Political Economics: Redistributive Policies 8899902070, 9788899902070

Citation preview

MyBook Political Economics Vincenzo Galasso

T

his book is written for undergraduate students in Economics and in Political Science who want to learn about the political economics of redistributive policies. It provides a positive analysis of the political process behind the design and implementation of redistributive policies, by using the minimum level of mathematical formalization required, in an attempt to be accessible to second- and third-year undergraduate students. The book does not span the entire domain of political economics but concentrates on redistributive policies. This includes monetary transfers, such as pensions and unemployment benefits, as well as regulations, in the labor and product market, which allow to redistribute economic rents. Although the book is mostly self-contained, readers are expected to have a basic knowledge of microeconomics.

Political Economics Vincenzo Galasso

Vincenzo Galasso is Professor of Economics at Bocconi University, where he is the Director of the BSc in International Politics and Government, Research Fellow at the Center for Economic Policy Research (CEPR), Member of the Council of Economic Advisors to the Italian Prime Minister and Editor of the European Journal of Political Economy. He has widely published in top academic journals. He received his PhD in Economics from the University of California at Los Angeles.

A Global Economic Redistributive Policies History

A Global Economic History

ISBN 978-88-99902-07-0

9 788899 902070 Euro 20,00

MyBook http://mybook.egeaonline.it MyBook is the gateway to access accompanying resources (both text and multimedia), the BookRoom, the EasyBook app and your purchased books.

Political Economics.indd 1

7 mm

05/09/17 10:31

Political Economics Vincenzo Galasso

The Basic Models Redistributive Policies

Frontespizio Political Economics.indd 1

05/09/17 10:31

Copyright © 2017 Bocconi University Press EGEA S.p.A.

EGEA S.p.A. Via Salasco, 5 - 20136 Milan, Italy Phone + 39 02 5836.5751 - Fax +39 02 5836.5753 [email protected] - www.egeaeditore.it

All rights reserved, including but not limited to translation, total or partial adaptation, reproduction, and communication to the public by any means on any media (including microfilms, films, photocopies, electronic or digital media), as well as electronic information storage and retrieval systems. For more information or permission to use material from this text, see the website www.egeaeditore.it Given the characteristics of Internet, the publisher is not responsible for any changes of address and contents of the websites mentioned.

First Edition: September 2017

ISBN Domestic Edition 978-88-99902-07-0 ISBN International Edition 978-88-85486-27-0 ISBN Pdf Edition 978-88-85486-28-7

Print: Digital Print Service, Segrate (Milan)

Table of Contents

Prologue 1. Introduction to Political Economics 1.1 The Political Economic Approach 1.2 Political Institutions 1.3 A simple Example: Indirect Utility Function, Single-Peaked Preferences and the Bliss Point 2.

Electoral Models 2.1 The Median Voter Model 2.2 Probabilistic Voting Model 2.3 An example of Simple Majority Voting and Probabilistic Voting

VII 1 3 7 12 15 17 21 29

3. Lobbying 3.1 Public Policy: A Local Public Good 3.2 Lobbying

33 34 38

4. Facts, Data and Relevant Issues of the Welfare State 4.1 Economic Policies and the Role of Government 4.2 Size and Composition of the Welfare State 4.3 Labor Market Issues 4.4 The Economic Approach to Government Intervention 4.5 The Political-Economics Approach to Government Intervention 4.6 Citizens’ Opinions on the Welfare State

43 44 48 52 56

5. Redistributive Transfers in the Welfare State 5.1 The Economic Model 5.2 The Redistributive System: Winners and Losers

65 66 68

60 61

VI

POLITICAL ECONOMICS

5.3 The Political Decision 5.4 Discussion 6. Pensions 6.1 The Economic Environment of the Voting Models 6.2 Individual Preferences over Pension Systems 6.3 Voting on Pension Systems 6.4 Population Aging and Pension Systems 89

72 73 77 79 81 87

7. Labor Market Policies 7.1 A simple Political Economy Model 7.2 Labor Market Institutions 7.3 The Political Game 7.4 Discussion

93 94 96 99 100

8. Structural Reforms 8.1 Describing Structural Reforms 8.2 Determinants of Structural Reforms 8.3 Designing Structural Reforms

103 104 108 111

9. The Politics of Anger 9.1 The Effects of Migration on the Natives: Theoretical Aspects 9.2 The Economic Effects of Migration: Some Empirical Evidence 9.3 Individual Preferences over Migration Policy 9.4 Parties Preferences on Immigration Policy 9.5 Only Migration Causes Anger?

115 117 120 122 126 126

Bibliography

129

Prologue

This book is written for undergraduate students in economics and in political science who want to learn about the political economics of redistributive politics. While being initially offered only at Master and PhD level, political economics has recently gained interest even among undergraduate students, as understanding the complex political processes behind the policy-making of public policies has become ever more important. Yet, classical textbooks in political economics tend to be highly formalized and thus to target more advanced students. To fill this gap, this book provides a positive analysis of the political process behind the design and implementation of redistributive policies, by using the minimum level of mathematical formalization required, in an attempt to be accessible to undergraduate students in their second or third year. The book does not span the entire domain of political economics but concentrate of redistributive policies. This includes monetary transfers, such as pensions and unemployment benefits, as well as regulations, in the labor and product market, which allow to redistribute economic rents. Although the book is mostly self-contained, readers are expected to have a basic knowledge of microeconomics.

The book begins with an introduction to the basic concept of political economics (chapter 1). The next two chapters develop the political economic tools needed to analyze redistributive policies. Special emphasis is given in chapter 2 to the understanding of two popular voting models: the median voter model, which has been the workhorse in political science for several decades, and the probabilistic voting model. Chapter 3 presents a simple model of decentralization of policy decisions and introduces the classical lobbying model. The focus on redistributive policies begins in chapter 4, which portrays a picture of the government intervention in the economy, provides a positive rational for this public involvement and shows its magnitude in different OECD countries and over time. A simple political economic theory of redistributive

VIII

POLITICAL ECONOMICS

policies is introduced in chapter 5, which analyzes the political process, driven by electoral concerns, may lead to redistribution from high to low income individuals. Yet, most current redistributive policies have an intergenerational cleavage. The politics of pension systems, which transfer resources from (young) workers to (elderly) retirees, is discussed in details in chapter 6. Regulations in the labor market may also induce redistribution (of working rights) from (young) unemployed or workers on temporary contracts to (elderly) senior workers. This aspect is examined with a theoretical model in chapter 7. The last part of the book summarizes the recent literature in political economics to discuss two hot issues in the political debate: structural reforms and migrations. Chapter 8 provides an overview of the structural reforms of labor and product markets that have been implemented in the last few decades and analyzes the economic and political drivers of the successful reform attempts. Chapter 9 addresses the recent politics of anger. The electoral results of the last few years indicate that many voters are upset with the current economic and political system. What are the drivers of this discontent? Chapter 9 summarizes the recent political economic literature to discuss the role of immigration and globalization in the recent populist wave.

1. Introduction to Political Economics

Economic policies differ widely across countries and – within the same country – even over time. In 2014, within developed countries – in particular among those belonging to the OECD (Organization of Economic Cooperation and Development) – government expenditure ranged from less than 40% in the US to almost 60% in Finland. Large differences also emerged in the size of welfare transfers, with social benefits below 15% in the US, but almost 30% in France. Economic investments by the public sector were between 4% and 6% of GDP in most OECD countries, but below 3% in Spain. Further large differences may be appreciated in economic policies related to the welfare state, such as the generosity and the design of the unemployment insurance system, the degree of employment protective legislation (EPL) (which regulates the obligations relating to employees’ rights, and the size and the design of social security systems. Among the economic policies not directly related to the welfare state, large variations can be found across countries and over time in monetary policies and public debt management. The many tools provided by economic theory generally fail to offer a complete and satisfactory explanation for these wide differences. Particularly in public economics, the aim is often to provide normative statements. The aim is to design the most efficient policies for different economic contexts in order to enhance the economic welfare of society. Economic theory is less successful at presenting positive explanations for the observed economic policies and their differences across countries. Yet, a recent stream of economic literature has devoted a large attention to the analysis of these differences and to the decision-making processes behind public policy. This new stream of literature goes under the name of political economics. Its distinctive aim is to provide an explanation of the observed public policies – particularly economic policies, and to evaluate how these policies are influenced by political factors. Its starting point is to recognize that economic policies do not need to be efficient to be adopted. In other words,

2

POLITICAL ECONOMICS

economic policies need not to increase the economic well-being of every individual in society, but rather to obtain enough political support to be adopted by the policy-makers and by the legislative body. Hence, labor market reforms need not to increase the well-being or the employment prospects of all current or potential individuals in the labor market. Rather, they have to be supported by crucial political players, such as voters, lobbies or “veto” players – for instance, trade unions and firms’ representatives. This political support makes their implementation expedient for elected policymakers, such as the Ministries of Labor and of Finance, who seek re-election. This chapter describes this new political economic approach, by exploiting its similarities with the economic approach. Individuals are examined in a double role: they are both economic and political agents. Hence, they have preferences over economic outcomes, which guide their economic decisions but also on political outcomes. These individual preferences determine their political behavior. Methodological tools are introduced to explain how the political process converts the individual preferences of each agent over a given public policy – for instance, the level of redistribution – into aggregate preferences. These aggregate preferences characterize the will of society as a whole, and eventually lead to an outcome, consisting of the implementation of a public policy. In this chapter, particular emphasis is placed on political institutions, where individual preferences are aggregated. The most common mechanism that aggregate preferences are elections. As we will discuss in later chapters, the specific characteristics of elections, such as electoral rules, will be crucial in determining political aggregation and hence public policy. The goal of this book is to apply the political economic approach to the analysis of redistributive policies in several OECD countries. We aim to provide a positive explanation of the differences observed in the size, design and development of the programs that compose the welfare state systems in OECD countries. Why, for example, do Spain and the UK protect their workers against the risk of being unemployed using different combinations of employment protective legislation and unemployment benefits? Why is the Italian welfare state composed almost entirely of pension benefits, whereas Scandinavian countries dedicate a much larger amount of public resources to active labor marker policies? These are the type of questions we aim to answer in the following chapters by adopting the political economic approach, and introducing specific methodological tools. The initial analysis will be based on political institutions – simple majority voting elections or lobbying – to aggregate individual preferences into a policy outcome. We shall examine whether differences in the economic and demographic scenarios induce different preferences in the electorate of the OECD countries, which translate into different economic policies. At a later stage, we shall argue that the unexplained differences in the size and the design of the welfare states in these countries, having ac-

1.

Introduction to Political Economics

3

counted for the demographic and economic differences, may be due to the existence of different political institutions. 1.1 The Political Economic approach Individuals make economic choices. Every day, they select goods to consume and services to use, they choose how much to save for future consumption, they select asset holdings in their portfolio or perhaps they decide to delegate these portfolio decisions to an economic advisor or a financial institution. Indeed, several economists – notably the Nobel Prize winner Gary Becker – suggest that economic motives are so pervasive in families’ lives to be behind such important decisions as who to marry, how many children to have, where to live, whether to migrate or not, and whether an individual becomes a criminal, a drug deal or a suicide bomber (Becker, 1981). For the purposes of this section, the relevant aspect to highlight is that most of these economic decisions are affected by public policies. Consider an individual deciding how to allocate his resources among different assets. Clearly, this agent will consider economic factors in selecting his portfolio, such as the risk profile of different assets, their expected returns, their duration and the co-variance among these assets, and with his own level of human capital. All these elements are determined in the financial markets, and the government – or the public sector at large – does not need to influence the working of these markets. Typically, however, the returns from these assets are taxed. The public sector intervenes by imposing a capital income tax on these returns, often distorting portfolio decisions. For instance, assume that among these assets there is a government bond, whose returns are not taxed. Clearly, this bond would be more attractive, and would hence be in demand thanks to its special fiscal status. Economic decisions are thus influenced by public policy. In this particular example, fiscal policy matters. Analogously, retirement decisions are strongly affected by the pension system’s design. Generous pensions for individuals who retired before reaching normal retirement age were very common in many European countries. Several studies – for example, Gruber and Wise (1999 and 2004) – show that individuals reacted rationally to incentives provided by the pension system, and retired as soon as they were entitled to. Again, a public policy – the design of the retirement incentives in the pension system – had a large impact on individuals’ economic decisions. Public policies may modify individuals’ well-being even when agents do not change their economic decisions. A policy awarding a monetary transfer to people living in a particular geographical area increases the resources of its in-

4

POLITICAL ECONOMICS

habitants, whose economic well-being will hence improve, but it does not produce relevant changes in their economic choices. 1 These examples suggest that, as public policies influence individuals’ well-being and may modify their economic decisions, individuals will care about these policies. To put it differently, they will form an opinion and have preferences over these public policies. For instance, low income workers will welcome redistributive policies that provide them with additional resources, whereas high income workers will oppose policies from which they stand to lose. A crucial message of political economic literature is that the adoption of public policy depends on the preferences of individual agents. As suggested before, and displayed in Figure 1, economic agents form their preferences over a public policy according to the policy’s impact on their economic well-being. This information becomes clear in the markets. When selecting portfolios, for example, individuals evaluate the net return – after taxes – from different assets, and assess how the existence of a tax-shielded bond affects their economic well-being. Typically, individual preferences for a public policy differ depending on the impact of the policy on each individual’s (economic) well-being. Indeed, even policies that unambiguously enhance economic efficiency may have redistributive effects. How are these individual preferences converted into a public policy? According to the political economic literature, there are several political institutions through which individual preferences are converted into aggregate (or societal) preferences, and eventually into public policies. In this chapter, we shall concentrate on elections, although several others political mechanisms have been studied in the literature. Individuals convey their preferences in elections, either directly through referenda (e.g., in Switzerland and in California), or by appointing their political representatives. Public policies will depend on the outcome of the election, and will coincide with the winning proposal, in the case of referenda, or with the policy decided by the winning candidate in a representative electoral system. Once public policies are implemented, economic agents react by adjusting their economic behavior, and hence the market equilibrium may change. As displayed in Figure 1, this behavior closes the circle from economic markets to political institutions and back to economic markets. Thus, the novelty of the political economic approach is to consider every individual in a double role as an economic and political agent. As economic agents, individuals take consumption, labor, saving – that is, economic – decisions, while considering the public policy as exogenously given.

1

Clearly, this policy may induce other individuals to move to this region to cash in on the transfer.

1.

Introduction to Political Economics

5

Figure 1: The political economic approach

As political agents, they express their preferences to determine those public policies, which they had regarded as exogenous given, when they took their economic decisions. The economic behavior in this two-sided analysis is displayed on the left of Figure 2. In choosing their optimal economic actions, agents understand that their individual action does not modify the existing economic and political situation, which has to be taken as given. Hence, in their economic behavior, tax payers do not try to change the tax system. Instead, they try to take advantage of any loophole to reduce their tax bill. Analogously, middle-aged individuals do not question, or try to amend, the design of the pension system. At most, they attempt to calculate when it is optimal – from their individual viewpoint – to retire. They calculate whether it is convenient to work one more year to increase their pension benefit, or to retire and enjoy a generous early retirement pension as well as leisure time. All these individual economic decisions, which create the demand and supply, determine the equilibrium prices and quantities in economic markets for exogenously given public policies. The individual behavior of the tax payers will thus determine the total amount of fiscal revenues. The interaction of the labor demand – by each single firm – and of the labor supply – by each worker –determines the equilibrium wage and employment rate, given the labor market policies, such as the level and duration of unemployment benefit, the degree of employment protective legislation, the tax rate on labor income and the payroll taxes levied to finance the welfare state. The political behavior in this two-sided analysis is displayed on the right of Figure 2. As political agents, individuals express their preferences over public policies. Individuals’ preferences depend on how the policy affects their utility or well-being.

6

POLITICAL ECONOMICS

Figure 2: Politico-economic equilibria

Low income workers, who are the net winners from redistributive policies, clearly express their support for this type of policies, whereas high earners oppose them. Individuals’ preferences over public policy are characterized by the indirect utility function. This indirect utility function is obtained in two steps. In the first stage, corresponding to the left side of the analysis, an individual optimizes with respect to the economic variables. The utility function is therefore maximized with respect to the economic variables (such as savings and labor supply), and the optimal values of these economic variables are obtained as a function of the public policy. In the second stage, the indirect utility function is obtained by substituting these optimal values back into the original utility function. The indirect utility function thus depends on the public policy only. Its interpretation is straightforward. The indirect utility function expresses the individual preferences over public policy when the economic decisions at the individual level have already been taken. Hence, in expressing their views over the public policy in the political arena (e.g. in elections or through lobbying), individuals will first determine their most preferred public policy by maximizing their indirect utility function with respect to this policy. There is a clear asymmetry in the economic and political behavior of individuals. In their economic decisions, agents understand that their individual action will not modify the existing economic and political scenario, which is therefore taken as given. This is because each individual is too small – too atomistic – to affect the economic aggregates. But in the political arena, individuals can be more powerful. If an individual is pivotal in an election – that is, he is among the few voters who determine the outcome– he may expect his most preferred public policy to be adopted by politicians, who seek to secure his pivotal vote to win the election. In this situation, the pivotal voter will act as a

1.

Introduction to Political Economics

7

“monopolist” of the public policy. In choosing his most preferred policy, he will consider that, if his policy is chosen and implemented by the politicians, it will modify the economic and political context for everyone. For instance, if a low income worker is the pivotal voter in an election, he may expect politicians to implement a highly redistributive policy. This policy will affect all workers, including those with a high income, and the whole economy, as a distortion due to taxation will emerge. After markets have aggregated individual economic decisions, and political institutions have aggregated individual preferences over public policies, a complete description of the economic and political scenario emerges. It is this combination of economic choices and political decisions over public policy that represents the distinctive mark of the political economic approach. 1.2 Political Institutions The political behavior illustrated in the right part of Figure 2 depends on the behavior of political institutions. Suppose that, after individual preferences over a public policy are formed, all political institutions generate the same policy outcome. In other words, suppose that – given individuals’ preferences – the public policy would be invariant to the type of political institution used to aggregate these preferences. As such, the political institution would be neutral. It would have no impact on the process of preferences aggregation and ultimately in determining the public policy. We would not need to examine political institutions, as they would play no role. Yet, things are not so simple. Political institutions do play a role. In fact, – for a given set of individuals’ preferences – the political institution used to aggregate these preferences shapes policy outcomes. This is an old result. In 1951, in his famous ‘Impossibility Theorem’, Ken Arrow showed that political institutions are not neutral, as no desirable political mechanism is able to aggregate individual preferences consistently. More specifically, Arrow’s ‘Impossibility Theorem’ stated that there is no democratic mechanism that allows individual preferences to be aggregated in a consistent way, that is, so that the properties of (i) Rationality; (ii) Unrestricted Domain; (iii) Weak Pareto Optimality; and (iv) Independence are satisfied. The first property – rationality – requires aggregate preferences to be complete and transitive. The political mechanism has to able to compare and rank all possible outcomes of the public policy; this raking has to be transitive so that if A is preferred to B, and B is preferred to C, then A has to be preferred to C. The second property demands that the political institution can accommodate any individual preference. Hence, for any configuration of individual

8

POLITICAL ECONOMICS

preferences over a policy – however irrational they may seem – the mechanism has to be able to produce an aggregate decision: a policy outcome. The third requirement is highly intuitive: if every individual weakly prefers A to B – that is, if individuals either prefer A to B, or they are indifferent between A and B, and no individual prefers B to A – then the mechanism has to rank A over B. Finally, the fourth property requires the mechanism to concentrate on the issues at stake. If the decision is between alternatives A and B, the individual ranking (or preference) over a third alternative C should not matter in determining the policy outcome between A and B. For example, the ranking between social security and health spending should not depend on how individuals rank these two expenses relatively to a third alternative – say, defense spending. As we shall see in the next section, to comply with the ‘Impossibility Theorem’, the political economic literature typically drops the unrestricted domain feature, as individual preferences are often required to be single-peaked. As an example of this theorem, we examine the policy outcome under different – albeit relatively similar – political institutions, for a given set of individual preferences over a public policy. The aim of using this simple example is to show that, for a given initial set of individual preferences, different policy outcomes can arise, depending on the political institution that aggregates individual preferences. We concentrate on a simple class of political institutions: elections. We exploit different timing of votes and types of vote-counting in elections. We begin with a simple majority voting election. We then compare its result – in terms of policy outcome – with an agenda setting election, in which the sequence of voting over different alternatives is a factor, and with a “Borda” election, in which each individual can vote for more candidates, assigning a different number of votes to each one. This allows individuals to express the intensity of their preferences. In this chapter, we assume that individuals vote sincerely, according to their true preferences over the public policy. 2 In our policy example, we consider 7 voters – we define them as voter 1, 2, … and 7 – and 4 alternatives, characterized by A, B, C and D. The voters could be seven ministers in a committee – each member with one vote. The alternatives may be the level of public spending in education or health care (and hence the tax revenue needed to finance public spending). In this case, A may represent no public spending, B low spending, C medium spending and D high spending. The ministers’ individual preferences are displayed in Table 1; the alternative at the top represents the most preferred policy, while the alternative at the bottom the least preferred. In close, small elections, individuals may decide to vote strategically. For instance, they may choose to vote for their second best outcome, if they realize that their first best outcome will not gain enough votes to win, and they could end up with their third best option.

2

1.

Introduction to Political Economics

9

Alternatives

Table 1: Individual preferences

Agents

1

2

3

4

5

6

7

best

A

A

A

B

B

C

C

B

B

B

C

C

D

D

C

C

C

A

D

A

A

D

D

D

D

A

B

B

worst

1.2.1 Simple Majority Voting Let us begin with a simple majority voting election. Every individual indicates his most preferred level of public spending – whether A, B, C or D – and the alternative receiving the most votes, that is, a simple majority of votes, becomes the policy outcome. In this case, ministers 1 to 3 prefer alternative A (no spending); ministers 4 and 5 vote for B (low spending); while ministers 6 and 7 vote for policy C (medium spending). None of the voters favors alternative D (high public spending). Accordingly, the policy outcome of this simple majority voting election is the alternative A, which receives three votes. Hence, there will be no public spending. In this case, the strong preferences of the first three ministers in favor of alternative A are sufficient for this policy to be adopted, despite opposition from the other ministers, who prefer at least two of the other alternatives to the winning policy (A). 1.2.2 Agenda Setting Consider now a political institution that utilizes a different type of election to determine public policy. The seven ministers are still the voters; however, the aggregation mechanism for preferences now includes agenda setting, so that alternatives are voted in a pair-wise comparison in a pre-established order. This voting sequence is determined by the following agenda: • • •

AGENDA I AGENDA II AGENDA III

A vs B - vs C - vs D D vs C - vs B -vs A A vs C - vs B - vs D

In agenda I, the pair-wise voting at the first stage is between alternatives A and B. A prevails five votes to two, thanks to the votes of ministers 1, 2, 3, 6 and 7. In the second round, policy A is matched with alternative C. Alternative C

10

POLITICAL ECONOMICS

wins by four votes (by ministers 4 to 7,) to three. In the third and last stage, alternative C competes against D and wins, as all ministers prefer C to D. Hence, alternative C (medium public spending) is the winner in this agenda setting election. In agenda II, the voting sequence changes. At the first stage, the pairwise voting is between alternatives D and C and C clearly prevails. In the second round, C is challenged by policy B, which wins by five votes (ministers 1 to 5) to two (ministers 6 and 7). In the final stage, alternative B is defeated by policy A by five votes to two. Hence, alternative A (low public spending) is the policy outcome in the election with agenda II. Finally, in agenda III alternative C prevails over alternative A in the first round by four votes to three, but is defeated by alternative B in the second round by five votes to two. In the third stage, alternative B wins over policy D by five votes to two, and becomes the policy outcome of this election with agenda III. This simple example shows that – in an election with agenda setting – the policy outcome hinges crucially on the role played by the agenda setter. Indeed, depending on the voting sequence selected by the agenda setter, one of three alternatives (A, B and C) can emerge as the policy equilibrium outcome. 1.2.3 Borda Voting The last type of political institution we consider is called ‘Borda’ voting. Ministers can still express their preferences through an election, in which the alternatives are simultaneously voted upon. The peculiarity of this preferences aggregation mechanism, in relation to a simple majority voting election is that voters can express more than one preference, thereby conveying the intensity of their preferences. Consider a situation in which each of our ministers can give two votes to one alternative and one vote to another alternative. Clearly, each minister will give two votes to his most preferred policy and one vote to the second preference. If we sum up these preferences, alternative A will receive six votes – as the first choice for ministers 1 to 3 – alternative B will receive 7 votes – determined by two first places (ministers 4 and 5) and three second places (ministers 1 to 3) alternative C will gain six votes – two first choices (ministers 6 and 7) and two second places (ministers 4 and 5) – while alternative D will only receive two votes, as the second choice of ministers 6 and 7. In this Borda counting election, alternative B (low public spending) would be the adopted policy.

1.

Introduction to Political Economics

11

Figure 3: Single peakedness

Further examples 3 can be given to show that – for given a set of initial individual preferences over public policies represented in Table 1 – the aggregation of votes though different political institutions can lead to different policy outcomes. It is now convenient to examine the characteristics of voters’ individual preferences – our ministers – which lead to different policy outcomes across these different types of elections. Figure 3 displays the individual preferences of our seven voters over the four alternatives (A, B, C and D). A quick look at these preferences suggests that they are all single-peaked, except ministers 6 and 7; in other words, they have a single maximum. Voters 1 to 3, whose preferences are characterized by the continuous line, prefer no public spending and their well-being drops as spending increases. Minister 4 has an interior maximum for a low level of public spending; no spending or a medium level decreases his utility. Minister 5’s preferences are also of this type: they peak for a low level of public spending, although – as a third choice – he prefers more spending (D) than minister 4 (who indicates A). Instead, the preferences of Ministers 6 and 7 are not singlepeaked. In fact, their most preferred choice is to have a medium level of public spending or – as a second alternative – high spending. As a third alternative, however, they prefer no public spending to low spending. In terms of their preferences, this creates an additional peak (or local maximum) at alternative A (no spending). While not single-peaked, these preferences are far from being irrational. In our example, they may be associated with an individual who prefers a medium (or high) level of public spending (i.e. high quality of healthcare). However, if the level is low, he prefers to use private healthcare and thus to have zero public spending.

