28

–170.48

221.62

–373.93

620.24

Observations R-squared

50 0.52

a Signiﬁcant at 5 percent. b Signiﬁcant at 10 percent. c Signiﬁcant at 1 percent.

50 0.42

59 0.3

59 0.23

50 0.31

In the Name of Entrepreneurship?

Table 3.7—Continued

State HI Mandates, CDHPs, and HSAs

105

Conclusion Small-business HI reform is a policy issue that is continually in the limelight. Since the majority of uninsured working Americans are employed in small businesses, extending HI coverage to small businesses is an important mechanism for reducing the number of uninsured. States have continued to adjust their small-group HI-reform packages to make them more eﬀective. However, these incremental pricing and access reforms cannot be expected to solve the fundamental problems of high administrative costs, adverse selection, and a shallow risk pool that aﬄict the small-group HI market. Regulations that restrict premium variation may lower prices for some but increase prices for others; they may drive some insurers out of the market. Evidence reveals that policy approaches focused on regulating the insurance market have not improved access to or aﬀordability of HI to small businesses across the board and have led to distortions in the size of businesses right around the regulatory threshold. In other words, the regulations have not only failed to achieve their core aims, but they also have had unintended consequences related to business operations. Research suggests that policymakers need to be aware that legislative size thresholds may have unintended consequences on business size. Furthermore, incremental legislation that makes only small changes in the small-group HI market is unlikely to have large-scale eﬀects on HI oﬀering among small ﬁrms. Solutions to the problem of HI access and aﬀordability will likely need to address fundamental issues driving the escalation in HI costs. Indeed, the policy debate has shifted in this direction. For example, the Small Business Health Plan legislation in the House and Senate proposes to improve access and availability to HI by allowing small businesses to band together to purchase HI through their industry associations (U.S. Chamber of Commerce, undated). Another solution that has been advocated by the Bush administration and by policy analysts is the development of CDHPs. These plans aim to control costs by increasing consumers’ ﬁnancial responsibility and involvement in their health-care choices. Since CDHPs are potentially less costly than traditional health plans and may appeal to younger workers with low

106

In the Name of Entrepreneurship?

health-care demand, these plans may be well suited to workers in small businesses (Laing, undated). However, despite the enthusiasm for such plans among small-business advocates, evidence to date suggests that small businesses have been no likelier than larger businesses to oﬀer such plans. We examine evidence from the KFF/HRET survey and show that, in general, small ﬁrms are no likelier to oﬀer CDHPs and that their uptake of CDHPs has not grown any more rapidly than larger ﬁrms. We do ﬁnd some evidence that small ﬁrms are likelier to add and drop CDHPs. More information on the implementation of CHDPs, particularly within smaller ﬁrms, would be valuable in assessing the causes of such churning and, ultimately, whether CDHPs are indeed a panacea for small businesses. Because the marketplace for such options is changing rapidly, it will be important to monitor changes over the next few years. For example, as more ﬁrms enter the marketplace oﬀering services to manage HSAs, a small business with a small beneﬁts oﬃce might ﬁnd it plausible to oﬀer HSAs by contracting with such a ﬁrm. It may simply take time for these providers to emerge and for small businesses to learn about them.

CHAPTER FOUR

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis John Mendeloff, Christopher Nelson, Kilkon Ko, and Amelia Haviland

In 2002, some 56 percent of Americans were employed in businesses with fewer than 100 workers. It has long been argued that the burdens of safety and health regulation fall more heavily on these ﬁrms. Adopting prevention technologies and processes often involves considerable ﬁxed costs, which are more diﬃcult for smaller operations to absorb. Similarly, small businesses are less likely than their larger counterparts to be able to hire in-house safety experts and often lack the resources to remain aware of voluminous and changing safety regulations. Concern about regulatory burdens on small businesses has not escaped the attention of policymakers. SBREFA (P.L. 104-121) and its predecessor, RFA (P.L. 96-354), seek to increase the weight given to small-business concerns in the regulatory rulemaking and enforcement processes. Similarly, OSHA exempts workplaces with fewer than 11 workers from regular, programmed inspections and considers ﬁrm size when assessing penalties for violations of its safety and health standards. For ﬁrms with fewer than 500 workers, OSHA developed a consultation program that provides services largely independently of the enforcement program. Yet, regulations and other policies toward small businesses should be guided both by concern for potential costs to small businesses and by an understanding of the magnitude of the risks they face and the potential beneﬁts of prevention activities. Therefore, it is important to understand whether working for a small business is any more or less risky than working for a large business. If working for a small business

107

108

In the Name of Entrepreneurship?

