Government Size and Implications for Economic Growth

Magnus Henrekson, Andreas Bergh

Description

As economists and policymakers strive to understand the causes of the global financial crisis, pinpointing the relationship between government size and economic growth is crucial. In this incisive economic study, Andreas Bergh and Magnus Henrekson find that in wealthy countries, where government size is measured as total taxes or total expenditure relative to GDP, there is a strong negative correlation between government size and economic growth-where government size increases by 10 percentage points, annual growth rates decrease by 0.5 to 1 percent. Bergh and Henrekson stress that statistical correlations, even when highly significant, are not law. Some countries with high taxes enjoy above-average growth, and some countries with small governments have stagnant economies. The Scandinavian welfare states, for example, have enjoyed steady growth over the last decade despite their large governments. However, these nations compensate for high taxes by employing market-friendly policies in other areas, such as

Sujets

• Economics
• Business
• International

Informations

• Publié par: AEI Press
• Date de parution: 16 juillet 2010
• Nombre de lectures: 0
• EAN13: 9780844743547
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Government Size and Implications for Economic Growth
Government Size and Implications for Economic Growth
Andreas Bergh and Magnus Henrekson
Distributed by arrangement with the Rowman Littlefield Publishing Group, 4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706. To order call toll free 1-800-462-6420 or 1-717-794-3800. For all other inquiries please contact AEI Press, 1150 Seventeenth Street, N.W. Washington, D.C. 20036 or call 1-800-862-5801.

This publication is a project of the National Research Initiative, a program of the American Enterprise Institute that is designed to support, publish, and disseminate research by university-based scholars and other independent researchers who are engaged in the exploration of important public policy issues.
Library of Congress Cataloging-in-Publication Data
Bergh, Andreas. Government size and implications for economic growth / Andreas Bergh and Magnus Henrekson.
p. cm.
Includes bibliographical references.
ISBN-13: 978-0-8447-4327-1 (cloth)
ISBN-10: 0-8447-4327-5 (cloth)
ISBN-13: 978-0-8447-4353-0 (pbk.)
ISBN-10: 0-8447-4353-4 (pbk.)
[etc.]
1. Expenditures, Public. 2. Economic development. 3. Taxation. I. Henrekson, Magnus. II. Title.
HJ7461.B47 2010 338.9-dc22
2010009573
14 13 12 11 10 1 2 3 4 5 6 7
2010 by the American Enterprise Institute for Public Policy Research, Washington, D.C. All rights reserved. No part of this publication may be used or reproduced in any manner whatsoever without permission in writing from the American Enterprise Institute except in the case of brief quotations embodied in news articles, critical articles, or reviews. The views expressed in the publications of the American Enterprise Institute are those of the authors and do not necessarily reflect the views of the staff, advisory panels, officers, or trustees of AEI.
Printed in the United States of America
Contents
L IST OF I LLUSTRATIONS
A CKNOWLEDGMENT
I NTRODUCTION : W HY G ROWTH I S I MPORTANT, AND W HY G OVERNMENT S IZE M AY M ATTER
1. H OW D O W E K NOW I F B IG G OVERNMENT I S G OOD OR B AD F OR G ROWTH ?
Why Seemingly Clear Evidence May Be Insufficient
Theoretical Expectations: Should We Expect Big Government To Be Good or Bad for Growth?
Arguments Pointing toward a Positive Link between Government Size and Growth
Arguments Pointing toward a Negative Effect
Weighing the Positive against the Negative Effects
2. W HAT D O E XISTING S TUDIES S HOW?
Early Cross-Country Studies
More Sophisticated Studies A: Bayesian Averaging of Classical Estimates (BACE)
More Sophisticated Studies B: Fixed Effects Panel Studies
Combining BACE with Panel Data
Is the Negative Correlation Due to Reverse Causality?
Results from Other Research Explaining Growth
Concluding Remarks
3. T HE G ROWTH E FFECTS OF I NSTITUTIONAL Q UALITY
Why Do Institutions Matter?
The Surprising Correlation between Big Government and Reforms Furthering Economic Freedom
Conclusion: Not Only Size Matters
4. D EFICIENT M ARKETIZATION OF H OUSEHOLD P RODUCTION IN H IGH -T AX S OCIETIES
C ONCLUSIONS AND P OLICY I MPLICATIONS
Policy Implications in General
Should Government Be Allowed to Grow in the United States?
N OTES
R EFERENCES
A BOUT THE A UTHORS
List of Illustrations
F IGURES
I-1
The Value of an Initial Investment of $1,000 with Annual Growth at 2 or 3 Percent
I-2
Top Marginal Income Tax Rate in Sweden and the United States, 1970-2003
I-3
Taxes as a Share of GDP in Sweden and the United States, 1925-2007
1-1
Cross-Country Correlation between Growth and Government Size Measured as Tax Revenue over GDP, Annual Pairs, OECD 1970-2005
1-2
Cross-Country Correlation between Growth and Government Size, Excluding Countries with Annual Growth Exceeding 6 Percent, OECD 1970-2005
1-3
Cross-Country Correlation between Growth and Government Size Using Averages over Decades rather than Annual Values of Growth and Government Size, OECD, 1970-2005
3-1
Increase from 1970 to 2000 in KOF Index Values of Globalization and Economic Freedom Index (EFI) Compared to Tax Share of GDP in 1970
4-1
Average Hours Worked Per Person 15-64 Years of Age, Sweden and the United States, 1956-2003
C-1
Economic Freedom in Sweden and the United States, 1970-2006
T ABLES
2-1
Some Early Cross-Country Studies
2-2
Thirteen Variables That Robustly Explain Average Annual Growth in 1960-1996
2-3
Recent Panel Data Studies
2-4
The Bergh and Karlsson Study (2010)
2-5
The Growth Effects of Four Variables Found Robust
2-6
Comparison of Estimates in Different Studies-Dependent Variable: Annual Growth Rate of Real GDP Per Capita
4-1
Paid Work, Unpaid Work, and Time with Children in Sweden and the United States, Latest Available Year from Time-Use Studies
Acknowledgment
The authors thank Daniel Hedblom for excellent research assistance.
Introduction: Why Growth Is Important, and Why Government Size May Matter
The debate regarding the relationship between government size and economic development has been intense for decades. The state of research is seemingly contradictory, with some scholars arguing that big government decreases growth, and some declaring this not to be the case.
A close look at the literature reveals that results are not as conflicting as they may at first appear. Different researchers have used different measures of growth and of government size, and they have studied different types of countries. When we focus on studies that examine growth of real gross domestic product (GDP) per capita over longer time periods, the research is actually close to a consensus: In rich countries, there is a negative correlation between total size of government and growth.
To arrive at this conclusion, we have reviewed a wide body of literature on the subject. Here we describe the state of research and discuss what findings can be trusted, as well as the most important policy implications of those findings. But first, some words on why the issue itself is important.
The growth effects of government size are crucial, both in countries with relatively small governments, such as the United States, and in countries like Sweden, where government spending exceeds 50 percent of GDP From one year to the next, the difference between annual economic growth at 2 percent or 2.5 percent is important enough, since it means several billions of dollars, more or less, in the hands of both households and politicians. From a longer perspective, the level of annual growth of GDP per capita is even more important: It ultimately determines which countries will grow rich and which will become or remain relatively poor.
As shown in figure I-1 , an annual growth rate of 2 percent means that the economic standard of living doubles in thirty-six years. But if the annual growth is instead 3 percent, a doubling of the standard of living takes a mere twenty-four years.
F IGURE I-1 T HE V ALUE OF AN I NITIAL I NVESTMENT OF $1,000 WITH A NNUAL G ROWTH AT 2 OR 3 P ERCENT