3 For instance, we could consider another Borda voting election, in which voters may attribute three votes to an alternative, two votes to another alternative and one vote to a third alternative.

12

POLITICAL ECONOMICS

This lack of single-peakedness is what determines the different outcomes, depending on the electoral system. As our example suggests, individual preferences have to be smooth in order to be single peaked. By moving away for his ideal alternative, the voter has to prefer alternatives that are closer to the ideal point over alternatives that are further apart from the ideal point. When individual preferences are not single-peaked, aggregate preferences may end up not to be transitive. As shown in the agenda setting voting system, aggregate preferences are not transitive in Table 1. In fact, A is preferred to B, and B is preferred to C, but A is not preferred to C. In what follows, we shall restrict individual preferences to being singlepeaked and will hence drop the second feature: the unrestricted domain in Arrow’s ‘Impossibility Theorem’. Fortunately, in most economic situations, individual preferences are single peaked (i.e. economic agents have smooth preferences), as shown in the following example. Yet, not always. 1.3 A simple Example: Indirect Utility Function, Single-Peaked Preferences and the Bliss Point Consider an economy populated by three groups of individuals. They consume a private good (C) and a public good (G). Their preferences over these two goods are the following: 𝑈𝑈𝑖𝑖 (𝐶𝐶, 𝐺𝐺) = 𝐶𝐶 + 𝛾𝛾𝑖𝑖 ln(𝐺𝐺)

(1)

𝑉𝑉(𝐺𝐺) = [𝑌𝑌𝑖𝑖 − 𝑇𝑇] + 𝛾𝛾𝑖𝑖 𝑙𝑙𝑙𝑙𝑙𝑙 = 𝑌𝑌𝑖𝑖 – 𝐺𝐺 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 ) 𝑙𝑙𝑙𝑙(𝐺𝐺)

(2)

where the parameter 𝛾𝛾𝑖𝑖 of the utility function depends on the agent’s income: 𝛾𝛾𝑖𝑖 = 𝐴𝐴 − 𝑌𝑌𝑖𝑖 . Agents differ according to their income, 𝑌𝑌𝑖𝑖 . They can be Poor (P), Middle Income (M) and Rich (R). In particular, their income is respectively 𝑌𝑌𝑃𝑃 = 1�2, 𝑌𝑌𝑀𝑀 = 2�3 and 𝑌𝑌𝑅𝑅 = 1 and their proportions in the population are α𝑃𝑃 = 45%, α𝑀𝑀 = 30%, and α𝑅𝑅 = 25%. The public good is financed with a lump-sum tax (T), so that the government budget constraint is 𝐺𝐺 = 𝑇𝑇. Finally, assume that 𝐴𝐴 = 1. In this simple case, individuals take no economic decision. They consume all their net income and the public good, G, received from the public sector. It is thus easy to obtain the indirect utility function of an agent i. Using the utility function of an individual i, where i represents his income type at equation (1), and substituting in the individual’s and the government’s budget constraints, respectively, 𝐶𝐶 = 𝑌𝑌𝑖𝑖 – 𝑇𝑇, and G = T, and 𝛾𝛾𝑖𝑖 = 𝐴𝐴 − 𝑌𝑌𝑖𝑖 , we obtain the following indirect utility function:

1.

Introduction to Political Economics

13

Figure 4: Individual preferences

Clearly, this indirect utility function depends only on the public policy, which in this example is summarized by the public good, G, or, analogously, by the lump-sum tax, T, given the government’s budget constraint, G=T. Individual preferences have graphical interpretations. If we consider the preferences over the consumption of private good, C, and public good, G, they are represented by the utility function, 𝑈𝑈𝑖𝑖 (𝐶𝐶, 𝐺𝐺). The indifference curves between C and G associated with this utility function are shown in Figure 4, while the straight line represents the individual budget constraint: 𝐶𝐶 = 𝑌𝑌𝑖𝑖 – 𝑇𝑇. The preferences over the level of public good, as measured by the indirect utility function, V(G), are displayed in Figure 5. Clearly, Figure 5 can be constructed from the data in Figure 4. For each value of G, the budget constraint determines the maximum amount of private consumption available, such that 𝐶𝐶 = 𝑌𝑌𝑖𝑖 – 𝐺𝐺. The indifference curve passing for this point (𝐶𝐶, 𝐺𝐺) defines the utility level associated with G. This information is reported in Figure 5, where for every value of G the indirect utility function, V(G), measures the corresponding utility value. As shown in Figure 5, the indirect utility function is single-peaked and has a maximum at G*. To guarantee that preferences over G are single-peaked, we need to show that V(G) is a concave function of G. This is done by checking that the second derivate of V(G) with respect to G is negative: 𝑉𝑉”(𝐺𝐺) = 𝑑𝑑 2 𝑉𝑉(𝐺𝐺)/𝑑𝑑𝐺𝐺 2 < 0. Since 𝑉𝑉(𝐺𝐺) = 𝑌𝑌𝑖𝑖 – 𝐺𝐺 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 ) 𝑙𝑙𝑙𝑙(𝐺𝐺), we have that: 𝑉𝑉’(𝐺𝐺) = 𝑑𝑑𝑑𝑑(𝐺𝐺)/𝑑𝑑𝑑𝑑 = −1 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 )/𝐺𝐺

(3)

14

POLITICAL ECONOMICS

and thus: 𝑉𝑉”(𝐺𝐺) = 𝑑𝑑2 𝑉𝑉(𝐺𝐺)/𝑑𝑑𝐺𝐺 2 = −(𝐴𝐴 − 𝑌𝑌𝑖𝑖 )/ 𝐺𝐺 2 < 0.

(4)

A useful concept to summarize (single-peaked) individual preferences is the bliss point: it represents the policy level at which individual preferences are maximized. To calculate the bliss point for an agent i, we simply maximize the individual indirect utility function with respect to the level of public good, G. max 𝑉𝑉(𝐺𝐺) = max 𝑌𝑌𝑖𝑖 – 𝐺𝐺 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 ) 𝑙𝑙𝑙𝑙(𝐺𝐺)

(5)

𝐴𝐴−𝑌𝑌𝑖𝑖 𝐺𝐺

(6)

{𝐺𝐺}

{𝐺𝐺}

The first order condition (FOC) becomes: −1 +

=0

and hence, the bliss point of a type-i individual is equal to: 𝐺𝐺𝑖𝑖∗ = 𝐴𝐴 − 𝑌𝑌𝑖𝑖

(7)

∗ = 1 − ⅔ = 1⁄3 and 𝐺𝐺𝑅𝑅∗ = 1 − 1 = 0. Hence, 𝐺𝐺𝑃𝑃∗ = 1 − ½ = ½, 𝐺𝐺𝑀𝑀

Figure 5: Indirect utility function

2. Electoral Models

The most common political events are undoubtedly elections. During elections, voters may choose directly public policy - as in referenda - or they may select their political representative for legislative bodies. In the latter case, individuals will base their choice – to some extent, if not exclusively – on the policy platform presented by the candidates. As such, they effectively provide their elected representatives with a mandate to adopt the political program they presented during the election campaign. What arises from an election is thus an incomplete contract between a majority of the electorate and its political representatives, in which the politicians govern based (also) on the interests of the majority of the electorate that voted for them. We shall discuss these political dynamics shortly when we introduce the concept of median voter to indicate the pivotal voter in these simple majority elections. However, voters may base their political decisions on other, additional factors. Besides their preferences over policy, they may feel closer to a particular candidate because of ideological reasons or just due to personal sympathy. Instead, candidates identify the section(s) of the electorate, (i.e. groups of voters) that they try to reach (and capture) by targeting their public policies. As we shall discuss later, the probabilistic voting model, used by Dixit and Londregan (1996) among others, analyzes this type of environment and suggests that political candidates should concentrate more on the ideologically neutral sector of the electorate, as these individuals are easier to convince through economic policy, and hence to “swing” their vote towards them. So far, we assumed that, once elected, political candidates keep the promises made during the political campaign and implement the policies publicized. Whether this is the case depends on the motivations of the politicians and on their career concerns. Politicians may be opportunistic and thus only care about being elected; or partisan, where they care about the policy being implemented. For opportunistic politicians, they offer a policy platform during the election campaign that maximizes their chances of winning the election.

16

POLITICAL ECONOMICS

Being in office may either provide these individuals with an ego-rent (e.g. being powerful, well known and often on television) or with an actual rents from bribes or legal contributions. Once in power, there are no incentives to change the policy announced during their electoral campaign, unless a new policy may increase their future reelection probability. Partisan politicians, on the other hand, may have an incentive to renegade on the platform presented at the election, in order to move closer to their most preferred policy. Thus, if these politicians’ true preferences over policies are known to voters, politicians campaigning on policies different from their own preferred policy will lack credibility. To address these issues, a strand of literature on elections has taken a deeper look at the entire political process, from the initial candidacy of a citizen to political position or office, to the election campaign and finally to the role of an elected representative as a policy-maker. These models, mainly due to Besley and Coate (1997) and Osborne and Slivinski (1996), suggest that these citizen candidates are motivated to become politicians and thus to run for elections by their specific preferences for public policy. If they believe that no other politician with a policy platform sufficiently close to theirs will stand for election, they have an incentive to enter the political arena and to participate as candidates to the elections, in order to have a chance to win and thus to implement their most preferred policy. These models may rationalize the decision of several non-professional-politician candidates, such as businessmen, sportsmen or actors, to enter politics. Another approach in this political economics literature has analyzed legislative, rather than electoral, models by concentrating on post-electoral politics. The main idea of these models is that – once politicians are elected – political decisions will largely be affected by the institutional constraints that characterize the decision making process. Barron and Ferejohn (1989) examine bargaining in agenda setting; while Shepsle (1979) considers the restrictions imposed on a multi-issue policy outcomes by the allocation of the decision process in policy jurisdictions. Finally, another strand of literature has stressed the role of lobbying in the decision-making process relating to economic policy. These lobbying models, based on research by Becker (1983) and Grossman and Helpman (1994) among others, suggest that besides electoral incentives and legislative restrictions, lobbying activities – consisting of financial contributions during election campaigns, pressure during the legislative process and even bribes – may strongly affect policy-makers’ decisions. We will return to these models on lobbying in the next chapter.

2.

Electoral Models

17

2.1 The Median Voter Model In this section, we introduce the first model of political aggregation discussed above, which uses a simple majority voting election. It is convenient to begin by considering a case of direct democracy – a referendum – in which voters directly determine a one-dimensional policy, such as the level of public spending (as in the example in chapter 1), the size of the welfare state (see chapter 5), or the degree of flexibility in the labor market (see chapter 6). These models assume commitment to the policy: once the policy is decided during the election campaign, it will be implemented. Individuals’ preferences are assumed to be single-peaked over the economic policy at stake, in order to comply with the requirement for the aggregate preferences to be rational (that is, complete and transitive), as discussed in chapter 1. As we discussed in the example at the end of chapter 1, single-peakedness is typically guaranteed in several types of economic decisions. This simple peak - or maximum - in the individuals’ preferences is denoted the bliss point, and represents the individuals most preferred economic policy. Finally, let us denote as the median voter, the voter whose bliss point divides the distribution of bliss points in the electorate in half: 50 percent to the right and 50 percent to the left. Clearly, this suggests that 50 percent of the voters would prefer to have a higher level of the policy, for instance more public spending, than the median voter; while 50 percent of the voters would prefer to have a lower level, for instance less public spending, than the median voter. One of the most powerful and widely used results of the median voter theorem is to identify the decisive voter in an election. The preference of the median voter coincides with the electorate’s overall decision. More precisely, the median voter’s theorem states that if the public policy is one-dimensional, and preferences are single-peaked over the public policy, the median voter’s bliss point represents the political equilibrium outcome in a simple majority voting election. To understand this result, notice that due to single-peakedness, individual preferences are smooth. In fact, as an individual moves away from the bliss point, his utility decreases. For instance, if an individual compares two points (i.e. policy levels) to the right of his bliss point (𝑥𝑥𝑏𝑏 < 𝑥𝑥’ < 𝑥𝑥’’, where 𝑥𝑥𝑏𝑏 is the bliss point), then the utility of the furthest point from the bliss point is lower than the utility of the point closer to the bliss point: 𝑈𝑈(𝑥𝑥𝑏𝑏 ) > 𝑈𝑈(𝑥𝑥’) > 𝑈𝑈(𝑥𝑥’’).

18

POLITICAL ECONOMICS

Figure 1: Median Voter’s Theorem

Now consider Figure 1: there are three voters, and the median clearly corresponds to voter B, whose bliss point is X*. According to the median voter’s theorem, X* is the policy which emerges in a simple majority voting over X. To see why X* prevails in a simple majority voting over any other alternative X, let us analyze a point to the left of X*, such as point 𝑥𝑥’. Clearly voters B and C – indeed any voter to the right of the median voters (i.e. whose bliss point is larger than the median voter’s) will prefer X* to 𝑥𝑥’, while voter A will prefer 𝑥𝑥’ to X*. Analogously, if we take any point to the right of X*, such as xc, two voters will still prefer X* to 𝑥𝑥𝑐𝑐 . In this case, voters A and B – and indeed any voter to the left of the median voters (i.e. whose bliss point is lower than the median voter’s) will prefer X* to 𝑥𝑥𝑐𝑐 , while voter C will obviously prefer 𝑥𝑥𝑐𝑐 . To summarize, 50 percent of the voters – those to the right of the median voter – prefer the median voter’s bliss point to any lower level; whereas the remaining 50 percent – those to the left of the median voter – prefers the median voter’s bliss point to any higher level. Hence, the median voter’s bliss point represents the equilibrium policy outcome of this simple majority voting election, since there will always be a majority of the electorate that prefers this bliss point to any other alternative. The median voter’s theorem may also be applied to a different type of majority voting election, where voters do not determine a specific public policy, as in a referendum, but select a political candidate to hold office.

2.

Electoral Models

19

Figure 2: Distribution of voters’ policy preferences

Consider a majority voting election for political candidates or parties, in which there are two opportunistic candidates (or parties or coalitions) running for office. The candidates’ decision consists of selecting a political platform, which may focus on a specific public policy (e.g. for promoting economic growth), or a broader agenda which fits along a traditional left-to-right scale. Clearly, individual voters have preferences over this political platform. In fact, we assume that there exists a distribution of voters’ preferences over the economic policy, displayed in Figure 2, which indicates the mass of voters who have different positions (or bliss points) over a traditional left-to-right scale. In order to win elections, each candidate has to position his policy platform on this left-to-right political spectrum in order to maximize votes, and thus the probability of winning the election, taking as given his opponent’s policy platform decision. Suppose that candidate A positioned his platform in Pa, what should candidate B do? Assume, as shown in Figure 2 that the distribution of policy over the left-to-right scale is skewed to the left. Then candidate B should locate his platform to the left of candidate A in order to gain the support of the voters who have the preferred policy to the left of Pa. This strategy would be sufficient to win the election, since they constitute a majority of the electorate. For instance, candidate B can choose Pb and win the election by obtaining the votes of more than 50 percent of the electorate. In this situation, candidate A will clearly have an incentive to reposition the platform to the left of candidate B in order to gain more than 50 percent of the electorate, and hence to win the election. The hypothetical process of re-positioning by the two candidates will finally stop when both candidates locate their platform at Pm, which represents the most preferred platform of the median voter. If candidate A is at Pm, candidate B will have no incentive to move from Pm, as he would certainly lose the election; and analogously for candidate B. Hence, both can-

20

POLITICAL ECONOMICS

didates would converge to Pm. 1. The most popular platform (or economic policy) would thus be Pm, but the identity of the winning candidate would be uncertain, since voters are indifferent between two candidates that propose identical platforms. This median voter model is thus robust enough to cope with different variations, and provides powerful empirical implications. First, it suggests that elections should lead to political (or policy) moderation, as both candidates or parties have a strong incentive to move towards the center i.e. towards moderate positions, while abandoning extreme positions. Second, it implies that both parties will campaign on the same (or at least on a similar) policy platform, thereby refrain from differentiating themselves. Despite its wide use in the political economy literature, this model has also some limitations. In particular, its application is restricted to elections contested between only two candidates or parties. This is typically the case in majoritarian electoral systems, such as in the United States. In fact, the predictions of the median voter model have been in line with the outcome of several elections – such as the close run US Presidential election in 2000, between Bush and Gore. However, its results – and intuition – do not extend to elections with more than two parties. Hence, elections in proportional systems, in which several parties run for elections, cannot be analyzed with this model. Additionally, the median voter theorem does not apply to elections in which the policy platform is considered to be multidimensional. In fact, if parties have to campaign on more than one specific issue, such as the typical leftto-right scale, no Nash equilibrium of this voting game will emerge in pure strategies. When campaigning on more than one issue (for instance, on two issues, such as public spending and foreign policy) parties will have an incentive to continuously modify their platforms to respond to their opponent. In the end, no platform will emerge as a political equilibrium outcome, and Condorcet cycles will emerge. Fortunately, many political scientists argue that the electoral platforms can indeed be reduced to a left-to-right scale. A final critique of the median voter model pertains to individuals’ voting choice. In this model, individuals have to vote for one party or the other. In reality, however, individuals can choose to abstain. Indeed many do so and purposely choose not to vote for any party. Introducing this abstention decision into the median voter model would modify its main results, as new tradeoffs may emerge. Consider for instance a right wing party, which has to decide whether to move further to the center in order to contend votes to the left wing party, or to remain on the right of the political spectrum. In the model without abstention described above, the party would be better off moving to the center – and in particular to the median voter’s bliss point. But if, by moving to the Stated differently, for both candidates, choosing the platform 𝑃𝑃𝑚𝑚 represents a Nash equilibrium of the voting game.

1

2.

Electoral Models

21

center, the party lost the votes of the more right wing voters, who prefer to abstain, a new trade-off would arise and the party may find it convenient to remain to the right of the median voter. This new factor could explain some of the polarization that we have observed in the latest elections, and also the decision by each party to target its own political base. 2.2 Probabilistic Voting Model In this section, we examine the outcome of a simple majority voting election in which voters base their political decision not only on their preferences over policies, but also on additional factors. In particular, we consider a simple majority voting election where the electorate votes into office a political candidate or a party. Unlike in the median voter’s model described above, however, the electorate does not exclusively consider the two candidates’ policy platform in their voting decisions. Their vote is also affected by the ideology of the voters, or by their general sympathy towards the candidates. 2.2.1 The candidates We consider an election with two opportunistic candidates (or parties) running for office. These candidates – call them A and B – have no preferences over the types of policies implemented after the election. In other words, they are not partisan. They are opportunistic and only seek to maximize the probability of being elected, perhaps because, if in office, they receive some private rents, such as ego-rents, bribes or contributions. During their election campaign, candidates have to select a policy platform, which identifies their position on a range of issues. For simplicity, we assume the platform is two-dimensional. Therefore, candidates have to express their views over two issues – call them X and Y – which may represent, for instance, the size of the welfare state and the type of foreign policy. Both candidates determine their policy platform simultaneously, yet independently. These platforms are represented by (XA,YA) and (XA,YA), respectively for candidates A and B. There is commitment over the platform: Once elected, a candidate is compelled to implement the platform they campaigned on. 2.2.2 The voters Voters in this model are quite sophisticated. They base their choice of the political candidate on an objective factor and on subjective perceptions. The former element in their voting decision is the policy component. Voters evaluate how each candidate’s policy platforms affects their indirect utility function (see chapter 1). The latter factors are instead more subjective. They capture the

22

POLITICAL ECONOMICS

ideological view of the voters, their idiosyncratic sympathy towards the candidates, and the average popularity of the candidates among the entire electorate. To make the election outcome interesting, we assume that in our society voters differ in their economic situation and in their ideology. The heterogeneity in the voters’ behavior will thus stem from the differences in their economic and ideological background. In particular, individuals are partitioned into three income groups. They can be poor (P), middle income (M) and rich (R), and their income is respectively YP, YM and YR, with YP < YM < YR. Moreover, the share of the total population in each group is respectively αP, αM and αR, with 𝛼𝛼𝑃𝑃 + 𝛼𝛼𝑀𝑀 + 𝛼𝛼𝑅𝑅 = 1. The potential policies of the two candidates will have the same effect on all individuals within the same group. This effect can be measured by the indirect utility functions, where UJ(XA,YA) and UJ(XB,YB) represent the indirect utility function induced respectively by the policy platforms of candidate A and B for an individual in group J. Clearly, these utilities differ if candidates choose different platforms. Furthermore, the same platform will have a different impact on individuals belonging to different social-economic groups. For instance, a platform involving a large welfare state will typically increase the well-being of poorer voters, while decreasing the welfare of the wealthy. Within each income group, individuals differ in their ideology or sympathy towards the two candidates. In particular, σiJ measures the ideology of a voter i in group J. If σ𝑖𝑖𝑖𝑖 > 0, voter i in group J is ideologically closer to candidate B, whereas if σ𝑖𝑖𝑖𝑖 < 0 the voter is ideologically closer to candidate A. Clearly, if σ𝑖𝑖𝑖𝑖 = 0, the voter is ideologically neutral, as he does not prefer either of the two candidates. Since voters have different ideologies within each income group, there will be voters close to candidate A (or B) among the poor, as well as among the middle-income and the wealthy. How are voters distributed according to their ideology within each group? Figure 3: Distribution of Ideology

2.

Electoral Models

23

For instance, how many voters are ideologically close to candidate A or B among the poor? To simplify the analysis, we assume that individuals are distributed according to a uniform distribution function, centered on an average ideology equal to zero, with a group-specific density, ΦJ. In other words, within each income group – say, among the poor – there are as many individuals who strongly prefer candidate A, as voters who are neutral or who have a slight preference for candidate B. However, the distribution of ideology may differ across income groups. For instance, middleincome voters may have a higher density than the other groups, Φ𝑀𝑀 > Φ𝑃𝑃 and Φ𝑀𝑀 > Φ𝑅𝑅 . In a uniform distribution function, the density, ΦJ, represents an important summary statistic to characterize the shape of the distribution of ideology. In particular, as the density, ΦJ, increases, the distribution function becomes more concentrated around the average (which we assumed to be equal to zero), and hence the group becomes less ideological, as fewer voters have a strong ideology or sympathy towards a candidate, while many voters are neutral, or almost neutral. Individuals’ voting decision is also affected by an additional subjective element: the candidates’ average popularity before the election. This average popularity is common to all voters, and measures the “mood” of the electorate as the election approaches. We measure this average popularity with δ. If δ > 0, candidate B is more popular than candidate A and vice versa for δ < 0. Clearly, if δ = 0, the candidates are equally popular (or unpopular). Typically, the average popularity before the election is determined by the outbreak of a scandal or news that may affect the whole electorate’s subjective perception of the candidates. If a negative shock hits candidate A - like in the 1990s when the media reported US President Bill Clinton’s affair with Monica Lewinsky - the candidate’s popularity will plunge, while candidate B’s popularity will rise, and hence δ > 0. Figure 4: Distribution of Popularity

24

POLITICAL ECONOMICS

Candidates cannot control their average popularity before the election, as they cannot entirely control the outbreak of a scandal or negative publicity. However, they are aware of the fact that scandals or breaking news – in their favor or against them – may occur. The average popularity – or the likelihood of a scandal occurring – is assumed to be distributed according to a uniform density function, centered around zero, and with density Ψ, as displayed in Figure 4. In other words, we assume that all scandals that affect candidate A or B are as likely to occur as no scandal at all, δ = 0, in which case both candidates share the same popularity. To summarize, voters consider three elements before deciding whom to vote for: 1. Policy: the utility induced by the candidates policy platform, UJ(XA,YA) and UJ(XB,YB). This factor is group specific. 2. Individual ideology: σiJ. 3. Average popularity: δ. Therefore, voter i in group J will vote for candidate B if the sum of the utility associated with the platform proposed by this candidate, his individual ideology (recall that σiJ can be positive or negative, with σ𝑖𝑖𝑖𝑖 > 0 indicating an ideology closer to candidate B), and the average popularity (in turn, δ can be positive or negative, with δ > 0 indicating an average popularity in favor of candidate B) is larger than the utility associated with the platform launched by candidate A: 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 ) + σ𝑖𝑖𝑖𝑖 + δ > 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 )

(1)

2.2.3 The timing of the game The process from the initial formation of individual preferences over policy outcomes to its final implementation encompasses four different stages: 1. Electoral Campaign. At this stage, candidates announce – independently and simultaneously – their policy platform (XA,YA) and (XB,YB). In taking this decision, they know the distribution of individuals’ ideology, and hence how many voters in each group are ideologically closer to candidate A or B. However, they do not know their average popularity, which is determined shortly before the elections, although they do know how likely they are to be caught in a scandal. 2. Before the election, the average popularity of the candidates, δ, is determined (i.e. a scandal may occur).

2.

Electoral Models

25

3. Election. Voters choose their favorite candidate, A or B. 4. Policy. After the election, the winner implements his announced policy platform. 2.2.4 The “Swing” voter Before analyzing how candidates’ decide on a political platform, it is convenient to introduce the concept of “Swing voter”. The swing voter is the voter who – after considering the policy platforms, his own ideology and the average popularity – is indifferent between voting for candidate A or B. Consider a swing voter in group J. He is indifferent between candidate A or B since he would receive the same utility from voting for either of the candidates. Thus, if we identify the swing voter in group J with σJ, we have that:

σ𝐽𝐽 = 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 ) − δ

(2)

Hence, σJ indicates the ideology of the swing voter in group J. Since we assumed a uniform distribution of ideology within a group, the mass of swing voters in each group will depend on the group density: the higher the density, the higher the number of swing voters. Why is this “swing” voter relevant? As we will show below, candidates’ platforms are designed to target the swing voter. This is because, by being indifferent between the two candidates , the swing voter is easier to convince – that is, to swing – for politicians. A small change in the policy platform of one of the candidates with the aim of increasing the swing voter’s utility is sufficient to gain his vote. On the contrary, a voter, who has a strong ideological bias in favor of candidate A, will be more difficult to convince. Candidate B would have to allocate many more resources to this voter’s group in his policy platform in order to convince this reluctant voter. In Figure 5, we show that the swing voter divides the overall number of voters into two groups. As the swing voter is indifferent between voting for candidate A or B, the voters to the left of the swing voter, σ𝑖𝑖𝑖𝑖 < σ𝐽𝐽 , who are ideologically closer to A, will vote for candidate A; while those voters to the right of the swing voter, σ𝑖𝑖𝑖𝑖 > σ𝐽𝐽 , who are ideologically closer to B, will vote for candidate B. Finally, notice that the identity of the swing voter is unknown to the candidates when they set out their policy platform as it depends on the average popularity which is determined immediately before the election. 2

For illustrative purposes, in Figure 5, the swing voter has a negative ideology,σ 𝐽𝐽 < 0. However, it may be the case that σJ ≥0.