is riskier, then we also need to understand whether the risk is due to the size of the establishment (i.e., the individual worksite) or to the size of the ﬁrm (i.e., the business organization, which consists of one or more establishments). Should policy eﬀorts be directed toward small ﬁrms, toward small establishments, or both? A better understanding of the distribution of risks can help policymakers design and target appropriate policies. Unfortunately, empirical research on the topic has been surprisingly scant, especially given the signiﬁcant number of policy initiatives targeting small businesses. While a small group of studies focuses on how risk changes with establishment size and a few look at the role of ﬁrm size, there has been no systematic attempt to disentangle the eﬀects of establishment and ﬁrm size. To shed light on these issues, we examined the relationship between the fatality rate, i.e., the number of deaths per 100,000 workers, and business size, both in terms of establishment size and ﬁrm size. Most of the analyses use fatalities investigated by OSHA between 1992 and 2001. We excluded the construction sector from most analyses due to concerns about the accuracy of the distinction between establishment and ﬁrm in these nonﬁxed work settings. As indicated in the chapter title, this work should be regarded as exploratory, not deﬁnitive. Data limitations and the scope of the problem limit our ability to understand fully the drivers of the size-fatality risk—and even, in some cases, to provide a full picture of the nature of the relationship. Rather than trying to provide the last word on the subject, our goal is to enrich the debate over safety in small businesses by providing a factual baseline and considering possible causal mechanisms and policy approaches. The remainder of the chapter proceeds as follows. First, we review some reasons we might expect the risk of injury or fatality to be higher at small ﬁrms than at large ﬁrms and discuss the results of previous research on the topic. We then brieﬂy describe the data and methods used in our analysis. We next present our ﬁndings, ﬁrst examining the simple relationships of risk to establishment and ﬁrm size, then considering the relationship between fatality risk and establishment and ﬁrm size, holding the other constant. We also discuss other issues, such

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis

109

as variables that might aﬀect the results, trends over time, the issue of underreporting, and the relationship between fatalities and violations of OSHA standards. Finally, we discuss the implications of our study for public policy and suggest directions for further research.

The Relationship Between Firm Size and Risk Previous research and theory lead us to expect that fatality risks will be higher in smaller ﬁrms and establishments. Understanding the reasons behind this expectation will help with data interpretation and policy implications. • Smaller ﬁrms might be expected to save less than larger ﬁrms do by preventing injuries. The limited actuarial experience at small ﬁrms means that they are subject to little or no experience rating by WC insurers. Thus, small ﬁrms will not see reductions in their WC premiums even if their injury losses decline. Small ﬁrms are also less likely to be unionized, and some evidence indicates that the presence of unions increases the probability that workers will receive higher wages to compensate for higher risks (Viscusi, 1983). Small ﬁrms also get reductions in OSHA ﬁnes, which also decreases the incentive to correct hazards. • Smaller ﬁrms are also likelier than larger ﬁrms to employ higherrisk workers (i.e., workers who are younger and unmarried and those have lower levels of education and experience) (Belman and Levine, 2004). They may not pressure management on safety issues as much as older or married workers would. These worker characteristics also may make it costlier for ﬁrms to achieve a given level of safety. • Both smaller ﬁrms and smaller establishments will be less able to realize economies of scale in the production of safety. Lacking in-house expertise, they may face higher marginal costs to obtain information about risks and how to reduce them. • Smaller establishments are less likely than large ones to be inspected, reducing the marginal beneﬁt of compliance.

110

In the Name of Entrepreneurship?

In sum, there appear to be good reasons to expect that both smaller ﬁrms and smaller establishments will exhibit higher levels of risk than larger ones. The reasons are more numerous and perhaps more powerful at the ﬁrm level. Tables 4.1 and 4.2 present a full list of potential factors considered.

Previous Research on Size and Risk A few studies have examined the relationship between establishment or ﬁrm size and risk. We brieﬂy review the ﬁndings from these studies here. Table 4.1 Factors Inﬂuencing the Predicted Effects of Establishment and Firm Size on Safety: Marginal Beneﬁts Affected Entity Firms

Factor Smaller ﬁrms have less experience rating in WC than larger ﬁrms have, reducing ﬁnancial incentive for investments in prevention. Smaller ﬁrms pay lower wages than larger ﬁrms do, reducing wagereplacement costs. Smaller ﬁrms are less likely than larger ﬁrms are to have unions and to pay high wage-risk premiums, reducing wage-replacement costs. Smaller ﬁrms face lower penalties than larger ﬁrms face from OSHA inspections and are less subject than larger ﬁrms to repeat violations if similar violations have been recently cited at other workplaces.