Focusing policy on economic growth and standard of living may seem narrow-mindedly materialistic to some. But growth is ultimately about becoming more efficient in getting what we want-including food, housing, and health care, as well as literature, arts, and leisure. It is hardly surprising, therefore, to find a strong positive correlation between GDP per capita and other measures of well-being, such as longevity and infant mortality, as demonstrated by Lant Pritchett and Lawrence Summers (1996) and visualized by Hans Rosling (2010) using statistics from the United Nations. Despite some claims to the contrary, a clear link also exists between income and happiness, as recently shown by Betsey Stevenson and Justin Wolfers (2008).
The misconception that growth is a materialistic concept probably comes from its typically being measured by looking only at GDP per capita. This measures economic activity in the formal sector of the economy without taking into account how many hours are actually worked. For example, Schmid (2008) shows that the United States is about 30 percent richer than Sweden-but that approximately half of the difference can be explained by the fact that Swedes work fewer hours with respect to market work.
The question is, how do we know if Swedes work fewer hours because they simply prefer leisure more than Americans do, or because they must pay much higher taxes when they work? The answer is suggested by the greater number of hours Swedes spend on unpaid household work as compared to Americans. If Swedes simply liked leisure more than Americans do, they would, presumably, work less than Americans both on the market and in the household. The greater amount of time they spend on unpaid household work strongly implies that taxes are part of the explanation. Taxes affect the choice between taxed market work and untaxed household work or leisure. When taxes are high, market work pays less, and leisure and household work become more attractive. 1 In the standard neoclassical growth model, this affects only the level of income and not its growth rate. But when services are provided by professionals, incentives emerge to invest in new knowledge, to develop more effective tools and superior contractual arrangements, and to create more flexible organizational structures. Thus, higher rates of personal taxation reduce the scope for entrepreneurial expansion into new market activities that economize on time use or supply close substitutes for home-produced services. And without the emergence of new service jobs replacing traditional manufacturing jobs, the demand for tax-financed transfers increases. This puts an upward pressure on the tax and expenditure ratios, which provides an additional channel resulting in slower economic growth.
When discussing taxes, it is important not to confuse marginal tax rates with average taxes. The marginal tax (for income) is the tax rate applie