2

26

POLITICAL ECONOMICS

Figure 5: Swing Voter

2.2.5 The candidates’ decisions As discussed above, candidates have to decide on their policy platform – simultaneously and independently from each other – before their average popularity is known, that is, before potentially negative publicity. Due to this uncertainty, candidates – despite being opportunistic – are unable to choose their platform in order to guarantee winning the election. They can only maximize the probability of being elected, but they remain subject to the occurrence of a scandal. In maximizing their chances of being in power, candidates select their platform in an attempt to please as many citizens as possible. However, public policies often have redistributive effects; their implementation can work for certain groups of voters more than others. In this respect, candidates use all the available information - such as the distribution of ideology within a group and the distribution of scandal probability - to guide their decision. Let us examine candidate A’s political decision. By using the swing voter in each group, we can identify the individuals who will vote for candidate A: the voters to the left of the swing voter. Hence, the mass of voters for candidate A in group J is given by the area displayed in Figure 6. The votes for A in group J are equal to: �σ𝐽𝐽 +

1

𝐽𝐽

2Φ

1 2

1 2

� Φ 𝐽𝐽 = σ𝐽𝐽 Φ 𝐽𝐽 + = + Φ 𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )] − δΦ 𝐽𝐽 (3)

2.

Electoral Models

27

Figure 6: Voters for Candidate A

By applying the same reasoning to all groups, and summing up votes across all socio-economic groups, we have that the total share of votes for candidate A in an election – which we indicate with ΠA – is equal to: 1

Π𝐴𝐴 = 2 ∑𝐽𝐽 𝛼𝛼 𝐽𝐽 + ∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ 𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )] − δ ∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ𝐽𝐽

(4)

When does candidate A win the election? In order to win the election, he has 1 to obtain the majority of votes, that is Π𝐴𝐴 > . Modifying the equation above, 2 we have: 1

1

Π𝐴𝐴 = 2 + ∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )] − δΦ > 2

(5)

since ∑ 𝛼𝛼𝐽𝐽 = 1 and we defined Φ = ∑ 𝛼𝛼 𝐽𝐽 Φ 𝐽𝐽 , as the average density among the three income groups, that is, the average ideology within the electorate. Thus, 1 candidate A wins the election, Π𝐴𝐴 > , if the following expression holds: 2

∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ 𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )] − δ Φ > 0

(6)

This expression, which depends on both candidates’ policy platforms, on the average popularity and on the ideology densities in the different groups, may also be written as a function of δ. This expression highlights that candidate A’s chances of winning the election depends on the realization of the average popularity:

δ < ∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )]/Φ = δ

(7)

Therefore, candidate A wins the elections, if the realization of the popularity shock that takes place before the election, δ, is below a certain threshold, δ.

28

POLITICAL ECONOMICS

The intuition is straightforward: candidate A wins the elections if he is not hit by a scandal or negative news that is “too large”. As discussed earlier, candidates are unable to control the outbreak of these scandals. However, they may select a particular policy platform in order to maximize their probability of winning the election or – analogously – in order to reduce the probability of losing in case a scandal breaks out. As shown in Figure 7, the probability of winning the election is the same as the probability that a 1 scandal (or negative news) below a certain threshold occurs, 𝑃𝑃𝑃𝑃 �Π𝐴𝐴 > � = 2

𝑃𝑃𝑃𝑃�δ < δ�. Moreover, notice that this threshold, δ, is endogenous, as it depends on the candidates’ platform. Candidate A will hence maximize the probability of winning the election by setting its policy platform, (XA,YA), in order to maximize the probability that the scandal is below this threshold: 1

1

Ψ

𝑃𝑃𝑟𝑟 �Π𝐴𝐴 > � = 𝑃𝑃𝑃𝑃�δ < δ� = + �∑𝐽𝐽 𝛼𝛼 𝐽𝐽 Φ 𝐽𝐽 [𝑈𝑈𝐽𝐽 (𝑋𝑋𝐴𝐴 , 𝑌𝑌𝐴𝐴 ) − 𝑈𝑈𝐽𝐽 (𝑋𝑋𝐵𝐵 , 𝑌𝑌𝐵𝐵 )]� (8) 2 2 Φ

As suggested by the equation above, the policy platform chosen by candidate A, (XA,YA), will aim to represent the interests of voters in the most numerous (higher αJ) and least ideological (higher ΦJ) groups. Clearly, economic policies targeting more numerous (socio-economic or income) groups (larger αJ) have the advantage of providing benefits to more potential voters. However, some voters in these more numerous groups may be highly ideological, and hence difficult to convince, despite the potential benefits provided to them by the platform. Hence, politicians may prefer to target less ideological groups (large ΦJ). Groups with a large proportion of nonideological voters – that is, with several swing voters – almost regardless of their size, are easy targets for politicians, who seek their vote by using their policy platforms accordingly. If this is the strategy for candidate A, it is important to notice that candidate B will follow the same strategy. Figure 7: Candidate A’s probability of winning the election

2.

Electoral Models

29

Candidate B will choose his policy platform, (XB,YB), in order to maximize his 1 probability of being elected 𝑃𝑃𝑃𝑃 �Π𝐵𝐵 > �. In a simple majority voting election 2 between two candidates, maximizing his probability of being elected clearly amounts to minimize the probability of candidate A winning the election: 1 𝑃𝑃𝑃𝑃 �Π𝐵𝐵 > � = 𝑃𝑃𝑃𝑃�δ > δ� = 1 − 𝑃𝑃𝑃𝑃�δ < δ�. Since the maximization problem of 2 candidate B mirrors that of candidate A, in equilibrium both candidates will set the same policy: (XA,YA) = (XB,YB). Hence, according to the probabilistic model, just like in the median voter model, both candidates will converge to a common policy platform. The electoral incentives will help to determine the public policy, which targets the less ideological – or swing – voters, rather than the median voter. Yet, the identity of the elected candidate (whether candidate A or B) will be determined by the popularity shock. 2.3 An Example of Simple Majority Voting and Probabilistic Voting Consider the economy described in section 1.3, populated by Poor (P), Middle Income (M) and Rich (R) individuals, with incomes equal to YP = ½ , YM = ⅔, and YR = 1 and proportions αP = 45%, αM = 30%, and αR =25%. The indirect utility function is:

Questions:

𝑉𝑉(𝐺𝐺) = [𝑌𝑌𝑖𝑖 − 𝐺𝐺] + γ𝑖𝑖 ln 𝐺𝐺 = 𝑌𝑌𝑖𝑖 − 𝐺𝐺 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 ) ln 𝐺𝐺

a) Calculate the political equilibrium in a majority voting election; b) Calculate the political equilibrium in a probabilistic voting model, in which the density of the ideology distribution function is equal to one in every income group, ΦJ=1 for all J, and the density of the average popularity, Ψ, is equal to one. Solution: a) Recall that the agents’ bliss points are Gi*= A-Yi; and hence GP* = 1½= ½, GM* = 1-⅔= 1/3 and GR*= 1-1= 0. Under majority voting, since preferences are single-peaked, the median voter theorem applies; hence the political equilibrium outcome corresponds to the middle-income voter’s bliss point: G* = GM*= 1/3. To see this, suppose that this policy, G* = 1/3, is compared to a lower level of public good, for instance, G = 0. Clearly, poor and middle-income voters would prefer G* to G = 0, while only the rich would prefer G = 0.

30

POLITICAL ECONOMICS

On the other hand, if this policy, G* = 1/3, is compared to a higher level of public good, for instance, G = ½, wealthy and middle-income voters would prefer G* to G = 1/2. Hence, G*=1/3 is the economic policy which becomes the political equilibrium in this simple majority voting election. Notice that for G*=1/3:

1 1 1 𝐶𝐶𝑃𝑃 = 𝑌𝑌𝑃𝑃 – 𝐺𝐺 = − = 2 3 6 2 1 1 𝐶𝐶𝑀𝑀 = − = 3 3 3 𝐶𝐶𝑅𝑅 = 1 −

and

𝑈𝑈𝑃𝑃 = 𝑈𝑈𝑀𝑀 =

1 2 = 3 3

1 1 1 + 𝑙𝑙𝑙𝑙 � � 6 2 3

1 1 1 + 𝑙𝑙𝑙𝑙 � � 3 3 3

𝑈𝑈𝑅𝑅 =

2 3

b) Probabilistic Voting. Recall that the three groups are such that: 𝑌𝑌𝑃𝑃 = ½ < 𝑌𝑌𝑀𝑀 = ⅔ < 𝑌𝑌𝑅𝑅 = 1, and 𝛼𝛼𝑃𝑃 = 0.45 > 𝛼𝛼𝑀𝑀 = 0.30 > 𝛼𝛼𝑅𝑅 = 0.25. Individual ideologies within each group are distributed according to a uniform distribution function with a zero mean and unitary density, Φ 𝐽𝐽 = 1 ∀𝑗𝑗: Analogously, the average popularity is distributed according to a uniform distribution function with a zero mean and unitary density, Ψ=1. How do candidates choose the level of public good, G? Let us identify the swing voter in group j: σ𝐽𝐽 = 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐴𝐴 ) − 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐵𝐵 ) − δ .

In any group j, the fraction of the voters in favor of candidate A is equal to: �σ𝐽𝐽 +

1

2Φ𝐽𝐽

1

1

� Φ 𝐽𝐽 = σ𝐽𝐽 Φ 𝐽𝐽 + 2 = σ𝐽𝐽 + 2.

Hence, by summing up over all the (income) groups, we have that the fraction of all voters for A is:

2.

Electoral Models

31

1 2

π𝐴𝐴 = �[𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐴𝐴 ) − 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐵𝐵 )]𝛼𝛼 𝐽𝐽 Φ𝐽𝐽 − δ � 𝛼𝛼 𝐽𝐽 Φ𝐽𝐽 + � 𝛼𝛼 𝐽𝐽 𝐽𝐽

𝐽𝐽

Since ∑ 𝛼𝛼 𝐽𝐽 Φ 𝐽𝐽 = 1, and Φ 𝐽𝐽 = 1, then:

π𝐴𝐴 = � 𝛼𝛼 𝐽𝐽 [𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐴𝐴 ) − 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐵𝐵 )] − δ + 𝐽𝐽

1 2

Candidate A wins the election if π𝐴𝐴 > , that is, if

� 𝛼𝛼 𝐽𝐽 [𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐴𝐴 ) − 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐵𝐵 )] − δ +

𝐽𝐽

1 2

1 1 > 2 2

δ < ∑ 𝛼𝛼 𝐽𝐽 [𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐴𝐴 ) − 𝑉𝑉 𝐽𝐽 (𝐺𝐺𝐵𝐵 )] = δ�

or,

1 1 Pr �π𝐴𝐴 > � = Pr�δ < δ� � = δ� + 2 2

Hence, candidate A maximizes: 𝑀𝑀𝑀𝑀𝑀𝑀𝐺𝐺𝐴𝐴

1 + � 𝛼𝛼 𝐽𝐽 [𝑌𝑌𝑖𝑖 − 𝐺𝐺𝐴𝐴 + (𝐴𝐴 − 𝑌𝑌𝑖𝑖 ) ln 𝐺𝐺] − � 𝛼𝛼 𝐽𝐽 𝑉𝑉(𝐺𝐺𝐵𝐵 ) 2

by taking first order conditions, we have: 𝛼𝛼𝑃𝑃 �−1 +

𝐴𝐴 − 𝑌𝑌𝑃𝑃 𝐴𝐴 − 𝑌𝑌𝑀𝑀 𝐴𝐴 − 𝑌𝑌𝑅𝑅 � + 𝛼𝛼𝑀𝑀 �−1 + � + 𝛼𝛼𝑅𝑅 �−1 + �=0 𝐺𝐺𝐴𝐴 𝐺𝐺𝐴𝐴 𝐺𝐺𝐴𝐴

Since 𝐴𝐴(𝛼𝛼𝑃𝑃 + 𝛼𝛼𝑀𝑀 + 𝛼𝛼𝑅𝑅 ) = 1 and 𝛼𝛼𝑃𝑃 𝑌𝑌 𝑃𝑃 − 𝛼𝛼𝑀𝑀 𝑌𝑌 𝑀𝑀 − 𝛼𝛼𝑅𝑅 𝑌𝑌 𝑅𝑅 = 𝑌𝑌� , we have: 𝛼𝛼𝑃𝑃 (𝐴𝐴 − 𝑌𝑌𝑃𝑃 ) 𝛼𝛼𝑀𝑀 (𝐴𝐴 − 𝑌𝑌𝑀𝑀 ) 𝛼𝛼𝑅𝑅 (𝐴𝐴 − 𝑌𝑌𝑅𝑅 ) + + 𝐺𝐺𝐴𝐴 𝐺𝐺𝐴𝐴 𝐺𝐺𝐴𝐴 𝐺𝐺𝐴𝐴 = 𝐴𝐴(𝛼𝛼𝑃𝑃 + 𝛼𝛼𝑀𝑀 + 𝛼𝛼𝑅𝑅 ) − 𝛼𝛼𝑃𝑃 𝑌𝑌𝑃𝑃 − 𝛼𝛼𝑀𝑀 𝑌𝑌𝑀𝑀 − 𝛼𝛼𝑅𝑅 𝑌𝑌𝑅𝑅

𝛼𝛼𝑃𝑃 + 𝛼𝛼𝑀𝑀 + 𝛼𝛼𝑅𝑅 = 1 =

Hence, there are two different equilibria. With respect to the level of the public good, we have: 𝐺𝐺𝐺𝐺𝐺𝐺 ∗ = 0.325 < 𝐺𝐺𝐺𝐺𝐺𝐺 ∗ = 0.33

Respectively in the probabilistic voting model and in the median voter model, which reflect the different decisions by the “swing” voter and by the median voter.

3. Lobbying

Several economic policies – such as pension issues – affect the interests of a vast majority of the population, and often become important issues in political candidates’ platforms during election campaigns. On the other hand, other policy decisions only affect a limited number of well-defined individuals, who may enjoy large benefits, and have only a small (sometimes negligible) impact on the other citizens. Examples of the latter include the provision of certain local public goods, targeted trade policy and the regulation of some specific markets –for instance taxis and notaries. Since these policy issues are not relevant to the large majority of the voters (and hence to the median voter) or to the undecided (swing) voters, they may fail to gain the central political stage, unless a specific group of well-organized political actors put these issues under the spotlight. These issues may thus fail to capture the policy-makers attention, unless organized, powerful political actors manage to promote these policy issues. When there is an opportunity for a relatively small and well-organized group to obtain large (and concentrated) benefits while spreading the cost on the entire society – perhaps imposing only a small (negligible) penalty on the citizens – this group may consider exerting pressure on the policy-makers – that is, to lobby them – in order to promote the policy that best suits their interests. In fact, lobbying is the process used by small groups to promote a policy which favors their interests, even though this policy might not benefit all of society or even a majority of the electorate. This chapter characterizes the lobbying activity as part of the political processes through which individual preferences are aggregated into policy outcomes. This political process has the distinctive feature of providing more power – and hence more favorable outcomes – to specific groups. To address these issues, this chapter will consider the policy decisions on the provision of a local public good, such as a bridge, a road or a local subsidy. To measure the impact of the lobbying activity on the

34

POLITICAL ECONOMICS

final policy outcome, we compare four alternative scenarios, which differ in how the policy-making process works. In the first scenario, we present the social optimum, chosen by a utilitarian social planner, who cares about all the citizens. The second scenario describes a case of complete fiscal federalism, in which the decisions on the local public good, and how to finance it, are taken at local level. In the third scenario, fiscal federalism is only partial. The spending decisions on the public good are decentralized while the financing of the public good is centralized. As we will see, this creates a situation known as “soft budget constraint” that encourages overspending. Finally, in the fourth scenario, we analyze a lobbying model in which a group of citizens exerts pressure on policy-makers. We consider a semibenevolent utilitarian social planner, who cares about all citizens, but also about the monetary contribution provided by the lobbies. We analyze how this lobbying activity modifies the policy-maker’s incentives, and how it may distort economic policies. 3.1 Public Policy: A Local Public Good To understand the impact of lobbying on the policy decisions, we consider a special case of economic policy, which typically provides benefits to a small, well-defined group of individuals, while spreading the cost to all citizens. We concentrate on the determination of a locally provided public good (a bridge on a river, a road or a local subsidy, such as the social income provided to citizens n Alaska), and on its redistributive effects. In this model, the economy is populated by three distinct groups of individuals, easily distinguishable by their relevant characteristics (e.g. occupation, income, geographic location). Consistent with an analysis of a local public good, it is convenient to assume that individuals differ in their geographic location. These three separate groups of individuals – identified as belonging to group (or area) 1, 2, and 3, have the same income 𝑦𝑦 > 1, but may be heterogeneous in their relative size. The relative size of group i is characterized by αi, with 𝛼𝛼1 + 𝛼𝛼2 + 𝛼𝛼3 = 1 Since groups differ in their geographic location, there may be three different local public goods, 𝑔𝑔𝑖𝑖 = {𝑔𝑔1 , 𝑔𝑔2 , 𝑔𝑔3 }, where 𝑔𝑔𝑖𝑖 benefits only group i and is provided in equal per capita amounts to individuals in group i. The utility function of an individual in group i is thus the following: 𝑈𝑈𝑖𝑖 = 𝐶𝐶𝑖𝑖 + ln 𝑔𝑔𝑖𝑖

(1)

3.

Lobbyng

35

where Ci represents the private consumption in group i. The economy is completely static, hence, consumers have no incentive to save for future consumption and consume all their net income, which is given by the difference between their income y and any tax that may be levied on them by the policymaker. To analyze the different incentives faced by politicians and hence the different policy outcomes, we consider four alternative processes of policy formation: 1. Social optimum. 2. Complete Fiscal Federalism. Local public goods decided at a local level. 3. Partial Fiscal Federalism. Local public goods decided at a local level, but financed at a central level. 4. Lobbying. 3.1.1 The Social Optimum In the social optimum case, we consider the decision of a social planner – that is, a benevolent dictator, who cares exclusively about the utility of all citizens. In particular, if the social planner is utilitarian, he will care about the sum of the utility of all citizens, weighted by their relative size. Hence, the social planner decision will be taken to maximize this overall utility with respect to the three local public goods, 𝑔𝑔1 , 𝑔𝑔2 , 𝑔𝑔3 , subject to the aggregate resource constraint. We thus have the following maximization problem:

𝑠𝑠. 𝑡𝑡.

max 𝑈𝑈 = 𝛼𝛼1 𝑈𝑈1 + 𝛼𝛼2 𝑈𝑈2 + 𝛼𝛼3 𝑈𝑈3

{𝑔𝑔1 ,𝑔𝑔2 ,𝑔𝑔3 }

(2)

𝛼𝛼1 (𝑔𝑔1 + 𝐶𝐶1 ) + 𝛼𝛼2 (𝑔𝑔2 + 𝐶𝐶2 ) + 𝛼𝛼3 (𝑔𝑔3 + 𝐶𝐶3 ) = 𝑦𝑦(𝛼𝛼1 + 𝛼𝛼2 + 𝛼𝛼3 ) (3)

where the utility function for each group is 𝑈𝑈𝑖𝑖 = 𝐶𝐶𝑖𝑖 + ln(𝑔𝑔𝑖𝑖 ), and the last equation represents the resource constraint, which equalizes the total amount of (private and public) goods consumed in the economy – the left hand side of the equation – with the total amount of resource, that is, the total income in the economy, which is on the right hand side. The resource constraint can be rewritten as follows: 𝛼𝛼1 𝐶𝐶1 + 𝛼𝛼2 𝐶𝐶2 + 𝛼𝛼3 𝐶𝐶3 = 𝑦𝑦 − 𝑔𝑔1 𝛼𝛼1 − 𝑔𝑔2 𝛼𝛼2 − 𝑔𝑔3 𝛼𝛼3

(4)

Using the utility function and the resource constraint, the maximization problem of the social planner becomes: max 𝛼𝛼1 (𝐶𝐶1 + ln 𝑔𝑔1 ) + 𝛼𝛼2 (𝐶𝐶2 + ln 𝑔𝑔2 ) + 𝛼𝛼3 (𝐶𝐶3 + ln 𝑔𝑔3 )

{𝑔𝑔1 ,𝑔𝑔2 ,𝑔𝑔3 }

(5.a)

36

POLITICAL ECONOMICS

or alternatively as: max 𝑦𝑦 + 𝛼𝛼1 (ln 𝑔𝑔1 − 𝑔𝑔1 ) + 𝛼𝛼2 (ln 𝑔𝑔2 − 𝑔𝑔2 ) + 𝛼𝛼3 (ln 𝑔𝑔3 − 𝑔𝑔3 )

{𝑔𝑔1 ,𝑔𝑔2 ,𝑔𝑔3 }

(5.b)

This maximization process yields the following set of first order conditions: 𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔1 ): 𝛼𝛼1 �

1 − 1� = 0 ⇒ 𝑔𝑔1 = 1 𝑔𝑔1

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔3 ): 𝛼𝛼3 �

1 − 1� = 0 ⇒ 𝑔𝑔3 = 1 𝑔𝑔3

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔2 ): 𝛼𝛼2 �

1 𝑔𝑔2

− 1� = 0 ⇒ 𝑔𝑔2 = 1

(6)

Thus, the benevolent utilitarian social planner provides every group with the same amount of the local public good, 𝑔𝑔𝑖𝑖 = 1 ∀ 𝑖𝑖. Rather than representing a realistic description of an actual political process of policy formation, the social optimum represents a useful benchmark for comparison. It is in fact the optimal policy. Any deviation from this policy will be due to a distortion in the process of policy formation. 3.1.2 Complete Fiscal Federalism In this second scenario, we consider a situation of complete fiscal federalism where the level of the local public good is determined entirely at the local level. In particular, each group, i chooses the size of the local public good, gi, and the lump-sum tax, τi, to levy on its members in order to finance the spending. Complete fiscal federalism implies that both spending and taxation concerning the financing of the local public good are decentralized at the local level. Hence, for every group i, the budget constraint becomes 𝐶𝐶𝑖𝑖 = 𝑦𝑦 − τ𝑖𝑖 , and the size of the public good becomes 𝑔𝑔𝑖𝑖 = τ𝑖𝑖 . Since individuals are identical within each group, we can deem the representative individual for every group i, who will take the policy decision over 𝑔𝑔𝑖𝑖 = τ𝑖𝑖 in order to maximize the utility of the (identical) individuals in the group. The maximization problem is thus: max 𝑈𝑈𝑖𝑖 = max 𝐶𝐶𝑖𝑖 + ln 𝑔𝑔𝑖𝑖 = max 𝑦𝑦 − 𝑔𝑔𝑖𝑖 + ln 𝑔𝑔𝑖𝑖 {𝑔𝑔𝑖𝑖 }

{𝑔𝑔𝑖𝑖 }

{𝑔𝑔𝑖𝑖 }

which yields the following first order condition and its related solution:

(7)

3.

Lobbyng

37

1

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔𝑖𝑖 ) : − 1 + = 0 ⇒ 𝑔𝑔𝑖𝑖 = 1 𝑔𝑔

(8)

𝑔𝑔 = 𝛼𝛼1 𝑔𝑔1 + 𝛼𝛼2 𝑔𝑔2 + 𝛼𝛼3 𝑔𝑔3 = 1

(9)

𝑖𝑖

Hence, the aggregate spending on all the local public goods is:

Interestingly, in these local public good decisions, complete fiscal federalism yields the same result as the benevolent social planner. This should not be surprising. When the level of local public good is decided entirely at a local level, and the citizens interested in the public good determine both spending, gi, and financing, τi , they fully internalize the costs and benefits from the local public good provision. 3.1.3 Partial Fiscal Federalism We now consider a situation in which the size of the public good is decided at a local level, but is funded centrally, through a common income tax levied on the income of all citizens in all geographic areas. As we shall see, results will look very different in this case. In this scenario, each group i chooses the level of its own local public good, gi, and the financing of all the local public goods is determined residually by setting a nation-wide lump-sum tax, τ. The combination of decentralized spending decisions and centralized financing gives rise to a “soft” budget constraint, since individuals in every group do not fully internalize the cost of their own group’s local public good, while they are determining its level. To see this, consider the aggregate budget constraint for all local public goods:

τ(𝛼𝛼1 + 𝛼𝛼2 + 𝛼𝛼3 ) = τ = 𝛼𝛼1 𝑔𝑔1 + 𝛼𝛼2 𝑔𝑔2 + 𝛼𝛼3 𝑔𝑔3

(10)

𝐶𝐶𝑖𝑖 = 𝑦𝑦 − τ = 𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3

(11)

Total spending on local public goods (the right hand side of the equation above) is financed through a nation-wide lump-sum tax, τ, (the left hand side of the equation above). Unlike the previous scenario, when both spending and financing were decided at the local level, here the spending is decided at a local level in every group. However, the financing – i.e. the lump-sum tax– is determined residually to balance the budget. The individual budget constraint for a citizen in group i appears as follows:

where the last equation is obtained by using the aggregate budget constraint in Equation 10.