Establishments

Smaller establishments are less likely than larger establishments to have unions and to pay high wage-risk premiums; therefore, a reduction in risk may not save them as much in compensation costs as larger establishments might save. Smaller establishments are less likely than larger establishments to be inspected; therefore, compliance will have lower expected beneﬁts for them than for larger establishments.

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis

111

Table 4.2 Factors Inﬂuencing the Predicted Effects of Establishment and Firm Size on Safety: Marginal Costs Affected Entity Firms

Factor Smaller ﬁrms have higher costs of capital than larger ﬁrms have, making investments in prevention costlier. Smaller ﬁrms are likelier than larger ﬁrms to have higher-risk workers.

Establishments

Smaller establishments are less likely than larger ones to have easy access to safety expertise, increasing costs of prevention. Smaller establishments are less likely than larger ones to engage in safety training.

Fatalities and Other Serious Injuries

Previous studies have found an association between establishment size and occupational injury and illness risk.1 A 1990 study of more than 14,000 OSHA fatality investigations from 1977 to 1986 showed that reported fatality rates were usually highest at smaller workplaces across all major industry sectors (Mendeloﬀ and Kagey, 1990). The fatality rates for the smallest establishments (1 to 19 employees) were about four times the rates for the largest (more than 1,000 employees). To investigate whether the result was due to a compositional eﬀect (i.e., industries with a high fatality rate just happen to be those that are dominated by small establishments), the study examined rates within detailed industry categories—four-digit standard industrial classiﬁcations (SICs)—in manufacturing.2 Similarly sharp drops for establishment with more than 20 workers were observed in most industries in the analysis.

In this chapter, injury, unless noted, will refer to both injuries and illnesses. However, none of the data sources examined can be expected to do a good job of capturing illnesses with long latency periods.

For example, food and kindred products is a two-digit category within manufacturing, meat products is a three-digit category within food, and sausages and other prepared meats is a four-digit category within meat products.

112

In the Name of Entrepreneurship?

Other studies have found an association between smaller establishments and serious injury. An examination of the 1990 Workplace Industrial Relations Survey (WIRS) of British manufacturing establishments with 25 or more employees (Nichols, Dennis, and Guy, 1995) cites earlier work by Thomas (1991), which found that the health and safety executive (HSE) major rate (which includes relatively serious categories of injuries, e.g., amputations) decreased with establishment size.3 Fenn and Ashby (2001), reporting on the ﬁndings from the 1998 WIRS of about 2,000 British establishments with more than 10 employees, found that doubling the number of employees at an establishment was associated with a 33-percent reduction in reported injuries and a 25-percent reduction in reported illnesses. Finally, Bennett and Passamore (1985) found that fatality rates in coal mining decreased as establishment size increased.4 Less Serious Injuries

Some research on less serious injuries has shown that small establishments (i.e., with 1 to 19 employees) have lower rates of such injuries than large establishments have. For example, the U.S. Bureau of Labor Statistics has regularly reported that small establishments have a relatively low lost-workday frequency rate.5 The rates increase from the smallest size category to the category with 100 to 250 employees and then decline with increasing size for establishments with more than 250 employees. In all sectors except construction and mining, the

Nichols, Dennis, and Guy (1995) also present data that indicate that establishments that are part of larger ﬁrms have higher HSE major rates than those that are independent, but the conclusion is based on small numbers and fails to control for industry composition.

Using the U.S. Bureau of Labor Statistics (BLS) National Census of Fatal Occupational Injuries (CFOI), Peek-Asa et al. (1999) analyzed fatalities in the retail trade sector, in which 89 percent of the deaths were due either to transportation accidents or to assaults. They found that establishments with fewer than 20 workers had higher-than-average fatality rates. It is plausible that workers in, for example, minimarts are more vulnerable to assaults than are workers in department stores. The possible patterns for car-crash deaths are less clear-cut and deserve further attention. In our analyses, we exclude fatalities from these two causes.

Lost-workday injuries includes both injuries resulting in one or more days away from work and injuries resulting only in restricted work activity.