38

POLITICAL ECONOMICS

Thus, in the political process, a representative citizen in each group will choose the level of their own local public good, gi, in order to maximize his group utility function, given the individual budget constraint above. We hence have the following maximization problem: max 𝑈𝑈𝑖𝑖 = max 𝐶𝐶𝑖𝑖 + ln 𝑔𝑔𝑖𝑖 = max 𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3 + ln 𝑔𝑔𝑖𝑖 (12) {𝑔𝑔𝑖𝑖 }

{𝑔𝑔𝑖𝑖 }

{𝑔𝑔𝑖𝑖 }

which yields the following first order condition and its related solution: 1

1

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔1 ): − 𝛼𝛼1 + = 0 ⇒ 𝑔𝑔1 = > 1 𝑔𝑔 𝛼𝛼 1

1

(13)

As expected, spending on the local public good is larger than the social optimum. This is because, when taking their political decision at the local level, individuals fail to fully internalize the cost of the public good. In fact, a unitary increase in the local public good for one group only increases its tax bill by a fraction αi. This leads to another interesting conclusion. Small groups are more likely to spend more, since they internalize a smaller percentage of the costs associated with their local public good than larger groups. When summing up for all the groups, the aggregate spending is three times larger than optimal: 𝛼𝛼

𝛼𝛼

𝛼𝛼

𝑔𝑔 = 𝛼𝛼1 𝑔𝑔1 + 𝛼𝛼2 𝑔𝑔2 + 𝛼𝛼3 𝑔𝑔3 = 𝛼𝛼1 + 𝛼𝛼2 + 𝛼𝛼3 = 3 1

2

3

(14)

To summarize, the concentration of benefits within each group – due to a local public good – and the dispersion of costs – due to general taxation – leads to overspending. 3.2 Lobbying We now analyze a scenario in which individuals in one of the three groups – that is, citizens in one of the three geographic locations – decide to exert pressure on the policy-makers through lobbying. This pressure may occur in many different ways. Citizens may spend time and/or money trying to convince politicians to pursue a specific policy that favors them. Citizens may use their time to write letters to politicians, or to organize sit-ins or other forms of protest in order to convince them to endorse their policy. Alternatively, individuals may provide resources (e.g. money) such contributions to election campaigns, gifts, or outright bribes. Indeed, possibly the most recurrent and effective lobbying activity is providing politicians with information about a policy or a particular issue at stake. This information can be technical, since it consists of data and predictions about the consequences of the alternative policies; but it can also

3.

Lobbyng

39

be political, to the extent that the lobbying group informs the politicians about the potential impact of the alternative policies on voters in their electoral districts. The aim of these lobbying activities is to steer politicians’ preferences – and consequently, policy decisions – away from considering the wellbeing of all citizens to pursue more policies closer to the interests of the lobbying group. As – otherwise benevolent – politicians are willing to modify their policy-decisions due to the influence of lobbies, the resulting policy will typically not represent the social optimum. To address this issue, we consider a simple environment in which a policymaker decides the level of local public goods, as described in the previous section. The political decision process resembles the social planner situation described in section 3.1.1, in that the policy-maker is a (partially) benevolent social planner. As before, there are three groups that differ in their geographic location. Yet, one of them – for simplicity group 1 – is a lobby, while the other two groups (2 and 3) are not. Clearly, the lobby (group 1) will try to influence the allocation of public goods by the decision maker – i.e. the benevolent social planner – in its favor, and hence against groups 2 and 3. In this section, we abstract from two additional and relevant issues that deserve further analysis. First, we do not address the question of which group decides to lobby. We simply assumed it to be group 1. Second, we do not discuss the free-rider problem in lobbying. Free riding may occur when any individual in a group refrains from pursuing a costly (in terms of money and/or time) lobbying activity and prefers to rely on the actions of the fellow group members. Clearly, if all individuals share this incentive, no lobbying will take place. In this section, we simply assume that the free rider problem does not occur. We consider a model in which a lobby tries to influence the policymaker’s decisions through bribery. In particular, the lobby (i.e. group 1) chooses a monetary contribution to the politicians, which depends on the policy outcome: M1(g1). It is convenient to describe this possible contribution as follows: 𝑀𝑀1 (𝑔𝑔1 ) = 𝑈𝑈1 (𝑔𝑔1 ) − 𝑏𝑏1 ≥ 0

(15)

where U1(g1) is the utility provided to group 1 by the level of local public good g1 and b1 is the reservation utility of group 1. It follows that the total utility of the individuals in group 1, after the monetary cost of lobbying M1(g1) and the realization of the policy outcome (g1), is given by the reservation utility: 𝑈𝑈1 (𝑔𝑔1 ) − 𝑀𝑀1 (𝑔𝑔1 ) = 𝑏𝑏1 . This implies that group 1 is willing to pay the legislator – at most – an amount equal to the entire difference between group 1’s utility from a policy g1 and the reservation utility, b1. This amount can hence be interpreted as an equivalent of the “consumer” surplus for the lobby.

40

POLITICAL ECONOMICS

The policy-maker, on the other hand, is a semi-benevolent social planner who cares about two issues, the social welfare and the lobby’s contribution. In particular, the policy-maker gives weight η to the social welfare, 𝑈𝑈 = 𝛼𝛼1 𝑈𝑈1 + 𝛼𝛼2 𝑈𝑈2 + 𝛼𝛼3 𝑈𝑈3 , where Ui represents group i’s utility, and weight (1-η) to the lobby’s contribution: M1(g1). The parameter η, with 0 ≤ η ≤ 1 is thus a measure of the policy-maker’s benevolence. To summarize, the policy-maker thus has the following objective function: 𝑊𝑊(𝑔𝑔1 , 𝑔𝑔2 , 𝑔𝑔3 ) = = η (𝛼𝛼1 𝑈𝑈1 + 𝛼𝛼2 𝑈𝑈2 + 𝛼𝛼3 𝑈𝑈3 ) + (1 − η)𝛼𝛼1 𝑀𝑀1 (𝑔𝑔1 ) = η (𝛼𝛼1 𝑈𝑈1 + 𝛼𝛼2 𝑈𝑈2 + 𝛼𝛼3 𝑈𝑈3 ) + (1 − η)𝛼𝛼1 [𝑈𝑈1 − 𝑏𝑏1 ] = 𝛼𝛼1 𝑈𝑈1 (𝑔𝑔1 ) + η[𝛼𝛼2 𝑈𝑈2 (𝑔𝑔2 ) + 𝛼𝛼3 𝑈𝑈3 (𝑔𝑔3 )] − (1 − η)𝛼𝛼1 𝑏𝑏1

(16)

where the last expression is obtained by using equation 15 to find the lobby’s contribution. The aggregate and the individual budget constraints remain the same as in the partial fiscal federalism case (see the previous section), respectively:

τ(𝛼𝛼1 + 𝛼𝛼2 + 𝛼𝛼3 ) = τ = 𝛼𝛼1 𝑔𝑔1 + 𝛼𝛼2 𝑔𝑔2 + 𝛼𝛼3 𝑔𝑔3

(17)

𝐶𝐶𝑖𝑖 = 𝑦𝑦 − τ = 𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3

(18)

𝑈𝑈𝑖𝑖 = 𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3 + ln 𝑔𝑔𝑖𝑖

(19)

and, for individuals in any group i:

The utility for a group i is thus:

Finally, the policy-maker chooses g1, g2 and g3 in order to maximize the following welfare function: 𝑊𝑊(𝑔𝑔1 , 𝑔𝑔2 , 𝑔𝑔3 ) = = 𝛼𝛼1 ( 𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3 + ln 𝑔𝑔1 ) +η 𝛼𝛼2 (𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3 + ln 𝑔𝑔2 ) (20) +η 𝛼𝛼3 (𝑦𝑦 − 𝛼𝛼1 𝑔𝑔1 − 𝛼𝛼2 𝑔𝑔2 − 𝛼𝛼3 𝑔𝑔3 + ln 𝑔𝑔3 ) − (1 − η)𝛼𝛼1 𝑏𝑏1

which is obtained by substituting the utilities at equation 19 in equation 16. The policy-maker’s maximization leads to the following first order conditions and results: 𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔1 ): 𝛼𝛼1 �

1 𝑔𝑔1

− 𝛼𝛼1 � − η 𝛼𝛼1 𝛼𝛼2 − η 𝛼𝛼1 𝛼𝛼3 = 0

(21)

3.

Lobbyng

41

⇒ 𝑔𝑔1 =

1 𝛼𝛼1 +η𝛼𝛼2 +η𝛼𝛼3

>1

1

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔2 ): − 𝛼𝛼1 𝛼𝛼2 + η 𝛼𝛼2 � − 𝛼𝛼2 � − η 𝛼𝛼2 𝛼𝛼3 = 0 (22) 𝑔𝑔 ⇒ 𝑔𝑔2 = 𝛼𝛼

η

1 +η𝛼𝛼2 +η𝛼𝛼3

𝐹𝐹𝐹𝐹𝐹𝐹(𝑔𝑔3 ) ⇒ 𝑔𝑔3 =

1, or not. Lobbying models are useful tools to analyze the occurrence of economic policies that deliver targeted redistribution - that is, policies that combine concentrated benefits to a few individuals with small costs spread out across all citizens (and thus the entire electorate). Such policies include regulations in specific markets that provide market power to a limited number of incumbent professionals or firms, while spreading the cost to all consumers and to potential entrants. A recent example is taxi drivers lobbying against the entrance of Uber into the market. Yet, as mentioned earlier, lobbying models typically have a drawback, since they fail to successfully address the free-rider problem. This free-riding problem within the lobbying group occurs when individuals within a group expect other members of the group to pay the monetary and/or time costs of lobbying and hence choose not to contribute. If all individuals were to follow this behavior, lobbying would never happen in equilibrium. A real-life example of this type of free-riding problem can be found in the labor market with the decrease in trade union membership. Over time, the share of workers affiliated to the labor unions has drastically fallen. Since unions act as a lobby for workers, one possible explanation for this pattern is the free-riding problem. All the more so, because, in many countries, unions have already established their bargaining power with the firms’ representative and the government; as such the benefits from negotiations accrue to all workers, while the monetary costs are borne by a smaller group – the union members, who pay the union membership fees. This leaves the question of how to reduce free-riding in lobbying.

42

POLITICAL ECONOMICS

Social pressure has often been viewed as a possible solution, especially when potential lobby members are physically close and the lobbying activity is observable (as in the case of the labor unions). Finally, but related to this issue, the lobbying models leave another question partly unanswered. Why do some groups start lobbying, while others do not? One of the suggestions from the literature is that groups in which free-riding is more difficult and where the concentration of benefits is higher are more likely to form a lobby.

4. Facts, Data and Relevant Issues of the Welfare State

Welfare states are one of the most universal institutions in modern societies. Indeed, all countries experience some degree of government intervention in order to modify the market allocation of resources. In socialist countries, public intervention has historically taken the form of a complete control of resource allocation, with no role for markets or private property. In market economies, government involvement has been more moderate and has taken less extreme forms. According to the former US President, Theodore Roosevelt, “the object of the government is the welfare of the people. The material progress and the prosperity of a nation are desirable chiefly so far as they lead to the moral and material welfare of all good citizens”. If the mission of the government is to improve the well-being of its citizens, then the welfare state can be defined as a set of programs and instruments used to improve the welfare of the citizens. Two important issues arise with respect to public intervention. The first – and most fundamental – question is whether individuals can trust the government. These doubts are clearly expressed in the words of another former US President, Thomas Jefferson: “Sometimes it is said that man cannot be trusted with the government of himself. Can he, then, be trusted with the government of others? Or have we found angels in the form of kings to govern him?” Secondly, even when government’s good intentions are taken for granted, one may ask whether it is possible to perform Pareto-improving public policies that increase the well-being of all citizens. Several of the existing welfare programs entail strong redistributive elements, which improve the economic well-being of some individuals, while decreasing others. The former question has partially been addressed in the previous chapters, in which governments – or policy-makers – were analyzed from a political economic

44

POLITICAL ECONOMICS

perspective, as office-seeking politicians trying to be re-elected or as partisan politicians aiming to implement their most preferred policies. 1 The latter issue will be analyzed in the following chapters, which examine some of the programs that typically constitute a welfare state, and their redistributive nature. The aim of this chapter is to provide a description of the government intervention in the economy. For this reason, we initially concentrate on the size of the government by analyzing the relative importance of different instruments of public intervention. Particular relevance is given to the size and composition of welfare state programs across countries and over time, and to the role played by the government in relation to the labor market both directly – through the use of unemployment benefits – and indirectly – via employment protective legislation. The chapter then provides a brief overview of the normative explanations given in the public economic literature to justify government intervention in the economy. The final part will present the political-economic perspective on this issue by relating the existence of a large welfare state – or more generally of extensive government intervention – to the demand for redistribution by political agents – typically, the voters. 4.1 Economic Policies and the Role of Government Government involvement in the economy takes several forms. The most obvious intervention – and perhaps the most disliked one – is taxation. The resources that enable the public sector to perform its different activities are mainly obtained through tax revenues. Taxation can be classified in three main groups: direct taxes, indirect taxes and contributions. Labor income taxes, capital income tax and corporate taxes belong to the first group. They are levied respectively on the income of workers, savers and companies. They are typically progressive, as higher marginal tax rates are applied to higher incomes; however, they may also be proportional, if the same marginal tax rate is applied, regardless of the taxpayers’ income, or even regressive, when the average tax rate is decreasing in the taxpayer income due, for instance, to tax deductions. Indirect taxes, instead, are not levied on agents’ income, but to consumption, production or trade. Typical examples of indirect taxes are value added tax (VAT), sales tax, and import tariffs. Some of these taxes, such as the sales tax, may turn out to be regressive. As individuals with a low income channel a larger share of their income to consumption – hence paying proportionally more taxes – than high income individuals. Finally, contributions – i.e. the third group of tax revenues –usually finance a specific welfare program, such as pensions, health care or unemployment benefits. They are typically proportional to labor earnings, although floors and ceilings on taxable earnings often apply. 1 In the public choice school, Brennan and Buchanan (1980) depicted the government as a Leviathan – a monster whose only objective is to maximize tax revenue.

4.

Facts, Data and Relevant Issues of the Welfare State

45

Clearly, government involvement in the economy also takes place on the more popular spending side. Tax revenues are typically used to provide transfers to different groups of individuals or firms, to finance public investments and to purchase goods or services. The purchase of products – such as provisions for the army, public hospitals and schools – and of services – such as labor services by teachers, nurses or physicians in public schools and hospitals, which is measured by their wages –is defined as government consumption. Government investment refers to the purchase of equipment or to the build-up of public infrastructure, such as bridges or highways. Most of the focus of the following chapters, however, will be on government transfers – such as old age pensions, noncontributory pensions, disability pensions and unemployment benefits – and more specifically on the welfare state. Apart from tax revenues and public spending, government intervention also includes imposing regulations on several markets – such as labor, product, and financial markets – and on the trade policy. In the labor market, for instance, the government may impose employment protective legislation, which regulate the dismissal of employees, such as, for instance, the length of the notice period and possible severance payments. In the product market, the government may impose barriers to entry in a specific market – such as, taxi or notary services – thereby reducing the degree of competition. Additionally, natural monopolies – such as electricity and gas, transportation networks or railways – may be regulated by governments through the establishment of specific authorities, in order to limit the monopolistic pricing behavior. Financial markets – e.g. stock exchanges, financial intermediaries – are also highly regulated, typically in an attempt to protect the individuals against the risks arising from asymmetric information between the financial institution and the individual saver. However, perhaps the most regulated area is international trade. Governments may affect the degree of openness of an economy by levying tariffs on imports, or they may decide to impose quotas on importing particular goods and services. In several countries, governments also own certain companies, typically those considered to belong to strategic industries or to be “national champions”. Finally, governments provide several public goods, such as defense, legal and judiciary systems. The degree of government’s involvement in the economy differs substantially across countries, while being relatively stable over time. Table 1 displays the size of several types of government expenditure, as a percentage of GDP, in different countries in 2014. European countries are divided in three groups: Continental, Mediterranean and Scandinavian; while the UK and the US belong to the Anglo-Saxon countries.

46

POLITICAL ECONOMICS

Table 1: Government expenditure Consumption

PropSubsidies erty In(%) come (%)

Social benefit and other transfers (%)

Gross Investment (%)

Total

Goods and services (%)

Wages (%)

Austria

52.5

6.3

10.6

1.4

2.5

26.4

5.3

Belgium

55.1

4.3

12.7

3.4

3.2

27.4

4.1

France

57.3

5.1

13.0

2.2

2.2

29.9

4.9

Germany

44.1

4.7

7.7

0.9

1.8

25.7

3.4

Netherlands Continental Europe

46.2

6.3

9.2

1.2

1.4

24.2

3.9

51.1

5.3

10.6

1.8

2.2

26.7

4.3

Greece

49.9

4.8

12.1

0.9

4.0

23.3

4.8

Italy

51.3

5.5

10.2

1.9

4.6

25.4

3.7

Portugal

51.7

5.8

11.8

0.7

4.9

22.5

6.0

Spain Mediterranean Europe

44.5

5.3

11.0

1.1

3.4

20.7

2.9

49.3

5.3

11.3

1.1

4.2

23.0

4.4

Denmark

56.0

9.3

16.5

2.1

1.5

22.4

4.2

Finland

58.1

11.5

14.2

1.3

1.2

25.4

4.4

Norway

45.6

6.2

14.0

1.9

0.6

18.2

4.7

Sweden Scandinavian Europe United Kingdom

51.8

8.4

12.6

1.7

0.8

23.6

4.6

52.9

8.8

14.3

1.7

1.1

22.4

4.5

43.8

11.2

9.4

0.5

2.7

16.6

3.4

38.0

6.3

9.9

0.3

3.5

14.7

3.3

Country

United States

Source: OECD National Account Statistics database (2014).

Scandinavian countries – Denmark, Finland, Norway and Sweden – have the largest share of total government spending, on average almost 53% of GDP, followed by Continental European countries – particularly France and Belgium – that also feature a large share of total government spending, on average more than 51% of GDP; while in Mediterranean countries, total government spending is roughly 49% of GDP. In the UK and the US, however, government intervention is limited, as total spending is equal to 43.8% and 38% of GDP, respectively. The heterogeneity in government spending is hence quite large, ranging from 38% in the US to more than 58% in Finland.

4.

Facts, Data and Relevant Issues of the Welfare State

47

In all countries, social benefits and other transfers represent the most relevant item. This element ranges from 14.7% in the US to 29.9% in France. Scandinavian countries spend a large share - roughly 9% - of their GDP both on goods and services, and more than 14% of GDP in wages – hence in public employment. Finally, gross investments receive a rather small share of the GDP, between 2.9% in Spain and 6% in Portugal. Table 2: Tax Wedge (including employer’s pension contributions)

Country Australia

Year 1981

1987

1993

2000

2004

2007

2010

2015

--

--

--

31.0

31.0

31.0

31.0

28.4

47.3

48.3

48.8

48.2

49.5

Austria Belgium

49.8

53.5

54.6

57.1

55.4

55.6

55.9

55.3

Canada

24.7

29.0

30.8

32.9

31.9

31.3

30.4

31.6

Denmark

42.7

47.7

46.9

42.1

38.9

39.1

36.4

36.4

Finland

42.4

45.5

49.3

47.5

44.2

43.9

42.3

43.9

France

--

--

--

50.4

50.3

49.7

49.9

48.5

41.9

45.1

46.4

52.9

52.2

51.8

49.1

49.4

24.1

22.2

25.8

27.5

Germany Ireland

34.7

42.7

40.0

28.9

Italy

47.3

49.4

49.2

47.1

46.3

46.4

47.2

49.0

Japan

17.3

21.4

21.2

24.7

27.3

29.3

30.2

32.2

Netherlands

48.3

49.5

45.7

40.0

38.8

38.7

38.1

36.2

Norway

43.1

42.6

36.8

38.6

38.1

37.5

37.3

36.6

Portugal

29.9

34.5

33.3

37.3

37.4

37.3

37.1

42.1

Spain

37.4

37.9

38.0

38.6

38.8

39.0

39.7

39.6

Sweden

50.8

51.7

45.6

50.1

48.4

45.3

42.8

42.7

22.2

22.4

22.1

22.2

Switzerland

29.1

28.5

28.7

22.9

United Kingdom

37.6

36.0

32.6

32.6

33.9

34.1

32.6

30.8

United States

35.3

30.6

31.2

30.8

30.5

30.9

30.7

31.7

Source: OECD Taxing Wages, 2015

Clearly, the differences in government spending are reflected in a similar heterogeneity in the level of taxation. Table 2 displays the tax wedge on labor income (including employer’s pension contributions) in different years from 1981 to 2015. In the early ‘80s, the tax wedge ranged from 17.3% in Japan to 50.8% in

48

POLITICAL ECONOMICS

Sweden. In 2015, the lowest tax wedge was in Switzerland, 22.2%, and the highest in Belgium, 55.3%. Table 2 also shows the evolution of these tax rates over time. During the ‘80s and in the early ‘90s, they had been increasing in several countries – most notably in Japan, from 17.3% in 1981 to 32.2% in 2015 – while they were reduced in other countries, such as the UK, from 37.6% in 1981 to 30.8% in 2015. Other notable cases of decreasing tax wedges in the last few decades are Denmark and the Netherlands. Table 3: Welfare State expenditure in 1980 and 2011 1980 France

Germany

Italy

Netherlands

Sweden

US

Welfare State Expenditures Pension Family Benefits Labor Market Health

20,80 9,60 2,40 0,00 5,40

23,00 10,90 2,00 0,50 6,30

18,00 8,90 1,00 0,60 5,30

24,10 6,70 2,30 2,10 4,80

28,60 8,40 3,50 1,60 7,20

13,30 6,30 0,80 0,90 3,50

Housing

0,40

0,10

0,00

0,20

1,10

0,20

2011 France

Germany

Italy

the Netherlands

Sweden

US

Welfare state Expenditures Pension

30,50

24,70

27,30

22,00

25,80

19,10

13,80

10,60

15,80

5,50

7,40

6,70

Family Benefits

2,80

2,10

1,30

1,50

3,50

0,70

Labour Market

2,30

1,80

1,70

2,40

1,80

0,70

Health

8,40

9,00

6,80

9,10

9,00

8,00

Housing

0,80

0,60

0,00

0,30

0,40

0,30

Source: Author’s calculation from OECD SOCX Database, 2016.

4.2 Size and Composition of the Welfare State The welfare state is the main focus of this chapter and the following chapters. The aim is to present the differences in the size and composition of welfare state across countries and to illustrate the motivations behind the introduction and the expansion of these schemes over time. Table 3 provides two snapshots of the size and composition of the welfare state in six OECD countries – France, Germany, Italy, the Netherlands,

4.

Facts, Data and Relevant Issues of the Welfare State

49

Sweden and the US – in 1985 and in 2011, using available OECD data. The composition of the welfare state in each country in these two years is displayed in Figures 1 to 3, which show the relative importance of each program within the welfare state. In 1985, Sweden was already spending 27% of its GDP on the welfare state, as opposed to only 12.8% in the US. All other countries in the sample were between 20% in Italy and 25% in France. In 2011, France had the largest welfare state expenditure –30.5 % of GDP – and the US the lowest with 19.1%. While the Netherlands and Sweden reduced their welfare state expenditure from 1985 to 2011, there was a remarkable growth in all other countries. Figure 1: Welfare State composition in France and Sweden in 1980 and 2011

As shown in Figure 1, the relative importance of pension and health spending in the French welfare state has increased from 65% to 73%. In Sweden, the relative importance of pension and health spending in total welfare spending has increased from 51% to 64%.

50

POLITICAL ECONOMICS

Figure 2 displays the situation in Italy and in the US. In Italy, the increase in welfare spending is largely due to a rise in pension expenditure, which accounted for 54% of welfare spending in 1985 and for 58% in 2011. In the US, the only significant upsurge was in public health spending: from 30% of total welfare spending in 1980, to 42% in 2011; while the relative importance of pension spending has largely decreased from 47% to 35%. Despite the large difference in size, the Italian and US welfare state share a common feature. Most resources are channeled to the elderly through pensions and health care, which in 2011 jointly amount to 77% of the total welfare spending in the US and 83% in Italy. Figure 2: Welfare State composition in Italy and the US in 1980 and 2011

Finally, Figure 3 depicts the evolution of welfare spending in Germany and the Netherlands. In the former, the slight increase in welfare spending – from 22.2% in 1985 to 24.7% in 2011 – is mostly due to an increase in health care spending

4.

Facts, Data and Relevant Issues of the Welfare State

51

– from 6.5% of GDP to 9%. Also in the Netherlands, health care spending increased significantly to more than 40% of total welfare spending in 2011. Figure 3: Welfare State composition in Germany and the Netherlands in 1980 and 2011

These figures suggest that – in most countries – the largest portion of the welfare budget is channeled towards the elderly through pensions and health care. In the following chapters, we will address the political decisions that determine the nature and scope of these welfare state programs, paying particular attention to the redistribution across generations. 4.3 Labor Market Issues One of the most important situations where the government intervenes both directly – by imposing taxes or providing transfers – or indirectly – by regulating

52

POLITICAL ECONOMICS

its structure or the behavior of its main players – is the labor market. Conventional economic theory analyzes labor services offered by the workers and requested by the firms. The interaction between labor demand and labor supply gives rise to an equilibrium wage and to an equilibrium level of employment, such that the number of workers willing to supply their services at the equilibrium wage is exactly equal to the number of jobs that the firms want to fill for that equilibrium wage. Yet, the labor market structure and the behavior of certain key players is affected by government policies. Workers may choose not to offer their labor services individually in a competitive fashion, but create a cartel representing all workers, which has a market power in bargaining with firms. Unions are an important example of such a cartel. Laws typically regulate their role in the economy and determine important characteristics such as union coverage and degree of centralization. Governments also determine the degree of regulation in the labor market, which establishes procedures for hiring and firing workers. Finally, governments may levy direct and large tax burdens on the labor market, as they impose labor income taxes and social contributions on earnings. Table 4: Unions’ characteristics in 2013

Country

Density1

Coverage2

Centralization3

1999

2008

2013

1999

2008

2013

1999

9.1

7.6

7.7

95.0

98.0

98.0

2.0

Germany

29.0

19.1

18.1

92.0

61.4

57.6

2.0

Italy

23.7

33.9

37.3

82.0

80.0

80.0

2.0

Japan

24.0

18.2

17.6

21.0

17.6

17.1

1.0

Spain

21.1

17.2

16.9

78.0

79.3

77.6

2.0

United States European Union

14.3

11.9

10.7

18.0

13.1

11.9

1.0

29.6

27.9

82.3

66.9

61.9

1.9

France

43.1

1. Percentage of workers enrolled in a union 2. Percentage of workers whose wage is determined by the union 3. Degree of centralization of wage bargaining 1 at firm level, 2 at industry level, 3 at central level Source: OECD, Economic Policy Reforms, 2016

The role of unions in Western economies dates back to the 19th century. Their current role may be characterized according to three main features: density, coverage and degree of centralization. The density rate is the percentage of the total workforce enrolled in a union, and hence measures the success of the union in attracting workers. This element is not directly related to the role of the state,

4.

Facts, Data and Relevant Issues of the Welfare State

53

although governments that are ideologically close to the unions (typically left wing governments) may adopt policies which encourage workers to become members. Over time, the density has decreased rapidly in most OECD countries as new generations of workers have proved less willing to join unions. In 1999, the average density in the EU was 43.1%, mainly due to the existence of Gant countries where membership of a union had a tangible impact on workers’ rights and entitlements. However, this number has fallen dramatically in recent years, plummeting to 27.9% in 2013. In 1999, large countries such as Germany, Italy and Spain, had a density rate between 21% and 29%, while it was less than 10% in France. In 2013, the differences were remarkable. In Germany and Spain, the recent values have dropped respectively to 18.1% and 16.9%. The same decreasing trend is evident in France with a density of only 7.7%, while Italy has experienced an upsurge to 37.3%. However, unions’ relevance - as defined by their impact in the wage bargaining - is better measured by their coverage, which is equal to the percentage of workers whose wages are determined by the unions. According to Table 4, 98% of the French workers in 2013 had their contract agreed by the unions, despite the unions “representing” only 7.7% of all workers. This phenomenon of extra-coverage - that is, the difference between coverage and density - estimates the importance of free-riding within the union’s membership. As the unions are already established, and hence bargain the contract and the wage for most workers, individual workers have no incentive to join because of the monetary cost (the membership) and the potential stigma of union membership in some working environments. As displayed in Table 4, the extra-coverage represents a pervasive European phenomenon. The final element characterizing a union is its degree of centralization, which describes whether the wage bargaining occurs at a company, sector or central level. Depending on their degree of centralization, unions internalize certain information and concerns in their bargaining process with firms’ representatives and the government, while leaving others aside. The US and Japan display a very decentralized model in which unions bargain at the company level, thereby sharing possible concerns and information about how labor cost may shape the competitiveness of individual companies. In continental European countries, unions, instead, typically bargain at the sector level, while certain Scandinavian countries have a highly centralized bargaining model. For the purpose of this book, it is important to stress that the second and third features of the unions – as described here – are established by law, and hence are considered under the direct influence of government policy.

54

POLITICAL ECONOMICS

Table 5: Degree of employment protective legislation. Quality measures increasing in the degree of rigidity from 0 to 6

Country

EPL - Temporary Work

EPL - Regular Workers

Overall Index

1985

2013

1985

2013

1985

2013

Australia

0.9

0.9

1.2

1.7

1.0

1.3

Austria

1.3

1.3

2.8

2.4

2.0

1.8

Belgium

4.6

2.4

1.8

1.9

3.2

2.1

Canada

0.3

0.3

0.9

0.9

0.6

0.6

Denmark

3.1

1.4

2.2

2.2

2.7

1.8

Finland

1.3

1.6

2.8

2.2

2.0

1.9

France

3.1

3.6

2.6

2.4

2.8

3.0

Germany

5.0

1.1

2.6

2.7

3.8

1.9

Greece

4.8

2.3

2.8

2.1

3.8

2.2

Ireland

0.3

0.6

1.4

1.4

0.8

1.0

Italy

5.3

2.0

2.8

2.7

4.0

2.3

Japan Netherlands

1.7

0.9

1.7

1.4

1.7

1.1

1.4

0.9

3.1

2.8

2.2

1.9

Norway

3.1

3.0

2.3

2.3

2.7

2.7

Portugal

3.4

1.8

5.0

3.2

4.2

2.5

Spain

3.8

2.6

3.5

2.0

3.6

2.3

Sweden

4.1

0.8

2.8

2.6

3.4

1.7

Switzerland United Kingdom United States

1.1

1.1

1.6

1.6

1.4

1.4

0.3

0.4

1.1

1.1

0.7

0.7

0.3

0.3

0.3

0.3

0.3

0.3

Source: OECD, Employment and Labour market statistics (2016)

One of the most important regulations the government imposes on the labor market is the so-called employment protective legislation, or EPL. The EPL is composed of laws, rules and regulations related to the dismissal of a worker by the employer. The OECD has constructed an index of EPL. Large degrees of EPL correspond to stricter regulations, and hence to a rigid labor market where is difficult to dismiss a worker. The rationale of the EPL is to protect workers

4.

Facts, Data and Relevant Issues of the Welfare State

55

against the risk of becoming unemployed, in particular, when dismisal is “unfair”. This legislation may take different forms. According to the OECD (1994), EPL measures are comprised of: (1) procedural inconveniences to dismissals (mainly capturing the complexity of the procedures needed to issue a dismissal notice); (2) notice and severance payments requirements (the time elapsed between the decision to dismiss a worker and his effective removal from the payroll); and (3) difficulty of dismissals, which measures the relevance of litigation costs and any possible bias in the judicial enforcement process. Bertola, Boeri and Cazes (2000) suggest that the third component of EPL – the difficulty of dismissal – accounts the most for the reduction in the dismissal rate associated with a stricter EPL. Hence, the first two elements have a severance payment component, which plays a role in insuring against the unemployment risk. However, the third element, the litigation costs, represents simply a deadweight loss. Figure 4: Changes in the degree of EPL for regular workers

Source: Author’s calculations on the basis of OECD, Employment and Labour market statistics (2016)

The degree of EPL in OECD countries differs widely. As displayed in Table 5, European countries, such as France, Norway and Spain, have stricter labor market regulations than, for instance, the US. These differences are apparent

56

POLITICAL ECONOMICS

when both regular and temporary workers are considered. For temporary workers, however, some flexibility was introduced in the 1990s, as shown in Table 5, particularly in the case of Germany, Italy and Sweden. On the other hand, labor market reforms for regular workers have proved extremely difficult. The most recent reforms of employment protection have been parametric, involving only marginal groups of the workforce. Moreover, they have been enforced “at the margin”, that is, only limited to new employees. This is confirmed by Figure 4, which shows the OECD index for the strictness of employment protection for regular workers in 1985 and 2013. The EPL for “regular” workers (i.e. workers with permanent contracts) has hardly changed at all in OECD countries over the decades, as most countries lie on the 45-degree line in Figure 4 - Finland, Greece, Portugal and Spain are notable exceptions. Hence, the degree of EPL seems to represent a stable political-economic equilibrium, since governments refrain from adopting any reforms. Characterizing these equilibria is the task for chapter 7. The last – but far from least – form of government intervention in the labor market is labor market taxation. Governments collect a large amount of revenues in the labor market in order to finance specific programs, such as public pensions, health care or unemployment benefits – these taxes are often called social contributions – and to finance general government spending. The tax wedge in Table 2 confirms the existence of a large difference among OECD countries. All these institutional features, such as the degree of EPL, the different strengths and characteristics of the unions, the size and redistributive element in the unemployment insurance system, and the labor market taxes, are all crucial elements, which, combined with the other features affecting the demand and supply of labor, determine the level of employment in the labor market and the unemployment rate. Table 6 presents the labor force participation and the unemployment rates in several OECD countries for 2016. Again, large differences occur, as the unemployment rate varies from less than 4% in Iceland– a country characterized by extremely large labor force participation – to more than 10% in France and Italy, and even more in the Spain, Greece and South Africa. 4.4 The Economic Approach to Government Intervention The economic literature has suggested a wide variety of reasons and situations in which government intervention in the market is justified or even needed. In any such occasion, markets do not operate efficiently, and the intervention of a benevolent government may increase the well-being of the citizens. These market failures occur in the case of public goods, externalities, asymmetric information or when the market structure is characterized by a monopoly.

4.

Facts, Data and Relevant Issues of the Welfare State

57

Table 6: Unemployment rate in the labor market

Country Australia Austria Belgium Brazil Canada Chile Colombia Czech Republic Denmark Estonia EU28 Finland France Germany Greece Hungary Iceland India Ireland Israel

76,95 75,50 67,58 72,60 77,97 66,76 74,48

Unemployment rate (%) 6,06 5,72 8,48 4,85 6,91 6,21 8,96

74,02 78,46 76,63 72,82 75,95

5,05 6,17 6,19 9,39 9,37

71,23 77,63 67,81 68,64 87,89 69,05

10,35 4,62 24,90 6,82 3,97 6,00

70,10 72,18

9,40 5,24

Labour Force participation (%)

Labour Force participation (%) 64,99 75,95 68,30 75,73 70,88 63,44 79,62

Unemployment rate (%) 11,89 3,38 3,64 9,88 6,66 4,33 6,87

Norway

78,40

4,30

OECD Avg Poland Portugal Russia Slovak Republic Slovenia South Africa Spain Sweden Turkey United Kingdom United States

71,32 68,11 73,40 73,41

6,77 7,50 12,44 5,57

70,91

11,48

71,75 58,51 75,45 81,73 56,09

8,96 25,35 22,06 7,43 10,24

77,63

5,30

72,61

5,29

Country Italy Japan Korea Latvia Luxembourg Mexico Netherlands

Source: OECD Economic Outlook, 2016

Public goods are those products or services that are non-rival in consumption and non-excludible. Because of these characteristics, competitive markets do not provide the correct incentives for firms to produce the efficient level of these goods. To see this, consider the non-rivalry in consumption feature. For typical goods – say, an apple – consumption by one individual prevents another person from consuming the same good. This aspect eliminates possible free-rider problems, as the second person will have to buy the good if he wants to consume it, and will not be able to free ride on the other individual’s purchase. A public good, for instance a lighthouse, is instead non-rival, as the “consumption” by a

58

POLITICAL ECONOMICS

boat of the service provided by the lighthouse does not prevent another boat from using it. In this case, the second boat does not have an incentive to “buy” the services of the lighthouse, but will find it more convenient to pretend not to need – or not to use – the service and thus to free ride on others. The latter characteristic, the non-excludability, prevents the seller from excluding individuals who did not buy the good from using it. In the case of the lighthouse, it is clearly impossible to exclude the second boat from benefiting from the service of the lighthouse if this boat is not willing to pay for the service. On these occasions, producers in competitive markets will have no incentive to provide any level of public good, since they will not be able to sell it due to the free-riding behavior of potential consumers. The conventional wisdom in economic literature is thus that the efficient provision of public goods, such as parks, railroads, the legal system and defence, has to come from the public sector. Another typical situation that calls for government intervention occurs when production or consumption externalities take place. A negative production externality represents a situation in which the production of a good or service has a negative impact on the welfare of other individuals, without the markets providing compensation for this negative effect. This typically occurs when, as a by-product of its production process, a firm pollutes nearby areas without taking care of its cleansing or paying appropriate compensation. In other words, the firm does not internalize the entire cost of production as it is not obliged to face the cost of polluting, and hence produces more than it is socially efficient. In this situation, the government may intervene in order to increase the welfare of individuals, either by imposing direct regulations that limits the firm’s production, or by levying taxes on production thereby inducing the firm to produce the socially efficient amount. A similar situation occurs – for the case of negative consumption externalities – when individuals smoke in a restaurant, thereby decreasing the utility that the other customers may obtain from consuming their dinner. Markets may also lead to inefficient allocations if some agents have market power – that is, the ability to affect prices. The extreme form of market power is monopoly, when there is only one seller in the market for a product. This monopolist will typically choose to increase prices above their marginal cost, by supplying fewer goods than a seller in a competitive market would do. This market failure is magnified in case of natural monopolies, that is, in situations in which the cost structure of the production sector naturally leads to the existence of a unique producer (and seller). This occurs when there are large initial fixed costs of production and only small marginal costs, such as, in the production and distribution of electricity or gas. In markets where monopoly power emerges because of large entry costs or non-competitive behavior by the incumbent firm, the government may guarantee a higher degree of competition

4.

Facts, Data and Relevant Issues of the Welfare State

59

by imposing competitive behavior through an anti-trust body. In natural monopolies, or when competition cannot be directly achieved, the government may instead try to regulate the market by imposing price caps. Another major reason for market failures is due to asymmetric information. Often the two sizes of a market, buyers and sellers, have different information on certain characteristics. This asymmetry of information may eventually lead to the disappearance of these potential markets. Consider a possible market for health insurance. A private insurance company offers a contract to cover health costs in exchange for a premium. Since the insurance company is unable to perfectly discriminate among health types – due to the asymmetric information problem – it will have to charge the same premium to all insured persons. In order to calculate the cost of this premium, the insurance company has to evaluate the average health status and the average health hazard of its subscribers. The larger the expected number of “good risks”, the lower the premium charged by the insurance company. However, people who perceive to have low health risk (notice that this is a private information) may not be willing to pay a high premium, and hence may not buy the insurance. Thus, a problem of adverse selection will occur, as the only subscribers to remain with the insurance company will be “bad health risks”. To compensate for this adverse selection of risk types, the insurance company will increase the premium. But for a high premium even the “bad risks” may be unwilling to buy the insurance and the market will disappear. Interestingly, the same type of reasoning applies to several insurance programs that belong to the welfare state. For example, the pension system which provides retirement income through annuities, and thus covers against the longevity risk – unemployment insurance and health care. The role of the government in these markets is to force all individuals to pay the insurance, as all citizens are, for instance, required to contribute to the health care system, regardless of their risk type (whether they are a good or bad heath risk). This compulsory behavior imposed by the government thus avoids any adverse selection problem despite the asymmetric information. The price to pay in this case is the redistribution introduced by the program. For instance, if all individuals contribute the same proportion to the health care system, but some of them use it more – as they are “bad risks” – the system redistributes in favor of the “bad risks”, who benefit more from the program. An additional problem due to asymmetric information is known as the moral hazard problem. Individuals covered by insurance – say, car theft insurance – have less incentives to decrease the probability that a negative event will occur, exactly because they are insured against this risk. The moral hazard problem remains present in the insurance programs offered by the government in the welfare state, and increases the cost of government intervention. For instance, free health care may induce people to seek medical care even when not needed. Or generous unemployment benefits

60

POLITICAL ECONOMICS

may allow workers to exert little effort, since they are ensured against the risk of losing their job. Economic literature also provides additional explanations for the government involvement in the economy – particularly the establishment of a welfare state. These motivations hinge on equity considerations and on a paternalistic view of the state. According to the equity motive, even if the market reaches an efficient allocation of resources, governments may still intervene in order to decrease the degree of income inequality in society or to reduce the share of the population below the poverty line. The underlying philosophical reason is that society may have a preference for less income or consumption inequality. Indeed, as shown at Table 7 - which reports the Gini coefficient for 2012 in several countries - income inequality can be quite large, particularly in developing countries. According to the paternalistic view, individuals are not farsighted enough to take care of themselves – for instance, they do not purchase insurance – and governments need to implement specific programs in order to force individuals to do so. This paternalistic justification has often been applied to pension programs, which effectively force individuals to set aside resources for old age consumption. 4.5 The Political-Economics Approach to Government Intervention The political economics approach to government intervention takes a completely different view. Government intervention needs not to be justified on efficiency, or equity reasons. Rather, government intervention hinges on political reasons. Politically appointed policy-makers set public policies in an attempt to increase their probability of winning an election. No equity or efficiency reasons are needed since policy-makers are only concerned about obtaining enough votes to win the elections. Public policies can be used as an instrument for politicians to demonstrate their competence (Besley, 2005 and 2006) or as a redistributive tool to convince certain groups of voters. Since the focus of the following chapters is on welfare states and labor markets, we consider different lines of redistribution introduced in welfare states that can be relevant for the policy-makers. Four types of redistributive cleavages emerge: (i) income, typically from the rich to the poor, as addressed in chapter 5; (ii) age, usually from the young to the elderly, as analyzed in chapter 6; (iii) employment status, between insiders (employed individuals with permanent contracts) and outsiders (unemployed or employed on temporary contracts), as studied at chapter 7; (iv) factors of production, between capital and labor. 2

2 The analysis of the class struggle between capitalists and workers goes back to Marx (1867), and has recently been revisited in modern terms by Piketty (2013).

4.

Facts, Data and Relevant Issues of the Welfare State

61

4.6 Citizens’ Opinions on the Welfare State Before turning to the political economics explanation for why some countries to establish and support a large welfare state – and highly regulated labor markets – it is useful to examine individual preferences about the size and the degree of redistribution in current welfare states in European countries. This analysis allows us to understand the overall support for welfare state programs, and the heterogeneous preferences of individuals depending on their socio-economic status. Figures 5 and 6 show the aggregate answers to statements about pension policies in a sample of EU countries as reported by the Euro-barometer, a survey run by Eurostat. In both cases, a large majority of the individuals favor the current level of pension benefits, and prefer to keep these generous pension entitlements, even if this implies higher taxes. Figure 5: Current pension levels should be maintained even if this means raising taxes or contributions

62

POLITICAL ECONOMICS

Figure 6: Contribution rates should not be raised even if this means lower pension levels

Additional survey evidence in Boeri, Boersch-Supon and Tabellini (2001) shows that a large majority of individuals in France, Germany, Italy and Spain are in favor of maintaining or even increasing the level of welfare state taxes and transfers. All this evidence suggests that current welfare states are supported politically. There are, however, large differences in individuals’ opinions over the optimal size of the welfare state. Interestingly, those opposed to the welfare state include the young and self-employed individuals; whereas employed, old, poor, medium or low skilled and union members support the welfare state. These aspects will be used to construct the theoretical models in the next chapters. Boeri, BoerschSupon and Tabellini’s (2001) study also analyze the intergenerational line of redistribution between young and old. They find that, in every country, a majority of those surveyed support the current pension system (that is, the status quo) or may even want to increase the pension benefits. Not surprisingly, elderly individuals, those with a permanent job, and with low levels of education are in favor of a more generous pension system, whereas the unemployed prefer more resources to be awarded to the unemployed and to young workers rather than to the retirees.

4.

Facts, Data and Relevant Issues of the Welfare State

63

Table 7: Income inequality as measured by the Gini Coefficient 2013 (disposable income after taxes and transfers)

Country

Value (%)

Country

Value (%)

Australia*

32,6

Korea

30,2

Austria

28,0

Latvia

35,2

Belgium

26,8

Lithuania

35,3

Canada

32,2

Luxembourg

28,0

Chile

46,5

Mexico*

45,7

Czech

26,2

Netherlands

28,0

Denmark

25,4

New Zealand*

33,3

Estonia

36,1

Norway

25,2

Finland

26,2

Poland

30,0

France

29,4

Portugal

34,2

Germany

29,2

Slovak

26,9

Republic

Republic Greece

34,3

Slovenia

25,5

Hungary*

28,9

Spain

34,6

Iceland

24,4

Sweden

28,1

Ireland

30,9

Switzerland

29,5

Israel

36,0

Turkey

39,3

Italy

32,5

United

35,8

Kingdom Japan*

33,0

United States

39,6

*2012 data Unit 0/1 scale where 0 represents highest equality level Source: OECD Social and Welfare Statistics: Income distribution, 2016

5. Redistributive Transfers in the Welfare State

The seminal contribution in the political economics of the welfare state dates back to mid ‘70s, with the research published by Romer (1975), Roberts (1977), and Meltzer and Richard (1981), which gave rise to a large corpus of literature aimed at linking the redistributive policies to their political rational and more specifically to electoral incentives. The basic idea in these initial contributions is that – in democracies – public policies are chosen by officeseeking policy-makers in an attempt to maximize their probability of winning the election. Politicians thus consider the preferences of a majority of voters, as indicated by the median voter theorem presented in section 2.1, or of the swing voters, as suggested by the probabilistic voting model in section 2.2, and determine their public policy accordingly. This chapter concentrates on a simple economic policy, consisting of an intra-generational income redistribution scheme. Individuals may be levied a proportional tax of their labor income, and these tax revenues may be redistributed to all citizens in a lump-sum fashion, that is, regardless of citizens’ specific characteristics, such as income, age, gender or geographic location. Thus, since every individual contributes in proportion to his income, while the benefits are not related to income, this system results in a redistribution from the rich, who pay large contributions, to the poor, who instead pay low taxes. The political environment is quite simple. A simple majority voting is considered. Since individual (i.e. voters) preferences are single-peaked, the median voter theorem applies. The economic model presented in the next section will illustrate the characteristics of this redistributive policy and the political environment. The crucial insight of this political-economic approach to redistribution relies on the hypothesis of selfish voters; they only care about their own well-being as provided, for instance, by consumption and leisure. Poor voters benefit from, and thus support, large redistributive programs, whereas richer voters, who are net contributors, oppose them.

66

POLITICAL ECONOMICS

Hence, as discussed in detail at the end of this chapter, the size of the welfare state is related to the degree of income inequality in the economy. In democracies where selfish voters determine the dimensions of the redistributive policy, a more unequal distribution of income induces more redistribution. These models propose a specific definition of income inequality: the difference between the median and the mean income in the economy. The main message is that the most relevant group in the political arena is the middle class, that is, the individuals with the median income. Therefore, if the middle class becomes relatively poorer, for example because of an increase in the income of the rich, demand for enlarging the welfare state increases. If the middle class becomes relatively richer, for example because of a reduction in the income of the poor, demand for the welfare state instead decreases. 5.1 The Economic Model To fully understand the features of this simple redistributive system, it is convenient to consider a simple economic model, as used by Meltzer and Richard (1981). The economy is populated by individuals, whose total mass (i.e. the population size) is normalized to one. This economy is static, and hence individuals only live for one period: today. Agents in this economy work and use their labor income, net of taxes and transfers, to purchase and consume the only produced good. In this model, redistribution takes place across income groups, and hence individuals have to differ along this dimension. In particular, individuals are assumed to be heterogeneous with respect to their ability to work, which is indicated with e. This working ability, e, determines the individual productivity and hence the amount of time that he will devote to work, and ultimately his labor income. As it will become clear after examining the optimization problem of the economic agents, there exists a close link between the individual ability type, e, and his labor income, n(e). Individuals abilities vary from the lowest, el, to the highest value, eu. They are distributed on the support [el, eu], with el < 0, and eu > 0, according to a cumulative distribution function G(e). The average ability in society is assumed to 𝑒𝑒 be equal to zero, 𝐸𝐸(𝑒𝑒) = ∫𝑒𝑒 𝑢𝑢 𝑒𝑒 𝑑𝑑𝑑𝑑(𝑒𝑒) = 0. In other words, the individual with 𝑙𝑙 average ability in society has an ability equal to zero, e=0, and a corresponding, typically positive, income n(0)>0. Furthermore, as shown in Figure 1, the distribution of abilities is assumed to be skewed, so that the median ability, eM, is below the average ability: eM < E(e) = 0.

5.

Redistributive Transfers in the Welfare State

67

Figure 1: Distribution of ability in the economy

The total endowment of disposable time that an individual has to divide between labor and leisure depends on the individual’s ability level. The individual time constraint is equal to 𝑙𝑙 + 𝑛𝑛 = 1 + 𝑒𝑒

(1)

𝑈𝑈 = 𝑐𝑐 + 𝑉𝑉(𝑙𝑙)

(2)

𝑐𝑐 = 𝑛𝑛𝑛𝑛(1 − τ) + 𝑇𝑇

(3)

where l is the amount of leisure, n represents the time spent working, the average time endowment in the economy is normalized to 1, and e is the individual’s ability. Individuals value consumption and leisure. Their preferences are summarized by a quasi-linear utility function:

where c denotes consumption, l is leisure. V(l) is a concave function, so that individuals always enjoy an increase in their leisure (or free time), but at a decreasing rate. In other words, the marginal utility of leisure is always positive, but decreasing. This characteristic of the utility function will be relevant in discussing the efficiency cost of taxation (and redistribution). Finally, the consumer budget constraint is given by his net labor income, since every individual will use all of his resources to consume. In fact, as there exists only one period in this economy, no saving will take place:

where n is the amount of labor supplied, w is the wage, which is normalized to one (w=1), τ is the tax rate on the labor income and T is the lump-sum transfer provided to every individual.

68

POLITICAL ECONOMICS

The fiscal policy is thus represented by a pair (τ, T), which every economic agent takes as given, while deciding over his labor supply, n, and hence consumption. As argued in chapter 1, individuals then take political decisions on the size of the redistributive system, by voting on the tax rate, τ. The income redistribution policy is characterized by a tax rate, τ, imposed on every worker’s labor income, and by a lump-sum transfer, T, provided to all individuals. Additionally, given the static nature of the economy, which prevents government borrowing, this redistributive policy is assumed to be budget balanced, that is, the total transfers paid out have to be fully financed by the tax revenues. Hence, the government budget constraint is 𝑒𝑒

𝑇𝑇 = τ 𝐸𝐸�𝑛𝑛(𝑒𝑒)� = τ ∫𝑒𝑒 𝑢𝑢 𝑛𝑛(𝑒𝑒) 𝑑𝑑𝑑𝑑(𝑒𝑒) 𝑙𝑙

(4)

Equation 4 simply states that all tax revenues collected from the labor income are transferred back to individuals in a lump-sum fashion. However, this equation plays a crucial role in determining the individual preferences over the fiscal policy and the individual voting decisions, as it allows the fiscal policy to be entirely summarized by a unique parameter, the tax rate τ. 5.2 The Redistributive System: Winners and Losers To understand this redistributive system, and hence to identify the winners and losers from redistribution, it is convenient to abstract from any economic decisions, and to assume that every individual works full time. In this scenario, leisure is set equal to zero for all individuals, regardless of their ability type, l=0. However, labor supply, and thus labor income, will still depend on individual ability. In fact, by using the time constraint at equation 1, it is easy to see that for ability type-e labor supply becomes equal to 𝑛𝑛(𝑒𝑒) = 1 + 𝑒𝑒. A type-e individual tax burden, that is, the amount of taxes paid by a type-e individual, is hence equal to τ (1 + 𝑒𝑒). In this case, the total amount of fiscal revenues – and hence the transfer – is equal to: 𝑒𝑒

𝑇𝑇 = τ 𝐸𝐸(1 + 𝑒𝑒) = τ ∫𝑒𝑒 𝑢𝑢(1 + 𝑒𝑒) 𝑑𝑑𝑑𝑑(𝑒𝑒) = τ 𝑙𝑙

(5)

In this scenario, as the utility function in equation 2 is linear for consumption and the amount of leisure was assumed to be equal to zero, the total level of consumption obtained by a type-e individual, and hence their utility, is given by the following expression:

5.