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis

113

smallest establishment-size category has the lowest rate. For those two exceptions, the smallest size has the second lowest rate, second only to establishments with more than 1,000 workers. In contrast to his ﬁndings for the HSE major injury category, Thomas (1991) found, in the same study, that the rates for a somewhat less serious injury category (more than three days oﬀ work but not in the HSE major category) increased with establishment size. One study of less serious injuries that did ﬁnd decreases in rates with larger sizes was by Haberstroh (1961). His study of 53 integrated steel mills found that, from 1948 to 1957, a 10-percent increase in employment, for both establishment and ﬁrm size, led to about a 3-percent decrease in the frequency of disabling injury rates. What could explain the disparity we usually ﬁnd between the size patterns for more and less severe injuries? The fairly consistent pattern we ﬁnd is that, as injuries become severer, the relative performance of smaller establishments worsens. Rates for fatalities and HSE major injuries show higher rates for the smallest establishments; rates for the less serious injuries in Britain and for the U.S. lost-workday rate show better performance there.6 One explanation could be that establishments in diﬀerent size categories truly diﬀer in their rates for more and less severe injuries. Another explanation could be that smaller establishments have a higher rate of underreporting but that the underreporting is less for more serious injuries. We discuss each explanation in turn. Accident Types and Size

It is plausible that diﬀerent types of injury-causing events might display diﬀerent frequencies across size groups. This result would require two elements: ﬁrst, that diﬀerent accident types vary in the probability that death will result; and second, that workplaces of diﬀerent sizes vary in the composition of these accident types.

One exception is a comparison we made between the rates from the BLS survey for lostworkday injuries and the rates from the survey for medical only cases, which do not involve time lost from work or restricted work activity. The patterns by establishment-size category were almost identical for the two groups of injuries; the relative rates for the smallest workplaces were only slightly higher for the severer category.

114

In the Name of Entrepreneurship?

The ﬁrst element is certainly present. The causes of fatalities do diﬀer considerably from the causes of nonfatal injuries and illnesses. Even ignoring highway motor-vehicle crashes and assaults (which are largely excluded from the database we examine here), we ﬁnd that other causal event types such as ﬁres and explosions also account for a much larger share of fatalities than they do of nonfatal injuries. Similarly, injuries caused by overexertion (e.g., sprains and strains) comprise about 40 percent of all lost-workday injuries but only a tiny share of deaths. Whether the rate of diﬀerent accident types varies for workplaces of diﬀerent sizes and whether these diﬀerences could account for major diﬀerences in fatality rates will be explored in our analysis of the relationship between establishment size and the causes of fatalities. Underreporting and Size

It seems plausible that more serious injuries might be less subject to underreporting. For example, Leigh, Marcin, and Miller (2004) reviewed many studies that indicate that the BLS survey substantially undercounts nonfatal injuries, perhaps by 40 percent for the sectors covered. They note that “evidence suggests that small ﬁrms are especially prone to underreport” (Leigh, Marcin, and Miller, 2004). Similarly, Seligman et al. (1988) reported that compliance with OSHA recordkeeping requirements was poorest at small ﬁrms and best at the largest ones. However, Glanzer et al.’s (1998) study of a large construction project found fewer instances of underreporting for injuries that involved lost workdays than for injuries without lost workdays. Oleinick, Gluck, and Guire (1995) suggest that lower reported rates for less severe injuries at smaller establishments in Michigan were probably due underreporting. They found that smaller establishments tended to have more risk factors for injury than did larger workplaces. For example, they found that smaller establishments had such risk factors as younger workers, a higher percentage of males, and more construction work.7 Because they found evidence that there were more risk 7

A higher turnover rate of workers has also often been linked to higher injury rates. Oleinick, Gluck, and Guire (1995) cite a study by Berkeley Planning Associates and SBA (1988)

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis

115

factors at smaller establishments, Oleinick, Gluck, and Guire (1995) concluded that the lower reported rates for less severe injuries at these workplaces were probably a result of underreporting. Morse et al. (2004) also concluded that both large and small businesses have underreported cases of occupationally related musculoskeletal disorders (MSDs) but that there appeared to be more underreporting in smaller businesses. (It was unclear in this study whether the survey responses pertained to establishment or ﬁrm size.) The researchers conducted a population-based survey in Connecticut that found that, controlling for age, gender, physical risks, and occupation, employees of smaller businesses had a marginally signiﬁcantly higher risk of occupationally related MSDs than did employees of larger businesses. The authors thus concluded that there was general underreporting of MSDs but that there appeared to be more underreporting in smaller ﬁrms. Summary

This review of studies done to date suggests that the rates for severe injuries (especially fatalities) are highest in the smallest establishments. For less serious injuries, in contrast, we ﬁnd somewhat lower rates in the smallest establishments. Regardless of whether the latter ﬁndings are an artifact of underreporting, the ﬁndings for deaths and severe injuries should generate concern about what is happening at smaller establishments.

Data and Methods Our analyses examine fatality rates, i.e., the number of fatalities during a period divided by the number of worker-years of exposure during that

that indicated that new hire rates were about 5 percent higher at small ﬁrms than at large ones, though another survey found no diﬀerences in turnover by size (Pedersen and Sieber, 1989).