Redistributive Transfers in the Welfare State

𝑐𝑐 = (1 − τ) (1 + 𝑒𝑒) + 𝑇𝑇 = (1 − τ)(1 + 𝑒𝑒) + τ = 1 + 𝑒𝑒 − τ𝑒𝑒

69

(6)

The expression above shows that the fiscal policy, as characterized by the tax rate, τ, may have a positive or negative impact on the agent’s utility – respectively increasing or decreasing his consumption – depending on whether the agent’s ability is below or above the average ability in the economy. In fact, for negative ability type (𝑒𝑒 < 0), a positive tax rate increases the agent’s consumption, and viceversa for a positive ability type (𝑒𝑒 > 0). Figure 2 provides a graphic interpretation of this result by separately displaying – for every individual – the tax burden from the transfer. The horizontal axis in Figure 2 measures the ability type, from the lowest ability type in the economy, el, on the extreme left on the axis, to the highest ability type, eu, on the extreme right. Clearly, the lower an individual’s ability – and thus income – the lower his tax burden, as displayed by the continuous line. Due to redistributive nature of the system, the transfer is unrelated to individual income, and is hence depicted as a horizontal (dotted) line. It is now easy to see that the winners from this redistributive policy, that is, those who contribute fewer taxes to the system than they receive in transfers, are located on the left of this graph. Clearly, they are low income individuals, 𝑒𝑒 < 0, for whom a positive tax rate increases the utility, since −τ 𝑒𝑒 > 0 (see equation 6). In fact, the lowest income type, el, enjoys the largest gain from redistribution, as measured by the difference between the dotted line (the transfer) and the continuous line (the taxes). As income increases, however, the gain decreases, as individuals contribute more in tax. In fact, as their ability type turns positive, 𝑒𝑒 > 0, an individual contributes more to the system than he receives, −τ 𝑒𝑒 < 0, and hence becomes a net contributor, that is, a loser in the redistributive game. The losers are therefore the high ability workers, that is, the rich individuals located on the right of Figure 2. According to the indirect utility function defined in equation 6, individual preferences over the tax rate, τ, are clearly single-peaked in the case of no economic decision, since agents are not allowed to choose how much to work. In particular, the preferences of low ability individuals – those with a negative ability type, 𝑒𝑒 < 0 – are increasing in τ, until a maximum at τ = 100%. Instead, the preferences of high ability individuals, 𝑒𝑒 < 0, are decreasing in τ, with the maximum at τ = 0. Therefore, if simple majority voting is the political environment, as discussed in chapter 1, the political equilibrium outcome coincides with the tax rate chosen by the median voter. In this context, the median voter is the agent with the median ability type, 𝑒𝑒𝑀𝑀 < 𝐸𝐸(𝑒𝑒) = 0, who chooses a positive tax rate, specifically, a tax rate equal to 100 %. In this simple economic environment, a sufficiently poor median voter, 𝑒𝑒𝑀𝑀 < 𝐸𝐸(𝑒𝑒) = 0, would thus support full redistribution, that is communism.

70

POLITICAL ECONOMICS

Figure 2: Winners and losers from redistribution

However, the analysis above does not consider the distortionary effect of taxation that a redistributive system introduces. In fact, the tax levied on the individuals income in order to finance the transfer typically introduces a distortion in the labor supply decisions. Instead, under the assumption that all individuals always work full time, the tax rate needed to finance the redistributive system does not modify the agents’ decision about how much to work; thereby leading to an extreme scenario where individuals preferred either no taxation or 100% taxation. Hence, either no redistribution or full expropriation of individuals’ income. If agents are allowed to decide how much to work, taxation will have a distortionary effect on the economy, as the opportunity cost of leisure increases with taxation, and hence individuals prefer to work less for higher tax rates. To see this, consider the economic decision of a type-e agent, who has to determine his optimal level of consumption, c, and of labor supply, n, for a given fiscal policy (τ,T). This optimization problem amounts to maximizing the individual’s utility function in equation 2, with respect to l and c, given the budget constraint in equation 3, and the time constraint in equation 1, for a given fiscal policy, (τ, T). By substituting these constraints in the utility function, the maximization problem can be written as follows: max(1 − τ)(1 + 𝑒𝑒 − 𝑙𝑙) + 𝑇𝑇 + 𝑉𝑉(𝑙𝑙) {𝑙𝑙}

(7)

which yields the following first order condition: 1−τ=

𝜕𝜕𝜕𝜕(𝑙𝑙) 𝜕𝜕𝜕𝜕

(8)

5.

Redistributive Transfers in the Welfare State

71

Equation 8 suggests that individuals maximize their utility by choosing a labor supply so that the marginal utility from leisure – that is, the right hand side of equation 8 – is equal to the marginal utility from the consumption acquired by working one additional unit of time, which is equal to the net wage (as shown on the left hand side of equation 8). The optimal level of labor supply chosen by an individual will have to satisfy this condition, which is shown graphically in Figure 3. The decreasing curve represents the marginal utility of leisure, which is decreasing in the level of leisure; whereas the horizontal line represents the marginal utility of consumption, which is independent of the level of leisure and is equal to 1 − τ. The optimal labor supply is determined by the intersection of these two curves at l*. From the time constraint in equation 1, the optimal labor supply for an agent with ability type-e is: 𝑛𝑛∗ (𝑒𝑒) = 1 + 𝑒𝑒 − 𝑙𝑙 ∗ . Figure 3: Individual labor supply

From Figure 3, it is easy to see that taxation introduces a distortion in the economy by reducing the labor supply and thus the level of economic production. In fact, an increase in the tax rate shifts down the horizontal line 1 – 𝜏𝜏, which hence intersects with the other curve at a higher level of l*. In other words, since the opportunity cost of leisure has decreased due to the increase in the tax rate, individuals will prefer to enjoy more leisure time and to work less. Every individual’s labor supply, n*(e), – and thus also average labor supply in the economy, 𝐸𝐸(𝑛𝑛 ∗ (𝑒𝑒)) – will be reduced. In this new scenario, which includes the distortionary effect of taxation, the government budget constraint has to be re-written to incorporate the individuals’ labor supply decision, and becomes equal to 𝑇𝑇 = 𝜏𝜏 𝐸𝐸(𝑛𝑛∗ (𝑒𝑒)). Under this specification, an increase in the tax rate, τ, gives rise to two distinct ef-

72

POLITICAL ECONOMICS

fects. On the one hand, there is a direct increase in the government fiscal revenues, as every individual is required to contribute a large share of his labor income. On the other hand, however, a higher tax rate induces a distortion in the economy, which decreases fiscal revenues as individuals react to the larger tax burden by working less, and hence reduce the tax base (defined as the total labor income over which taxes may be levied). The combination of these two effects produces the so-called Laffer Curve, which relates the total amount of tax revenues to the tax rate. As shown in Figure 4, for low tax rates, fiscal revenues are positive, yet small. As the tax rate increases, revenues increase, yet at a decreasing rate because the distortionary effect of taxation reduces the total tax base. For a medium tax rate, labeled τL in Figure 4, fiscal revenues reach their maximum and any additional increase in the tax rate produces a reduction in revenue, as the distortionary impact of taxation prevails over the direct effect. For a large level of taxation, the tax base shrinks to the point that fiscal revenues eventually become zero. Figure 4: Laffer Curve, the distortionary effect of taxation

5.3 The Political Decision As in the previous section, individual preferences over the redistributive system, summarized by the tax rate, τ, will depend on how the system affects the economic well-being of the agent. Under this new scenario, however, the entire utility function has to be considered as the trade-off between labor and leisure has been re-introduced.

5.

Redistributive Transfers in the Welfare State

73

Therefore, given the economic decision described in equations 7 and 8, every individual will take his political decision by maximizing the following indirect utility function with respect to the tax rate, τ: (1 − 𝜏𝜏) 𝑛𝑛∗ (𝑒𝑒) + 𝜏𝜏 𝐸𝐸 (𝑛𝑛∗ (𝑒𝑒)) + 𝑉𝑉 (𝑙𝑙 ∗ (𝑒𝑒))

(9)

where the first term refers to the net labor income, the second to the transfer received and the third to the utility associated with leisure. Because of the Envelope theorem, in his political decision, an ability type-e agent will leave aside the effect on 𝑛𝑛∗ (𝑒𝑒) and 𝑙𝑙 ∗ (𝑒𝑒) and optimally trade-off the costs and benefits arising from the impact of the tax rate on the following three elements: (i) the direct, negative cost of taxation – since an increase in τ raises his tax burden by n*(e); (ii) the direct, positive benefit from the transfer – since a rise in τ increases his transfer by 𝐸𝐸(𝑛𝑛∗ (𝑒𝑒)); and (iii) the cost of the distortion introduced by the taxation, since a rise in τ reduces the tax base, and hence the transfer 𝐸𝐸(𝑛𝑛∗ (𝑒𝑒)), as discussed above. As in the previous scenario, winners and losers from the redistributive system can be easily defined by their ability, by using equation 9. Low ability types, 𝑒𝑒 < 0, i.e. low income individuals, gain from redistribution, and hence vote for a positive tax rate, 𝜏𝜏 > 0. High ability types, 𝑒𝑒 > 0, lose and thus oppose redistribution, 𝜏𝜏 = 0. In this case, however, even very low ability types do not vote for full expropriation - that is, for 𝜏𝜏 = 100% - as, due to the distortionary effect of taxation on the tax base, very high taxes do not lead to larger transfers. Again, since preferences are single-peaked, the median voter’s theorem applies, and the political equilibrium tax rate corresponds the most preferred tax rate of the agent with median working ability, 𝑒𝑒𝑀𝑀 . Since the ability distribution was assumed to be skewed, in order to resemble actual labor income distribution, the median voter has a lower ability (and income) than the average ability (and income) in the economy – that is 𝑒𝑒𝑀𝑀 < 0, and 𝑛𝑛∗ (𝑒𝑒𝑀𝑀 ) < 𝐸𝐸(𝑛𝑛∗ (𝑒𝑒)). Hence, the median voter supports a positive tax rate, and thus some level of redistribution. 5.4 Discussion According to this model, the amount of redistribution in the economy depends crucially on the degree of income inequality, as measured by the difference between the median and the average income. More income inequality thus leads to more redistribution, since the pivotal (median) voter becomes relatively poorer vis-à-vis the average income, and hence benefits more from redistribution.

74

POLITICAL ECONOMICS

However, if inequality is not measured as the difference between the median and the average income, but rather using a more conventional measure such as the Gini coefficient (see Table 7 in chapter 4 -, greater inequality does not to necessarily lead to more redistribution. To see this, consider two possible increases in income inequality. First, when rich individuals become richer, income inequality – as measured by the Gini coefficient – increases. The average income in the economy also increases, while the median income does not to vary. The difference between median and average income widens, thus leading to more political demand for redistribution, as the median voter becomes relatively poorer. Second, consider the case of poor individuals becoming poorer. In this situation, inequality – as measured by the Gini coefficient – still increases, yet the difference between the median and the average income shrinks, as the former does not change while the latter decreases. The median voter is thus relatively richer, and requires less – rather than more – redistribution. Since its initial development, several attempts have been made at examining how well the positive theory of redistribution explains the data (see Milanovic, 1999). Two crucial questions in this respect are the following: can this theory explain the large cross-country difference in income redistribution, with Scandinavian countries implementing large redistributive programs and Anglo-Saxon countries limiting redistribution to a safety net consisting of meanstested schemes? Can the theory account for the dynamics of Welfare State expenditure, and its large expansion since World War II? To some extent, the theory of redistribution successfully explains the early growth of the welfare state, if combined with two additional features. First, the extension of voting rights to include poor voters - which decreased the median ability among voters without affecting the average income among the citizens – has, according to the model, led to more redistribution. Second, the reduction in the cost of collecting taxes has also led to more redistribution by decreasing the distortionary effects of taxation. Nevertheless, the theory of redistribution is less successful in explaining the expansion in welfare state expenditure from World War II until the 1990s and the large cross-country differences. Part of this lack of success may be due to the fact that intragenerational redistribution, of the type analyzed in this chapter, is not the most relevant kind of redistribution. Indeed, income is not the only source of difference among individuals, and modern welfare systems seem to channel redistribution more through pension systems - which privilege the age component - than through other programs, as discussed in chapter 4. Another reason for this partial failure may be attributed to the stationarity of the model examined here. In reality, voting does not occur only once, and welfare systems and taxation vary over time. Moreover, taxation has dynamic effects on capital accumulation and economic growth. Along these lines, for instance, Krusell and Rios-Rull (1999) embed this static redistributive

5.

Redistributive Transfers in the Welfare State

75

model in a more dynamic economic and political environment to show that “dynamic distortions” lead to lower political demand for redistribution and the welfare state. The critique of the model’s stationarity can also be interpreted in a completely different way. It could be argued that, while welfare state programs are strongly persistent over time, 1 individuals’ economic and social conditions may change during their life-time. Hence, poor individuals may manage to increase their income, while rich agents may find themselves in poor economic conditions. As soon as the welfare state is persistent over time, while the winners and losers may change, individuals have to trade-off their current status, with the probability of finding themselves in a different socio-economic situation in the future. 2 This will soften the demand for welfare among the lowincome individuals, who take into account the possibility of having to pay into the system if they become rich. Analogously, the preferences of the current rich is modified, as they realize that – despite being current loser from the welfare state – they may become winners, if their economic conditions become worse in the future. Hence, different perceptions about social mobility across individuals or countries may lead to different demands for redistribution (see Alesina et al. 2016). Recent research has further examined this element of intertemporal insurance to understand the difference in the preferences for redistribution between the US and (most) European countries. According to an argument made by Alesina and Angeletos (2005), individuals in different countries attribute their (socio-economic) success in (professional) life to different reasons. In the US, most individuals credit talent and hard work for professional success. By isolating an aspect – hard work – which is under the direct control of the individuals, this view limits the need for insurance against the risk of becoming poor – and thus reduces the overall demand for redistribution. In other (European) countries, individuals tend to ascribe professional success to luck or family connections, and thus to factors which are not under the direct control of the individuals. In this case, individuals may be more interested in obtaining insurance against bad professional luck (or lack of good family connections), and thus demand more redistribution. Finally, the median voter theory hinges crucially on the hypothesis that individuals behave selfishly, as they only care about their own economic wellbeing. Hence, poor voters support large redistributive programs, as they benefit, whereas richer voters, who are net contributors to these programs, oppose them. In reality, we observe at least some individuals behaving unselfishly, as several people – for instance – contribute to private charity. This behavior is con1 This policy persistence is itself very relevant from a political-economic perspective, as it asks the obvious question of why these programs are persistent, even though they could be modified over time by using the same procedure with which they were introduced. 2 This dynamic scheme is discussed in chapter 7, when we analyze the labor market institutions.

76

POLITICAL ECONOMICS

sistent with altruistic preferences: people may want to “feel good” about giving to others or they may want to reduce inequality in society. In the latter case, these individuals may prefer to encourage a public income redistribution program that obliges all citizens – not just those willing to give to charity – to contribute. As argued in Galasso (2003), however, in an economy populated with a majority of selfish individuals, who only value their own well-being, and some unselfish voters, who oppose inequality in the society, the message that more inequality leads to more redistribution may be strengthened. In fact, the existence of fair (or unselfish) voters limits the political relevance of the middle class. For instance, if the distance between the poor and the middle class increases because poor individuals become poorer, the level of redistribution does not necessarily decrease, as predicted in the standard version of the model. The increased poverty may push the inequality averse, fair agents to support more redistribution, which may balance the opposite voting behavior of the selfish voters. Finally, an empirical study by Corneo and Gruner (2002), based on survey data from twelve countries, shows that agents may selfishly pursue other personal – yet non-economic – goals, such as maintaining or improving their social status. Moreover, individuals may support redistributive programs that reduce their private economic and social well-being because of “public values”. In particular, Corneo and Gruner (2002) find that agents are more favorable to redistributive policies if they believe that the current income distribution in the society is determined more by nepotism (e.g., family background) than by meritocratic values (individual ability or effort).

6. Pensions

All developed countries have a pension system. An overwhelming majority of these systems are unfunded (or pay-as-you-go, PAYG). In these PAYG systems, the revenues obtained through payroll tax levied on current workers’ labor income are immediately used to provide a pension benefit to current retirees. Typically, these retirees are former workers, who have at least reached the minimum retirement age and have an entitlement to an old-age pension. The inter-generational links embedded in the working of a PAYG system are presented graphically at Figure 1. Figure 1: Pay-as-you-go pension systems

78

POLITICAL ECONOMICS

The size of these unfunded pension systems has increased over the last few decades. In 1985, they absorbed 6% of GDP in the US, 10.8% in Italy, 10.3% in France, and 7.2% in Sweden. In 2011, pension spending reached 15.8% of GDP in Italy and 13.8% in France. As discussed in chapter 4, there are several economic reasons behind the introduction of pension systems, and thus a wide body of literature focusing on the economics of pensions has emerged. 1 However, decisions about pension policies, from the initial design of the system to later modifications, often go beyond economic theory into the realm of politics. To address this aspect, we consider political economy models that analyze the institution of the unfunded pension systems and their development into the most widespread instrument of social insurance. The challenge is to understand why pension programs that transfer resources from a majority of young and middle-aged workers to a minority of elderly individuals exist. What are the benefits for these young and middleaged workers from a PAYG pension system? To answer this question, let us consider the impact of a mandatory pension system on these individuals. The main, almost mechanical, economic effect is to tilt the net income profile of each individual towards fewer resources in youth and more in old age. In fact, pension contributions decrease net labor income during an individual’s working life, while pension benefits provide additional income in retirement. Clearly, this purely mechanical shift induces behavioral changes, as individuals modify their economic decisions – such as their saving behavior or their labor supply. Since a pension system constitutes a forced saving mechanism, which moves individual resources from youth and adulthood to old age, the impact of such a system on individual economic well-being can be assessed by considering the returns that pensions provide, relative to alternative assets. In the absence of a PAYG pension system, individuals would save for old age consumption, using alternative instruments available on the financial markets. Hence, each individual can evaluate the economic convenience of a PAYG system by comparing its internal rate of return with the returns that can be obtained on alternative assets. Along these lines, five major economic reasons for young and/or middle-aged agents to support pension systems can be identified: (i) dynamic inefficiency (Aaron, 1966); (ii) a reduced time horizon in evaluating the pension program (Browning, 1975); (iii) the crowding-out of aggregate savings by the pension system (Cooley and Soares, 1999); (iv) the within-cohort redistribution element shared by many pension schemes (Tabellini, 2000); and (v) an altruistic motive (Hansson and Stuart, 1989). The agents’ individual preferences over the pension system induced by these economic factors can be aggregated through several political mechanisms: majority voting, veto-power or constitutional rules, and interest-group 1

Useful references are Diamond (1996), Barr and Diamond (2010) and Blake (2006).

6.

Pensions

79

models. This chapter combines two lines of classification—economic factors and political institutions—to analyze why pension programs that transfer resources from young and middle-aged to retirees exist. 6.1 The Economic Environment of the Voting Models In this section, we introduce a simple yet quite general economic environment in order to examine some of the economic factors that may lead young and middle-aged individuals to support unfunded pension systems. This setting abstracts from altruistic motives. In Section 6.2.5, we discuss the relationship between altruistic preferences and the sustainability of unfunded systems (see Hansson and Stuart, 1989, and Tabellini, 2000). We consider an overlapping generations model with capital accumulation in a closed economy. In every period, three generations are alive. We call them ‘‘Young’’, ‘‘Middle-Aged’’, and ‘‘Old’’. Population grows at a constant non-negative rate µ. It follows that in any period for every ‘old’ there are (1 + µ) ‘middle-aged’ individuals and (1 + µ)2 ‘young’. As in previous chapter, we assume that individuals differ in their ability e, which is distributed on the support [el, eu], with el < 0, and eu > 0, according to a cumulative distribution function G(e). The average ability in society is as𝑒𝑒 sumed to be equal to zero, 𝐸𝐸(𝑒𝑒) = ∫𝑒𝑒 𝑢𝑢 𝑒𝑒 𝑑𝑑𝑑𝑑(𝑒𝑒) = 0, and the median ability, eM, 𝑙𝑙 to be below the average ability: eM < E(e) = 0. In every working period, the individual time constraint is thus equal to 𝑡𝑡 𝑡𝑡 𝑙𝑙𝑡𝑡𝑡𝑡 + 𝑛𝑛𝑡𝑡𝑡𝑡 = 𝑙𝑙𝑡𝑡+1 + 𝑛𝑛𝑡𝑡+1 = 1 + 𝑒𝑒

(1)

where l is the amount of leisure, n represents the time spent working, the average time endowment in the economy is normalized to 1, and e is the individual ability. Moreover, subscripts indicate calendar time and superscripts indicate the period when the agent was born. Retirement is mandatory in old age. Non-altruistic agents value leisure, 𝑙𝑙, and consumption, c, according to a time separable utility function: 𝑡𝑡 𝑡𝑡 ) 𝑡𝑡 ) 𝑈𝑈(𝑙𝑙𝑡𝑡𝑡𝑡 , 𝑐𝑐𝑡𝑡𝑡𝑡 ) + β𝑈𝑈(𝑙𝑙𝑡𝑡+1 , 𝑐𝑐𝑡𝑡+1 + β2 𝑈𝑈(𝑐𝑐𝑡𝑡+2

(2)

where β ∈ ℜ+ represents the individual time discount factor, and 𝑈𝑈 ′ > 0 and 𝑈𝑈 ′′ < 0, so that individuals value leisure and consumption, but at a decreasing rate. The budget constraints of a type 𝑒𝑒𝑡𝑡 individual born at time 𝑡𝑡 during his lifetime are: 𝑐𝑐𝑡𝑡𝑡𝑡 + 𝑎𝑎𝑡𝑡𝑡𝑡 = 𝑤𝑤𝑡𝑡 (1 − τ𝑡𝑡 )𝑛𝑛𝑡𝑡𝑡𝑡

80

POLITICAL ECONOMICS

𝑡𝑡 𝑡𝑡 𝑡𝑡 𝑐𝑐𝑡𝑡+1 + 𝑎𝑎𝑡𝑡+1 = 𝑤𝑤𝑡𝑡+1 (1 − τ𝑡𝑡+1 )𝑛𝑛𝑡𝑡+1 + 𝑎𝑎𝑡𝑡𝑡𝑡 (1 + 𝑟𝑟𝑡𝑡+1 )

(3)

𝑡𝑡 𝑡𝑡 (1 = 𝑎𝑎𝑡𝑡+1 𝑐𝑐𝑡𝑡+2 + 𝑟𝑟𝑡𝑡+2 ) + 𝑃𝑃𝑡𝑡+2

𝑡𝑡 where 𝑎𝑎𝑡𝑡+1 represents asset holding at the beginning of period 𝑡𝑡 + 1 for an individual born at time 𝑡𝑡, and 𝑟𝑟𝑡𝑡 , 𝑤𝑤𝑡𝑡 , τ𝑡𝑡 and 𝑃𝑃𝑡𝑡 are, respectively, the real interest rate, the wage rate, the payroll tax rate and the pension transfer at time 𝑡𝑡. These period budget constraints can be used to obtain a life-time budget constraint, in which the discounted life-time consumption (on the left hand side) is equal to the discounted life-time net income (on the right hand side): 𝑐𝑐 𝑡𝑡

𝑐𝑐 𝑡𝑡

𝑡𝑡+1 𝑡𝑡+2 𝑐𝑐𝑡𝑡𝑡𝑡 + 1+𝑟𝑟 + (1+𝑟𝑟 = 𝑤𝑤𝑡𝑡 (1 − τ𝑡𝑡 )𝑛𝑛𝑡𝑡𝑡𝑡 + )2

𝑡𝑡 𝑤𝑤𝑡𝑡+1 (1−τ𝑡𝑡+1 )𝑛𝑛𝑡𝑡+1

1+𝑟𝑟

𝑃𝑃

𝑡𝑡+2 + (1+𝑟𝑟 )2

(4)

Agents maximize the utility at equation (2), with respect to asset holdings and labor supply, subject to the budget constraints at equation (3), and to the time constraints at equation (1), and taking the pension system as given. The pension system imposes a payroll tax, τ𝑡𝑡 , on the labor earnings of active generations – young and middle-aged – and transfers the collected amount of resources lump sum to the current retirees, 𝑃𝑃𝑡𝑡 . We consider that the pension budget is balanced in every period. Thus, the pension transfer to retirees, 𝑃𝑃𝑡𝑡 , is equal to the payroll tax rate, τ𝑡𝑡 , multiplied by the average labor earn𝑛𝑛𝑡𝑡 , where the average labor supply in the economy, ings in the economy, 𝑤𝑤𝑡𝑡 ��� ���𝑡𝑡 and of the middle���, 𝑛𝑛𝑡𝑡 depends on the average labor supply of the young, 𝑛𝑛 𝑡𝑡 aged workers, ������ 𝑛𝑛𝑡𝑡𝑡𝑡−1 . Hence, the pension system’s (balanced) budget constraint is: 𝑡𝑡−1 ������ 𝑃𝑃𝑡𝑡 = τ𝑡𝑡 𝑤𝑤𝑡𝑡 ��� 𝑛𝑛𝑡𝑡 = τ𝑡𝑡 𝑤𝑤𝑡𝑡 �(1 + µ)2 ��� 𝑛𝑛𝑡𝑡𝑡𝑡 + (1 + µ)𝑛𝑛 𝑡𝑡 �

(5)

An increase in the tax rate directly increases pension benefits. However, it typ���𝑡𝑡 /𝜕𝜕τ𝑡𝑡 < 0, due to the distortionically decreases the average labor supply, 𝜕𝜕𝑛𝑛 ary effects of taxation discussed in chapter 5. The pension system entails an element of within cohort redistribution. This feature is examined in section 6.2.3, but it is disregarded in all other sections by assuming homogeneity in abilities across agents. On the production side, a constant-returns-to-scale production function uses capital and labor, measured in efficiency units, to produce units of a consumption good. Total capital per capita in the economy is obtained by aggregating the net savings, or end-of-period asset holdings among generations. From the profit maximization problem of the competitive firms and the equilibrium conditions, factors of production are paid their marginal products. It is useful to assume that the economic productivity grows at an exogenous rate, g. Hence, wages will also grow at the same rate: 𝑤𝑤𝑡𝑡+1 = (1 + 𝑔𝑔)𝑤𝑤𝑡𝑡 .

6.

Pensions

81

The definition of economic equilibrium is standard. For a given sequence of pension tax rates, an economic equilibrium is a sequence of allocations and prices, such that, in every period, the consumers problem is solved for each type-e individual in each generation, firms maximize profits, the pension budget constraint is balanced, and labor, capital and goods markets clear. For a given sequence of pension tax rates, we identify the utility level obtained in an economic equilibrium by every agent with the agent’s indirect utility function. At time 𝑡𝑡, the preference relations over current (and future) tax rates are characterized by the following indirect utility functions: 𝑢𝑢𝑡𝑡𝑡𝑡 (𝑒𝑒; τ𝑡𝑡 , τ𝑡𝑡+1 , τ𝑡𝑡+2 ) for type-e young, 𝑢𝑢𝑡𝑡𝑡𝑡−1 (𝑒𝑒; τ𝑡𝑡−1 , τ𝑡𝑡 , τ𝑡𝑡+1 ) for type-e middleaged, and 𝑢𝑢𝑡𝑡𝑡𝑡−2 (𝑒𝑒; τ𝑡𝑡−2 , τ𝑡𝑡−1 , τ𝑡𝑡 ) for type-e old agent. 6.2 Individual Preferences over Pension Systems To examine the different economic factors that shape individual preferences, we focus on constant sequences of pension tax rates,τ𝑡𝑡 = τ ∀𝑡𝑡. This amounts to assume that young and middle-aged agents form their preferences about the pension tax rate under the assumption that the current scheme will not be modified in the future. Young individuals, who do not expect the system to be in place in their old age, perceive the current tax rate as a net cost, and generally would not be willing to support the system. 2 In the next section, we shall discuss the political arrangements under which a sequence of positive tax rates may arise. Obviously, elderly individuals support pension systems, which award them a pension with no current cost. When do young and/or middle-aged agents favor a PAYG pension system? A marginal increase in the (constant sequence of) tax rates has three effects on the indirect utility function of a type-e young individual, through (i) a direct increase in the labor income tax in the two working periods, (ii) a net increase in the pension transfer, and (iii) a variation in factor prices, due to changes in the stock of capital and in the average labor supply. The same forces affect a type-e middle-aged individual, but over a reduced time horizon. We can now analyze the previously listed economic factors that may justify the support for the pension system. 6.2.1 Dynamic inefficiency Pension systems can improve the welfare of every individual if the economy is dynamically inefficient, that is, if the implicit rate of return from pensions is larger than the real rate of return from capital accumulation (as initially arThe exception is Tabellini (2000), see Section 6.1.5, in which young agents support the pension system due to their weak altruism.

2

82

POLITICAL ECONOMICS

gued in Samuelson, 1958, and Aaron, 1966). To see this, assume that there is no heterogeneity – so that we can normalize e=0. If we concentrate only on the discounted life-time net income (on the right hand side of equation 4), and disregard the effects of taxation on individual labor supply, n, the effect of pension contributions and benefits on the discounted life-time net income can be summarized in the next equation: −𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡 −

𝑡𝑡 𝑤𝑤𝑡𝑡+1 𝑛𝑛𝑡𝑡+1

1+𝑟𝑟

+

𝑛𝑛 𝜕𝜕�������� τ𝑤𝑤𝑡𝑡+2 � 𝜕𝜕𝑡𝑡+2 � 𝑤𝑤𝑡𝑡+2 ������� 𝑛𝑛𝑡𝑡+2 τ + 2 2 (1+𝑟𝑟) (1+𝑟𝑟)

+

(1+𝑔𝑔)2 [(1+µ)2 +(1+µ)] � (1+𝑖𝑖)2

(6)

The first two (negative) terms represent the cost of increasing taxation, the third term is the increase in pension benefits due to higher taxation, while the fourth term represents the distortionary effect of taxation on the average income in the economy, which is negative, since 𝜕𝜕𝑛𝑛 ������/ 𝑡𝑡+2 𝜕𝜕τ < 0. Equation 6 contains all the crucial elements to understand whether the discounted life-time net income increases or decreases when the contribution rate increases, and thus whether young individuals are willing to provide political support to a PAYG pension system. To simplify the analysis, consider that the individual labor supply in youth and in adulthood is the same, and that it is the same among all workers, 𝑡𝑡 𝑡𝑡+1 since there is no heterogeneity in ability. Then we have 𝑛𝑛𝑡𝑡𝑡𝑡 =𝑛𝑛𝑡𝑡+1 = 𝑛𝑛𝑡𝑡+2 = 𝑡𝑡+2 𝑡𝑡+2 2 ) 𝑛𝑛𝑡𝑡+2 , and the average labor supply at t+2 becomes: 𝑛𝑛 ������ = 𝑛𝑛 [(1 + µ + 𝑡𝑡+2 𝑡𝑡+2 (1 + µ)]. Moreover, recall that the productivity grows at a rate g, so that 𝑤𝑤𝑡𝑡+2 = (1 + 𝑔𝑔)𝑤𝑤𝑡𝑡+1 = (1 + 𝑔𝑔)2 𝑤𝑤𝑡𝑡 . If we abstract from the distortionary effect of taxation on the average income in the economy (the forth term in equation 6), we can rewrite equation 6 in order to calculate the internal rate of return from the (outgoing and incoming) embedded monetary flows in the pension system – namely, contributions and pension benefits. The internal rate of return, i, is the return that equalizes the discounted values of contributions to the discounted value of the pension benefits from the following equation: 𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡 �−1 −

Simple algebra shows that

(1+𝑔𝑔)

1+𝑖𝑖

(1 + 𝑖𝑖) = (1 + µ)(1 + 𝑔𝑔)

=0

(7)

(8)

Hence, the return provided by a PAYG pension system to a young individual, who contributes a proportion of his labor income to the system during the working life, and receives a pension at retirement, depends on the rate of economic growth, g, and on the population growth, µ. Whether a young individu-

6.

Pensions

83

al prefers to save through a PAYG pension system or through private savings thus depends on a comparison between the two rate of returns, respectively, (1 + µ)(1 + 𝑔𝑔), for the PAYG pension system and (1+r) for the private savings, where r represents a market return on private capital. In a dynamically inefficient economy, we have (1 + µ)(1 + 𝑔𝑔) > 1 + 𝑟𝑟. Hence, there exist and economic convenience to introduce a pension system, and young individuals would support it and be willing to pay current contributions with the expectation of receiving future benefits. However, explaining support for unfunded systems amongst the young by relying only on this economic element is not too appealing. In the past, some countries may have had periods of dynamical inefficiency. This may occur because of a complete lack of financial markets, so that saving was not possible. Or it may happen in periods, such as after World War II, when economic growth and fertility rates were both high. However, in normal times, empirical evidence seems to reject the idea that economies are dynamically inefficient. 6.2.2 Reduced time horizon Even in a dynamically efficient economy, however, as initially suggested by Browning (1975), middle-aged individuals may choose a positive level of social security, since they only value current and future contributions to, and benefits from, the system, whereas past contributions are a sunk cost. In other words, middle-aged individuals do not take into account the entire cost of social security, since they only consider a reduced time horizon. To isolate this effect, consider a dynamically efficient economy, with no heterogeneity, 𝑒𝑒 = 0. Again, let us abstract from the distortionary cost of taxation. Then the implicit return factor from a PAYG pension system for a middle-aged person who considers previous contributions as a sunk cost, denoted by 1 + 𝑖𝑖 𝑀𝑀𝑀𝑀 , equalizes the discounted values of contributions to the discounted value of the pension benefits from the following equation:

It is easy to see that

𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡 �−1 +

(1+𝑔𝑔)[(1+µ)2 +(1+µ)] 1+𝑖𝑖 𝑀𝑀𝑀𝑀

�=0

(1 + 𝑖𝑖 𝑀𝑀𝑀𝑀 ) = (1 + µ)(2 + µ)(1 + 𝑔𝑔)

(9)

(10)

Thus, a middle-aged individual favors a positive amount of social security, τ > 0, if the implicit return from the system (calculated on the reduced time horizon) is larger than the real return on capital accumulation, which may occur in a dynamically efficient economy if (1 + µ)(2 + µ)(1 + 𝑔𝑔) > 1 + 𝑟𝑟 > (1 + µ)(1 + 𝑔𝑔) .

84

POLITICAL ECONOMICS

6.2.3 Within cohort redistribution Many studies have suggested that young individuals may support a PAYG pension system because of within-cohort redistributive elements. For instance, Boskin et al. (1987) have shown that the US pension system redistributes within cohorts across different family types, by yielding higher returns to lowincome rather than high-income individuals. A direct consequence of this intragenerational redistribution is that, even in a dynamically efficient economy, pension may be more profitable than capital accumulation for low income individuals. To see this, we need to re-introduce heterogeneity into the model so that the individual labor supply 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒) depends on e. However, to keep the analysis simple we consider a two period overlapping generation model, so that there are only young workers and retirees. As in the previous two examples, we concentrate only on the discounted life-time net income and disregard the effects of taxation on the individual labor supply, n(𝑒𝑒). The effect of pension contributions and benefits on the discounted life-time net income can be summarized in the next equation, which contains only three terms as we refer to a model with two generations: 𝑛𝑛 𝜕𝜕�������� τ𝑤𝑤𝑡𝑡+1 � 𝜕𝜕𝑡𝑡+1 � 𝑤𝑤𝑡𝑡+1 ������� 𝑛𝑛𝑡𝑡+1 τ 𝑡𝑡 (𝑒𝑒) −𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡 + 1+𝑟𝑟 + 1+𝑟𝑟

(11)

If we abstract from the distortionary effect of taxation on the average income in the economy, we can rewrite equation 11 in order to calculate the internal rate of return from the pension system for a type-e young individual. The internal rate of return, ie, is the return that equalizes the discounted values of contributions to the discounted value of the pension benefits from the following equation: 𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒) �−1 −

(1+𝑔𝑔)(1+µ) ������� 𝑛𝑛𝑡𝑡+1 �=0 1+𝑖𝑖𝑒𝑒 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒)

(12)

Hence, the return for a type-e young individual is

1 + 𝑖𝑖 𝑒𝑒 = (1 + µ)(1 + 𝑔𝑔)

������� 𝑛𝑛 𝑡𝑡+1 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒)

(13)

Equation 13 shows that the return for a low ability type-e individual, who has 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒) < 𝑛𝑛 ������, 𝑡𝑡+1 is larger than the average return, (1 + µ)(1 + 𝑔𝑔). Hence, even in a dynamically efficient economy, low ability young individuals may support a ������� 𝑛𝑛 , due to the pension system, if (1 + µ)(1 + 𝑔𝑔) < 1 + 𝑟𝑟 < (1 + µ)(1 + 𝑔𝑔) 𝑛𝑛𝑡𝑡𝑡𝑡+1 (𝑒𝑒) 𝑡𝑡

redistributive nature of the system. On the contrary, in a dynamically efficient

6.

Pensions

85

economy, high ability type-e individuals, with 𝑛𝑛𝑡𝑡𝑡𝑡 (𝑒𝑒) > 𝑛𝑛 ������, 𝑡𝑡+1 will be even less willing to support a pension system, because they are net contributors. These models typically suggest that in a Beveridgean system, in which the intragenerational redistribution component is large, more income inequality leads to more social security. 6.2.4 Crowding out The existence of an unfunded pension system has general equilibrium effects on the economy, which may give individuals additional reasons to support pensions. Some studies (Cooley and Soares, 1999, and Boldrin and Rustichini, 2000) argue that the existence of intergenerational redistribution schemes, such as PAYG pension systems, tends to crowd out capital, thus reducing real wages and increases real returns to capital. This mechanisms creates redistribution in favor of assets-holders (‘‘capitalist’’) and against individuals who rely on labor income (‘‘workers’’). To understand this effect, consider a two-period version of our dynamically efficient economy, with no heterogeneity, 𝑒𝑒 = 0, and with exogenous labor supply, 𝑛𝑛𝑡𝑡𝑡𝑡 . 3 Moreover, assume that the utility function is logarithmic. Then equation 2 becomes: 𝑡𝑡 ) 𝑙𝑙𝑙𝑙(𝑐𝑐𝑡𝑡𝑡𝑡 ) + β𝑙𝑙𝑙𝑙(𝑐𝑐𝑡𝑡+1

(14)

and the budget constraints are:

𝑡𝑡 = 𝑤𝑤𝑡𝑡 (1 − τ𝑡𝑡 )𝑛𝑛𝑡𝑡𝑡𝑡 𝑐𝑐𝑡𝑡𝑡𝑡 + 𝑎𝑎𝑡𝑡+1

𝑡𝑡 𝑡𝑡 𝑐𝑐𝑡𝑡+1 = 𝑎𝑎𝑡𝑡+1 (1 + 𝑟𝑟𝑡𝑡+1 ) + 𝑃𝑃𝑡𝑡+2

(15)

Let us consider the economic optimization, before turning to the political aspects. Maximizing equation 14 with respect to the asset holdings, 𝑎𝑎𝑡𝑡𝑡𝑡 , gives 𝑡𝑡 raise to the common Euler equation, 𝑐𝑐𝑡𝑡+1 = 𝑐𝑐𝑡𝑡𝑡𝑡 β(1 + 𝑟𝑟𝑡𝑡+1 ), which describes the individual optimal level of current and future consumption as a function of the subjective time discount, β, and of the (endogenous) market return of the asset, 𝑟𝑟𝑡𝑡+1 . Substituting the budget constraints of equation 15 into the Euler equation, we can obtain the optimal asset holdings, 𝑎𝑎𝑡𝑡𝑡𝑡 , which correspond to the optimal savings of the young: β 𝑃𝑃𝑡𝑡+1 𝑎𝑎𝑡𝑡𝑡𝑡 = 𝑤𝑤𝑡𝑡 (1 − τ𝑡𝑡 ) − (1+ )(1+𝑟𝑟 (16) 1+β

β

𝑡𝑡+1 )

By assuming exogenous labor supply, we disregard the distortionary effects of the taxation needed to finance the pensions.

3

86

POLITICAL ECONOMICS

Equation 16 can be used to understand the effect of the exogenous pension policy (τ𝑡𝑡 , 𝑃𝑃𝑡𝑡+1 ) on the individual economic decision. In particular, an increase in the contribution rate, τ𝑡𝑡 , and/or in the pension benefit, 𝑃𝑃𝑡𝑡+1 , reduces individual savings, and thus the total savings in the economy. This aspect is crucial in individuals’ political decisions. In fact, when deciding over the pension policy, the individuals will take into account that this policy – represented by τ𝑡𝑡 and 𝑃𝑃𝑡𝑡+1 – affects the overall savings in the economy, and thus also the stock of capital. Even more importantly, from an individual perspective, the change in the stock of capital induced by the pension system, affects wages, 𝑤𝑤𝑡𝑡 , and the rate of interest, 𝑟𝑟𝑡𝑡+1 . This represents an additional channel through which the pension system affects the utility of individuals. Before turning to the political decisions of young individuals, it is convenient to show the pension system’s (balanced) budget constraint: 𝑃𝑃𝑡𝑡 = τ𝑡𝑡 (1 + µ)𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡

(17)

The effect of pension contributions and benefits on the discounted life-time net income can be summarized in the next equation, which contains new additional terms, even though we refer to a model with two generations: −𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡𝑡𝑡 +

𝑡𝑡+1 (1+µ)(1+𝑔𝑔)𝑤𝑤𝑡𝑡 𝑛𝑛𝑡𝑡+1

1+𝑟𝑟

+

�𝑎𝑎𝑡𝑡𝑡𝑡

𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 + � 𝜕𝜕𝑘𝑘 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕τ

1+𝑟𝑟

(18)

where the sum of the first two terms is negative in the case of dynamic efficiency, and the third term represents the overall general equilibrium effect. This is the sum of the positive effect of an increase in the tax rate on the returns of young individual’s assets, and the negative impact on their future pension due to the decrease in the wages. In fact, an increase in the tax rate – and the corresponding increase in the pension benefit due to the budget constraint in equation 18 – reduces aggregate savings, and hence the stock of capital, k, in the economy – that is, (𝜕𝜕𝜕𝜕/𝜕𝜕τ) < 0. A lower stock of capital increases the marginal product of capital, which has become a scarcer resource, and thus the rate of return increases, (𝜕𝜕𝜕𝜕/𝜕𝜕𝜕𝜕) < 0. On the other hand, less capital reduces the marginal product of labor and hence wages, (𝜕𝜕𝜕𝜕/𝜕𝜕𝜕𝜕) > 0. In the end, with large asset holdings, 𝑎𝑎𝑡𝑡𝑡𝑡 , and a large enough crowding out effect, (𝜕𝜕𝜕𝜕/𝜕𝜕𝜕𝜕) (𝜕𝜕𝜕𝜕/ 𝜕𝜕τ)>0, a young individual may be willing to use a pension system, even though its internal rate of return is lower than the alternative asset, in order to increase these returns on private assets.

6.

Pensions

87

6.2.5 Altruism Young agents may also have altruistic preferences towards the old. Hansson and Stuart (1989) suggest that young agents, in choosing their current savings, take into account the behavior of altruistic young individuals in the future. In doing this, agents recognize that, if they have not saved enough for old-age consumption, the young in the future will be willing to provide them with an old-age transfer. This Stackelberg behavior leads to an inefficient allocation of resources. In fact, agents rationally ‘‘undersave’’ to induce the young in the future to grant them a transfer. However, every individual would benefit from shifting private resources from consumption in youth to old age consumption. In this context, a pension system would arise under unanimity rule to improve allocation by shifting resources into the future. Weakly altruistic preferences can also be combined with intragenerational redistribution. In a model introduced by Tabellini (2000), heterogeneous, altruistic agents vote every period on the current pension level, which they believe to be unrelated to any future benefit. Since their altruism is weak, young individuals are not willing to provide a private transfer to the elderly. However, the low-income young may favor a positive pension level, since the utility from their parents receiving a pension outweighs the direct (utility) cost of the tax. 6.3 Voting on Pension Systems When voting on pension systems, individuals also vote on pension tax rates, and the policy outcome corresponds to the tax rate that receives the majority of votes. The tax rate then also determines the pension benefit, according to the pension system budget constraints in equations 5 and 17 –for the model with three and two generations, respectively. Selfish young and middle-aged agents are not willing to sustain a pension system unless they expect it to remain in place until their old age. Early models, such as Browning (1975), disregarded this problem by considering a once-and-for-all election. This is equivalent to assuming that there exists full commitment over future policies, and that future reforms of the system, even Pareto improving (i.e., beneficial to individuals of all generations) are not feasible. It is clearly more realistic to consider that elections take place every few years, and that previous policies can be changed at zero (or at low) cost (see Sjoblom, 1985). This scenario is more difficult, but can be modelled using a simple, yet quite intriguing idea. The voting game gives rise to a social contract, which implicitly defines a system of rewards and punishments. Young voters may agree to transfer resources to current retirees because they expect to be rewarded with a corresponding transfer in their old age. Failures to com-

88

POLITICAL ECONOMICS

ply with the contract, and therefore to provide pensions to current retirees, are punished with no transfers in old-age. These contracts can be enforced in an overlapping generations model by a sequence of trigger strategies. To understand this, it is useful to introduce a simple majority voting game over pension tax rates. We use the two-period version of the overlapping generations model described in the previous section with no heterogeneity. The players in this voting game are all agents alive at every election. For each generation at time t, we identify a representative play𝑦𝑦 er, young and old. An action at time t for a young player is a tax rate, 𝑎𝑎𝑡𝑡 ∈ [0,1], and analogously for an old player, 𝑎𝑎𝑡𝑡𝑜𝑜 ∈ [0,1]. At time 𝑡𝑡, the public history of the game is given by the sequence of tax rates until 𝑡𝑡 − 1: ℎ𝑡𝑡 = (τ0 , τ1 , … , τ𝑡𝑡−1 ) ∈ [0,1]𝑡𝑡 . A time 𝑡𝑡 strategy for a young voter is a mapping 𝑦𝑦 from the history into the action space, σ𝑡𝑡 : ℎ𝑡𝑡 → [0,1]; and analogously for an old voter, σ𝑜𝑜𝑡𝑡 : ℎ𝑡𝑡 → [0,1]. In a majority voting game, the political outcome has to be preferred to any other outcome by a majority of voters. Thus, since the young constitute a majority of voters, the political outcome is determined by 𝑦𝑦 the actions of the young, τ𝑡𝑡 = 𝑎𝑎𝑡𝑡 . For a given sequence of action profiles, 𝑦𝑦 𝑦𝑦 𝑦𝑦 𝑜𝑜 (𝑎𝑎0 , 𝑎𝑎0𝑜𝑜 , … , 𝑎𝑎𝑡𝑡 , 𝑎𝑎𝑡𝑡𝑜𝑜 , 𝑎𝑎𝑡𝑡+1, 𝑎𝑎𝑡𝑡+1 , … ), and corresponding outcomes, (τ0 , … , τ𝑡𝑡 , τ𝑡𝑡+1 , … ), the payoff function of a young player at time 𝑡𝑡 is given by his indirect utility function, 𝑣𝑣𝑡𝑡𝑡𝑡 (τ𝑡𝑡 , τ𝑡𝑡+1 ), and analogously for an old player, 𝑣𝑣𝑡𝑡𝑡𝑡−1 (τ𝑡𝑡−1 , τ𝑡𝑡 ). The equilibrium concept we use is subgame perfection. 4 In this voting game, what kind of strategy profile supports an implicit contract, and thus a positive level of pension as a subgame perfect equilibrium ∞ 𝑦𝑦∗ outcome? Consider any strategy profile �σ𝑠𝑠 , σ𝑜𝑜∗ 𝑠𝑠 �𝑠𝑠=𝑡𝑡 such that:

σ𝑠𝑠𝑦𝑦∗ = �

τ∗𝑠𝑠 , 𝑖𝑖𝑖𝑖 τ𝑠𝑠−𝑖𝑖 = τ∗𝑠𝑠−𝑖𝑖 𝑓𝑓𝑓𝑓𝑓𝑓 𝑖𝑖 = 1, … , 𝑠𝑠 − 𝑡𝑡 0

𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒

(19)

and call (𝑣𝑣𝑠𝑠∗ )∞ 𝑠𝑠=𝑡𝑡 the resulting payoffs for the young. Since old voters are a minority, their actions are irrelevant as they cannot affect the outcome of the game. The young are required to vote for a tax rate τ∗𝑠𝑠 , at time s, if the se𝑠𝑠−1 quence of tax rates (τ𝑖𝑖∗ )𝑖𝑖=1 has been played in the past, and to vote for a zero tax rate otherwise. This strategy profile is an equilibrium if no agent has an incentive to deviate from it. That is, no young person has to be the first to vote for a tax that is different from the optimal policy τ𝑠𝑠 ≠ τ∗𝑠𝑠 , and it has to be incentive compatible to punish defectors. The best deviation for a young individual is to vote for a zero tax rate, τ𝑠𝑠 = 0, in which case he receive no pension in old-age, and the associated payoff is his indirect utility level 𝑣𝑣𝑠𝑠𝑠𝑠 (0,0); whereas the payoff from the strategy

4 A subgame perfect equilibrium is a refinement of a Nash equilibrium, which requires a strategy profile to be a Nash equilibrium of every subgame of the game.

6.

Pensions

89

σ𝑠𝑠𝑦𝑦∗ is 𝑣𝑣𝑠𝑠∗. Thus, if 𝑣𝑣𝑠𝑠∗ ≥ 𝑣𝑣𝑠𝑠𝑠𝑠 (0,0), a young person will not deviate from the op𝑠𝑠−1 timal strategy (τ∗𝑖𝑖 )𝑖𝑖=1 . The utility from punishing a defector is again 𝑣𝑣𝑠𝑠𝑠𝑠 (0,0),

since the punisher effectively dismantles the system, which exceeds the utility from not punishing the defector because 𝑣𝑣𝑠𝑠𝑠𝑠 (0,0) ≥ 𝑣𝑣𝑠𝑠𝑠𝑠 (τ∗ , 0) for τ∗ ≥ 0. Hence, if the existence of a pension system provides at least the same utility to the young as no pension (or more), i.e. ν∗𝑠𝑠 ≥ ν𝑠𝑠𝑠𝑠 (0,0) ∀𝑠𝑠, the above strategy pro∞ 𝑦𝑦∗ file, �σ𝑠𝑠 , σ𝑜𝑜∗ 𝑠𝑠 �𝑠𝑠=𝑡𝑡 , is a subgame perfect strategy profile. The economic factors underlying this result were reviewed in the previous section. Hence, if it is economically convenient to vote for a pension system, due to any of the reasons reviewed in section 6.2, a PAYG pension system will be supported in a repeated voting game. 6.4 Population Aging and Pension Systems In the last few decades, PAYG pension systems have received large attention due to the implications of the aging process on these systems. Aging tends to increase the proportion of retirees, who draw resources from the pension system, and reduces the proportion of workers who contribute to it. Hence, this demographic trend puts pressure on the financial sustainability of the PAYG pension systems, particularly when the current economic growth is not strong enough to compensate for the negative demographics. At same point, revenues from contributions will not be sufficient to cover the pension benefits awarded under the current rules and reforms, such as higher contribution rates or lower pension benefits, will be needed to balance the budget. The aging process is due to a combination of higher life expectancy and lower fertility rates. These two elements have created a substantial increase in the proportion of elderly people in society, and, more relevantly for PAYG pension systems, increased the old age dependency ratio, which is defined as the ratio of people aged 65 or more compared to people aged 18 to 64. Figure 2 shows the values of this important measure in 2015 and the projections for 2050. The average dependence ratio among OECD countries in 2015 is around 25%, which corresponds to four individuals at working age for every elderly (65+) individual. Yet, in 2050 the average ratio among OECD countries will increase to 50%, corresponding to two workers per retiree. These averages mask large differences. As shown at Figure 2, populations in countries such as Japan, Spain and Italy will age much faster than others. Furthermore, the old age dependency ratio may actually underestimate the ratio of retirees to workers; in many countries individuals retire earlier than the official retirement age – and in general earlier than 65. What is the impact of aging on political support for social security? According to the theoretical framework introduced in previous sections, there

90

POLITICAL ECONOMICS

are two major channels through which the aging process can affect pensions: economic and political. The economic channel works through the reduction of the average return from a PAYG system, (1+n)(1+g), induced by the demographic process. In fact, aging increases the number of retired recipients while reducing the proportion of working contributors. The corresponding drop in the average return diminishes individuals’ support for pensions. The political channel, however, allows the aging process to modify the relative importance of the different age groups in the political arena. An older electorate is characterized by more political power for (elderly) individuals who benefit the most from pensions. In other words, due to aging, the pivotal individual on the political process– the median voter – becomes older. Hence, the economic and political effects of aging on PAYG pension systems pull in different directions. Which effect will dominate? To answer this question, the political economy models presented in the previous sections are simulated to provide a quantitative evaluation of the overall effect of aging on social security. The logic of the simulation exercise is first to calibrate the model to an initial economic, demographic and political scenario. It then simulates the political equilibrium which would arise in the future given the predicted demographic, economic and political characteristics (see Galasso and Profeta, 2004, and Galasso, 2006). The results from these simulations suggest that the political aspect – namely, the increased political influence of elderly voters – tends to dominate. In all countries analyzed by Galasso and Profeta (2004), the pension contribution rate – hence the size of the system – increases. Despite this increase, however, the replacement rate, i.e., the ratio between the pension benefit and the (latest) wage income, decreases. This is because aging increases the dependency ratio. So even if pension contributions become higher, as fewer workers contribute and more retirees draw benefits, pension benefits are less generous. The simulations however show that, in all countries, an increase in the retirement age mitigates the rise in contribution rates.