116

In the Name of Entrepreneurship?

period. If an industry employs an annual average of 1,000 workers over a 10-year period, then there are 10,000 worker-years of exposure.8 A full discussion of our data and methods appears in Appendix B. Here, we present a brief review of our approach. We begin by describing how fatality rates were derived for our analysis, ﬁrst for the numerator (the number of deaths) and then for the denominator (exposure to the risk of death). We then brieﬂy describe the regression analyses used to add more control variables to the analysis. Number of Fatalities (Numerator Data)

Our data on fatalities come from the OSHA Integrated Management Information System (IMIS) and are generated by the accident investigations that OSHA conducts when work fatalities (and some other serious injuries) are reported. Employers are required to telephone OSHA within 24 hours after the death of an employee. Exclusions have been made for deaths due to highway crashes, intentional violence, and some other causes. (These categories account for almost half of the deaths identiﬁed in the BLS CFOI.) On average, OSHA has investigated somewhat fewer than 2,000 deaths per year. Its investigations provide information on both the number of employees at the establishment (worksite) where the death occurred and the total employment at all worksites of the ﬁrm. Most of our analyses focus on deaths OSHA investigated from 1992 to 2000, a sample that includes 17,481 fatalities. However, we also look at deaths back to 1985 for a subset of states for which there is continuous reporting. Exposure to the Risk of Death (Denominator Data)

Data on the number of workers employed are needed to provide a measure of exposure to risk and thus serve as the denominator in the fatality-rate calculations. For industry employment, we relied on county business patterns (CBPs) (U.S. Census Bureau, undated[b]) and a spe8

Our data do not distinguish between full-time and part-time workers. This shortcoming will give rise to a tendency to overestimate rates for industries and size categories that have more employees working part time.

Small Businesses and Workplace Fatality Risk: An Exploratory Analysis

117

cial table prepared for us by the U.S. Census Bureau that distributed establishment employment by size category and industry into ﬁrmsize categories. The table tells us, for example, how many employees in meatpacking establishments with 20 to 49 workers are employed by ﬁrms with 20 to 49 workers, how many are employed by ﬁrms with 50 to 99 workers, and so on. Regression Analyses

To see whether our conclusions about the eﬀects of ﬁrm and establishment size might be biased due to the omission of variables with which they might be correlated, we conducted regression analyses that allowed us to control for the eﬀects of some other variables. To examine whether unionization or metropolitan location aﬀected death rates, we constructed a data set that included both accident investigations and programmed inspections. The latter were scheduled randomly during the period used in this analysis but were limited to establishments with more than 11 employees in industries whose lost-workday injury rates exceeded the industrywide average. So we limited the accident investigations in the same way and investigated, using Poisson regression, how the probability of a fatal accident varied with establishment size and ﬁrm size, holding other variables constant.

Findings In this section, we review the ﬁndings of our analysis. We ﬁrst provide a description of the data in terms of the number of deaths investigated over time, by industry, and by establishment and ﬁrm size. Then we examine the relationship between fatality rates and establishment size, both for various industry sectors and then for a selected set of detailed industries. Next, we consider the relationship between fatality rates and ﬁrm size at the level of the industry sector and the eﬀects of establishment size while holding ﬁrm size constant and vice versa. We also present the results of some analyses that control for additional factors that may aﬀect the relationship between fatality rates and ﬁrm and establishment size and will examine whether there are size-based

118

In the Name of Entrepreneurship?

diﬀerences in the causes of fatalities, especially in the role played by serious health and safety violations. Finally, we review trends in establishment-size fatality rates over time. Overall Patterns in the OSHA Fatality Data

From 1992 to 2001, OSHA investigated a total of 17,481 fatalities. Nearly 39 percent of these deaths occurred in the construction sector, with manufacturing a distant second. Table 4.3 shows the percentage distribution of deaths in diﬀerent employment-size categories for both establishments and ﬁrms. A separate distribution is shown for construction because these size categories have a diﬀerent meaning in construction. Even outside of construction, almost 42 percent of deaths investigated by OSHA were in establishments with fewer than 20 employees, though only 27 percent occurred in ﬁrms of that size. In contrast, only 6 percent of deaths Table 4.3 OSHA-Investigated Fatalities in Each Establishment- and Firm-Size Category, All States, 1992–2001 Nonconstruction (N = 10,742) (%) Employees

Establishment Size

Construction Only (N = 6,739) (%)

Firm Size

Establishment Size

Firm Size

In the Name of Entrepreneurship? The Logic and Effects of Special Regulatory Treatment for Small Business - PDF Free Download (2024)