6.

Pensions

Figure 2: Current and projected old-age dependency ratio

Source: OECD Pensions at a Glance (2015)

91

7. Labor Market Policies

In chapter 4, we argued that a government often intervenes in the labor market and that this intervention has important effects on labor market outcomes, such as unemployment and labor force participation rates. Unemployment benefits (UBs) and employment protection legislation (EPL) protect individuals against the risk of losing their job. Yet, while EPL protects the employed individuals and does not levy any tax, UBs transfer income to the unemployed and are funded by taxes on labor income. Across OECD countries there is a large variation in the use of EPL and UBs. Plotted against each other, various measures of the two institutions point to a trade-off between EPL and UBs. While "flexicurity" countries adopt a combination of flexible labor market (low EPL) and large income security for the unemployed (generous UBs), other countries, notably in Southern and Continental Europe, adopt strict EPL and smaller UBs programs, as shown in Figure 1. The location of countries along this trade-off corresponds to stable political-economic equilibria. Reforms of EPL are mostly marginal. As discussed in chapter 4, they are confined to introducing more flexible contractual types "at the margin", rather than modifying rules for workers who already have a permanent contracts. Reducing the fiscal cost of flexicurity is also proving very difficult. UBs systems are adjusted by modifying enforcement rules (e.g. via activation schemes) rather than statutory replacement rates or the maximum duration of benefits. This type of adjustment shows strong political opposition to structural reforms that would move a country along this tradeoff. Why do different countries resort to alternative combinations of employment protection and unemployment insurance to protect individuals against the risk of being unemployed? Why is it so uncommon to reform these institutional configurations?

94

POLITICAL ECONOMICS

Figure 1: The Unemployment benefit –EPL Trade-off

This chapter builds on a growing literature on the political economy of the labor market by providing an explanation for the use of these two labor market policies and discusses the trade-off between EPL and UBs across countries. Current literature deals mainly with the relationship between insiders and outsiders, where the insiders are the workers (with permanent contracts) and the outsiders are the unemployed (or employed with temporary contracts). These two groups have a different employment status, different preferences and would hence benefit from different policies. For instance, regular workers typically prefer a high degree of employment protective legislation, which gives them more job security; the unemployed prefer a low degree of EPL, which increases their probability of finding a job. With regards to unemployment benefits, preferences are reversed: the unemployed – who receive the transfers – prefer higher benefits than the workers, who pay income taxes or contributions to finance the unemployment benefits. 7.1 A Simple Political Economy Model We introduce a simple political-economic model to analyze this issue. The economy is populated by two types of agents: Employed (E) and Unemployed (U). Agents live forever – alternatively, one may think of different dynasties of

7.

Labor Market Policies

95

agents, with current agents caring about the welfare of their entire dynasty (sons, grandsons and so on), which gives them an infinite time horizon. Workers have a fixed labor supply normalized to one unit per period (e.g. 40 hours a week). Agents value their current and future (expected) consumption according to a utility function, 𝑈𝑈(. ), which is increasing and concave in the level of consumption (an example of such a function is given, for instance, by a logarithmic utility function, 𝑈𝑈(. ) = 𝑙𝑙𝑙𝑙(. )). In other words, agents are risk averse, and are thus willing to purchase insurance against the risk that characterizes this economy, that is, the risk of becoming unemployed. Net individual’s income, Yti, is determined by the labor market status as follows, respectively for the employed and unemployed individuals: Employed: Unemployed:

𝑌𝑌𝑡𝑡𝐸𝐸 = 1 − τ𝑡𝑡 𝑌𝑌𝑡𝑡𝑈𝑈 = 𝐼𝐼𝑡𝑡

(1)

where τt is the tax rate levied on labor income (notice that labor income is normalized to one, w=1), and used to finance unemployment benefits; and It is the unemployment benefit paid at time t to an unemployed person. To study the political determination of EPL and UBs, we consider a simple economy in which there are no capital markets – that is, no banks, financial intermediaries or loans – and hence individuals are unable to save. This implies that – for all agents – current consumption equals current income. Another important assumption in this simplified framework is that taxation and unemployment benefits do not create any economic distortions, unlike those discussed in chapter 4. Clearly, this assumption does not real to real world situations. In fact, taxation typically reduces labor demand by firms – thereby decreasing wages and employment (or creating unemployment) – while generous unemployment benefits tend to reduce individuals incentives to search for a job, and therefore the probability of finding a job when unemployed – again decreasing wages and employment (or increasing unemployment). The labor market dynamics in this simple environment are summarized in Figure 2. At every period, a percentage f of employed individuals is fired and hence moves from being employed to being unemployed. Thus, f represents the (firing) rate at which an employed worker loses his job. At the same time, a share h of the unemployed is hired and becomes employed. Thus, h represents the (hiring) rate at which an unemployed individual finds a job. It follows that the dynamics of the equilibrium unemployment rate over time is: 𝑢𝑢𝑡𝑡 = (1 − ℎ)𝑢𝑢𝑡𝑡−1 + 𝑓𝑓(1 − 𝑢𝑢𝑡𝑡−1 )

where ut is the unemployment rate at time t.

(2)

96

POLITICAL ECONOMICS

Figure 2. The relationship between Employment and Unemployment

7.2 Labor Market Institutions To analyze the impact of the EPL on the labor market outcome and how EPL is determined in the political system, we examine a case in which both hiring and firing rates depend on the degree of Employment Protection Legislation, s. In particular, we consider that the firing rate decreases linearly with the EPL, f(s). This relationship captures the restrictions placed on the firms’ capacity to dismiss a worker by the EPL, which thus reduces the firing rate. But the hiring rate is also affected: it decreases non linearly as the degree of EPL increase. This is in accordance with empirical evidence (OECD, 1999) and with economic theory (e.g., Bentolila and Bertola, 1990) predicting that, when labor markets are rigid, employers hire fewer workers in upturns in order to reduce the cost of dismissals during downturns. The relationship between firing rate, hiring rate and the degree of EPL is summarized in Figure 3, where the degree of EPL, s, is assumed to range between 0 (the lowest level of EPL) and 1 (the highest). In the dynamic version of the model, the unemployment rate is determined as follows: 𝑢𝑢𝑡𝑡 = 𝑓𝑓(𝑠𝑠)(1 − 𝑢𝑢𝑡𝑡−1 ) + �1 − ℎ(𝑠𝑠)�𝑢𝑢𝑡𝑡−1

(3)

In order to calculate the steady state equilibrium value, we need to consider the situation in which the unemployment rate is constant over time, that is, when 𝑢𝑢𝑡𝑡 = 𝑢𝑢𝑡𝑡−1 . Using these two equations, it is possible to turn the dynamic version of the unemployment equation into a static equation:

7.

Labor Market Policies

97

𝑢𝑢 = 𝑓𝑓(1 − 𝑢𝑢) + (1 − ℎ)𝑢𝑢

(4)

𝑢𝑢(1 + 𝑓𝑓 − 1 + ℎ) = 𝑓𝑓

It follows that the equilibrium unemployment rate at the steady state is equal to: 𝑓𝑓(𝑠𝑠)

𝑢𝑢(𝑠𝑠) = 𝑓𝑓(𝑠𝑠)+ℎ(𝑠𝑠)

(5)

Figure 3. The relationship between f(s), h(s) and s

In Figure 4, 𝑠𝑠̂ is defined as the rate of EPL that minimizes unemployment. Since the effect of EPL on the firing rate (more EPL reduces the firing rate) is supposed to start immediately, while the effect on the hiring rate (more EPL reduces the hiring rate) takes place for higher levels of EPL, the relationship between unemployment rate and EPL is U-shaped, as shown at Figure 4. Besides the EPL, the second policy instrument in the labor market is given by the UBs .This welfare scheme provides a transfer to an unemployed individual, financed by labor income. More specifically, the revenues consist of the labor tax imposed on all workers, τ (1 − 𝑢𝑢), where 1 − 𝑢𝑢 represents the total share of workers in the economy. The insurance system then pays a benefit I to all unemployed individuals (their share is equal to u). Thus, the budget constraint of the unemployment insurance program is: 𝐼𝐼 𝑢𝑢 = τ (1 − 𝑢𝑢)

(6)

98

POLITICAL ECONOMICS

Figure 4. The determination of the optimal rate of s

Once the equilibrium unemployment rate has been determined (recall that it may depend on the degree of EPL), it can be substituted into this equation to find the following expression: 𝐼𝐼 = τ

1−𝑢𝑢 𝑢𝑢

= τ

𝑓𝑓 𝑓𝑓+ℎ 𝑓𝑓 𝑓𝑓+ℎ

1−

= τ

𝑓𝑓+ℎ−𝑓𝑓 𝑓𝑓+ℎ 𝑓𝑓 𝑓𝑓+ℎ

⇒ 𝐼𝐼 = τ

ℎ 𝑓𝑓

(7)

It is now convenient to analyze the (indirect) utility function of the different types of individuals. Since all individuals are equal, with the notable exception of their current state of employment, it is easy to obtain the utility of an employed and an unemployed worker. The workers who are currently employed have the following utility: 𝑉𝑉 𝐸𝐸 = 𝑈𝑈(1 − τ) + β [(1 − 𝑓𝑓)𝑉𝑉 𝐸𝐸 + 𝑓𝑓𝑉𝑉 𝑈𝑈 ]

(8)

Whereas individuals who are currently unemployed have the following value function: 𝑉𝑉 𝑈𝑈 = 𝑈𝑈(𝐼𝐼) + β [ℎ𝑉𝑉 𝐸𝐸 + (1 − ℎ)𝑉𝑉 𝑈𝑈 ]

(9)

In both these value functions, β represents the discount rate. These expressions combine the utility during the current period, with the expected discounted utility from consumption in the following periods. Clearly, the value functions only differ in the state of the individual, and there is no difference in calendar time. When combining these two expressions, one can get the following functions.

7.

Labor Market Policies

99

Employed: (1 − β)𝑉𝑉 𝐸𝐸 = 𝑈𝑈(1 − τ) − [1−

Unemployed:

𝑓𝑓β [𝑈𝑈(1 − τ) − 𝑈𝑈(𝐼𝐼)] β+β(𝑓𝑓+ℎ)]

ℎβ

[𝑈𝑈(1 − τ) − 𝑈𝑈(𝐼𝐼)] (1 − β)𝑉𝑉 𝑈𝑈 = 𝑈𝑈(𝐼𝐼) + [1−β+β(𝑓𝑓+ℎ)]

(10)

(11)

It is important to notice for the unlikely case, in which β → 1, these two expressions would be identical and equal to: (1 − β)𝑉𝑉 𝐸𝐸 = (1 − β)𝑉𝑉 𝑈𝑈 = ℎ𝑈𝑈(1 − 𝑡𝑡) + 𝑓𝑓𝑓𝑓(𝐼𝐼)

This is because for β→1 the current labor market condition (employed or unemployed) has the same value as the infinite sequence of future conditions. When β < 1, instead, future conditions are discounted more heavily, and the current status of the individual influences their preferences. 7.3 The Political Game These two institutions of the labor market are decided through a political economic voting process. We analyse the case in which these two instruments are determined separately as two different policy decisions. First, we analyze the political decision over the amount of EPL by voting on s. Second, we consider how the level of UBs (𝐼𝐼, τ) is determined, by voting on τ. In both voting procedures, individual preferences are assumed to be single-peaked and the election is assumed to be once-and-for-all voting, that is, once in place, the policy cannot be modified in the future. An increase in the amount of EPL reduces the hiring and firing rates, so that labor market dynamics will slow down, the market will become more rigid, and both hiring and firing will become less frequent. For the employed workers, this implies that the probability of being fired decreases – hence they will prefer a high level of EPL. For unemployed individuals, instead, a high level of EPL is unattractive, because hiring is reduced and thus the probability of an unemployed person finding a job decreases. Hence, the employed can be expected to prefer higher EPL than the unemployed. If the unemployment rate is below 50 %, the median voter is an employed worker and thus the level of EPL will be high as it is chosen by the employed. Regarding the UBs, employed people trade-off the current cost of the program, due to the current contributions they have to pay, with future bene-

100

POLITICAL ECONOMICS

fits, since they receive a transfer if they become unemployed. Currently unemployed individuals instead trade-off a current gain from the insurance, consisting of the current transfer, with a future cost, occurring when they find a job and pay UBs contributions. Since the current status matters, β < 1, the employment insurance reduces the difference in utility between being employed and being unemployed. Hence, the preferred level of UBs will be higher for the unemployed than for the employed. Yet, since the median voter is a worker, the actual level of UBs will be the one chosen by a worker. To summarize, when voting over the level of EPL, we have 𝑠𝑠 𝐸𝐸 > 𝑠𝑠̂ > 𝑈𝑈 𝑠𝑠 , where 𝑠𝑠̂ is the level of EPL that minimizes the unemployment rate. Hence, the political system delivers a level of EPL that is larger that the level that minimizes unemployment. In equilibrium, there will be more unemployment than required. For the UBs, we have τ𝐸𝐸 < τ𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 < τ𝑈𝑈 , where τ𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 is the case of full insurance, with β = 1 - that is, it is the level of UBs that provides individuals will a complete level of insurance against the income risk related to becoming unemployed. Hence, in this case the political system delivers less than full insurance. The political equilibrium of this separate voting system delivers a high degree of EPL but a low level of UBs. 7.4 Discussion The political game described so far with a separate voting system delivers political equilibria with high degrees of EPL and a low level of UBs. However, this is not what we observed in Figure 1, where we see European countries with higher expenditure on UBs (and Active Labor Market Policy) per unemployed person having less EPL, and vice versa. Only Anglo-Saxon countries seem to have little of both. Moreover, moving along this trade-off has proved to be very difficult, as shown empirically in Chapter 4 for the EPL, as changes have mostly been limited to more flexible contract types for new employees, with no changes for workers who already have a permanent contract. Analogously, UBs systems have experienced only modest changes in replacement rates and duration of benefits. Hence, some countries choose to protect the jobs more than unemployed individuals and continue to do so; while other countries do the opposite, they support the unemployed rather than protecting their (former) jobs. What, then, explains the location of different countries along this trade-off, that corresponds to stable political-economic equilibria? The economic literature acknowledges that both EPL and UBs provide insurance against labor market risks. However, EPL does not impose a tax burden on the worker, whereas UBs is financed with a tax on labor income and transfers income to the unemployed. Thus, the latter policy is potentially redistributive. Political economic models, which study the joint determination

7.

Labor Market Policies

101

of EPL and UBs in the political arena, introduce two conflicts of interest. One conflict is between insiders and outsiders and arises in the transitions between employment and unemployment, which are affected by the EPL. The other conflict is the traditional redistribution between rich and poor (or high and low ability types), which occurs because both EPL and UBs involve a degree of redistribution across ability types. Boeri, Conde-Ruiz and Galasso (2012) consider a political system in which the entire electorate votes simultaneously on the size of the UBs and on the strictness of EPL, using an issue-by-issue voting framework (see Shepsle, 1979). In this case, the median voter on the UBs is a low or high ability insider (i.e., employed individual). However, the median voter on the degree of EPL may be a low ability insider or outsider or a high ability individual, depending on how the population is distributed by ability level. Boeri, Conde-Ruiz and Galasso (2012) suggest that the flexicurity configurations, with low EPL and high UBs, emerge in societies with a large (but not necessarily majoritarian) share of educated individuals. The reason is that, in these countries, the median voter over the EPL decision (and perhaps also on UBs) will not be a low ability insider. Thus, the political equilibrium will feature more a flexible labor market and more UBs.

8. Structural Reforms

In the last few decades, structural reforms have often been advocated as a crucial instrument to increase productivity growth and to reduce the increasing divergence in economic performances among, for instance, EU Member countries. In fact, structural reforms modify the rules of the game in markets or welfare state programs in a fundamental way, and typically have important consequences. Indeed, many OECD countries have reformed their labor and product markets. Yet, significant cross-country differences in the speed and scope of these reforms have emerged. Economic advisors and international organizations are almost unanimous in recommending reforms that liberalize markets in order to reduce existing rents and enhance economic growth. Why have structural reforms been so difficult to implement? And what explain such large differences in the reform process across countries? As many other policies, reforms are undertaken by policy-makers, who respond to political incentives. Hence, a positive analysis of the reform process has to consider the political costs and benefits attached to the reforms In fact, structural reforms in labor and product markets have political costs, since they tend to have redistributive effects and thus produce winners and losers. Why do some politicians risk political capital to undergo reforms? While ideological motivations may play a role, more often politicians are influenced by economic reasons. For instance, some features of the labor and product market may no longer be economically sustainable in a changing environment. Often this occurs during a sudden economic crisis, but it may also happen before to prevent such a crisis from occurring. The economic drivers of reforms include population aging and its negative impact on the financial sustainability of PAYG pension systems, as analyzed in chapter 6. Efficiency considerations aimed at improving the market allocation of labor often drive labor-market reforms, as discussed in chapter 7. Equity reasons have also been put forward as an argument against strong employment protection legislation (EPL), which may give rise to labor-market dualism, characterized by a strong separation

104

POLITICAL ECONOMICS

between (senior) workers with permanent contracts and (younger) workers with temporary or no contracts. Structural reforms of product markets, such as liberalizations and privatizations, are also supported on efficiency grounds (Bassanini and Ernst, 2002, Scarpetta et al., 2002, Schiantarelli, 2005), in an attempt to increase economic productivity. 8.1 Describing Structural Reforms Structural reforms largely modify the rules of the game in markets or welfare state programs. Typically, after such a reform there are changes in the balance of power in a market, in the identity of the beneficiaries, or in the amount of their benefit. This chapter focuses on structural reforms in a pension system, and in product and labor markets. Figure 1: Changes in the implicit tax on continuing to work

Source: OECD Database Note: Implicit tax rates on continued work for workers in the age cohort 60-64 years in early retirement pathways and in old-age pension schemes. Latest available data for 2009. For more details, see Duval (2004)

To respond to population aging, several pension reforms have recently been implemented in order to limit the increase in pension spending and to maintain the systems financially sustainable. The typical reform measures have includ-

8.

Structural Reforms

105

ed: an increase in the retirement age, as in Italy in 1992, 1995, 2004 and 2011, in France in 2003, and in Germany in 1992, 1997 and 2003; modifications of benefit formulas to reduce pension generosity, as in Italy in 1995; and fiscal incentives to increase reliance on private pensions, as in Italy in 1995 and 2004, in France in 2003 and in Germany in 2001. Of particular relevance are reforms designed at postponing the age of retirement. This goal has been achieved through an outright increase in the legal retirement age, and through a reduction in incentives to retire early. The OECD measures these incentives for individuals over 60 years old, as an implicit tax on continuing to work one more year rather than retiring. Figure 1 reports fiscal incentives to retire early for several countries in 1985 and 2009. The numerous countries that are below the 45° line modified their pension systems in order to reduce these incentive and thus to increase the effective retirement age. Yet, not all countries took this action. Truly structural pension reforms that that fundamentally changed the system were adopted in Italy and Sweden in the mid-90s, and in Poland a few years later. Figure 2: Average Trend in Product Market Liberalization

Source: OECD Database. Note: OECD summary indicator of regulatory impediments to product market competition in seven non-manufacturing industries: gas, electricity, post, telecommunications, passenger air transport, railways (passenger and freight services) and road freight. For more details, see Conway and Nicoletti (2006).

106

POLITICAL ECONOMICS

The existing defined-benefit (PAYG) pension schemes were substituted, in some cases over a long transition period, with a (still PAYG) notional defined contribution scheme. 1 The most common structural reforms in OECD countries have occurred in the product market. Yet, even in this case, the magnitude varies across countries and over time. The intensity of these reforms is measured by the OECD indicators of anti-competitive regulation, which considers seven non-manufacturing sectors (electricity, telecommunications, natural gas, postal and railway services, road and air transport). As shown in Figure 2, which uses the average indicator for OECD countries, product market liberalization (displayed as a decrease in the anti-competitive regulation index) has occurred since the late 1980s. Figure 3: Catching-up in Product Market Liberalization

Source: OECD Database. Note: The OECD summary indicator of regulatory impediments to product market competi-tion across seven non-manufacturing industries: gas, electricity, post, telecommunications, passenger air transport, railways (passenger and freight services) and road freight. For more details, see Conway and Nicoletti (2006). 1 In NDC schemes, the calculation of an individual’s pension is directly linked to his lifetime contributions, although these contributions are not accrued in a funded system, but are used to pay current retirees’ pensions, as in all PAYG systems. This is why the system is only a “notional” defined contribution.

8.

Structural Reforms

107

From Figure 3, which shows the change in this measure between 1975 and 2013 for each OECD country, it clearly emerges that the deregulation process was stronger in countries that were more regulated in 1975. The negative relationship between the initial degree of regulation and the subsequent liberalization suggests a process of catching-up by the most regulated countries. Structural reforms in the labor market have perhaps proved the most difficult to implement. As shown at chapter 4, changes in the regulation of permanent labor contracts did not take place in OECD countries, with the ex-ception of Portugal, Spain, to a lesser extent Finland, and in Italy during 2016. On the contrary, measures that deregulated temporary contracts and affected the unemployed with low employment probability, (i.e., the modern labour-market “outsiders“) have been adopted in several countries since the 1980s. As a result, the combination of liberalized temporary contracts and rigid perma-nent labor contracts has created a large duality in the labor market between insiders and outsiders. This duality can be seen in Figure 4, which displays the share of temporary contracts by age groups in several European countries. While more senior workers (the “insiders”) rely on regular contracts, the share of temporary contracts is much higher among young individuals. Figure 4: Proportion of workers in temporary employment by age

Source: OECD, Employment Outlook, 2015.

108

POLITICAL ECONOMICS

8.2 Determinants of Structural Reforms Structural reforms can be unpopular. Existing economic rents in labor and product markets and a generous welfare state create a large political constituency. Welfare program beneficiaries and bureaucrats support the status quo and oppose reforming measures. So do labor market insiders, as well as workers and firms who benefits from the economic rents stemming from highly regulated markets. The potential losers from structural reforms are easily identified, and vocal in opposing any reforming efforts. Furthermore, despite being a minority, these potential “losers” stand to lose large stakes from reforming measures and are therefore willing to lobby against them. On the other hand, potential winners from structural reforms are typically outsiders (in the labor market) and consumers (in product market), who may have less to gain. In this scenario of concentrated (large) costs for potential losers, and dispersed (small) benefits for the potential winners, policy-makers have to decide whether or not to make reforms. As discussed in previous chapters, incumbent politicians, who act as policy-makers, adopt public policies in pursuit of their political career. In pension reforms, as well as in labour market policies, broad electoral interests are important, and politicians may use the reforming measures to target their core constituency of voters or undecided (swing) voters (see Galasso, 2006). However, in product markets reforms, liberalizations have a strong (negative) impact on a small number of losers, and a large number of marginal winners. Here, the collective actions problem may emerge, as winners are too diffuse to become an organised electorate (Olson, 1965). On the other hand, losers have strong incentives to lobby in order to block potential reforms by offering political support and financial contributions (Grossman and Helpman, 2001). The lobbying models were discussed in chapter 3. Two additional factors may hinder structural reforms: information and timing. First, individuals may be uncertain about their status at the time the reform takes place. In order words, they may not know whether they will win or lose. In risk averse individuals, this uncertainty reduces the aggregate support for reforms. In fact, even potential winners may prefer to oppose reforms from which they are not certain to benefit. In this scenario, even if a majority of the population gains ex-post from the reforms, it is difficult to find an exante majority that supports the reforms. Hence, in absence of perfect information, a status quo bias may emerge (Fernández and Rodrik, 1991). Second, the timing of the cost and benefits in the process of structural reforms may be politically unfortunate. In fact, reforms measures typically generate current costs and future (expected) benefits. This is the case with labor market liberalizations, which immediately increase the unemployment risk, as firms find it easier to dismiss workers; whereas the new employment opportunities materialize later (Coe and Snower, 1997). Given the short time horizon of politicians

8.

Structural Reforms

109

due to the electoral cycle, this cost-benefit time distribution further decreases the political appeal to make reforms. 8.2.1 Economic Crisis Do economic crises facilitate or hinder structural reforms (see Haggard and Kaufman, 1992, and Drazen, 2000)? For political economic theorists, the answer is uncertain since two different channels can play a role. On one hand, deep crises require quick reactions, particularly if the existing programs or regulations are, at least partially, accountable for the deteriorating of the economic conditions. Thus, deep recessions call for measures designed to enhance economic efficiency in the product market; high and persistent unemployment puts pressure on labour market reforms, and financial crises lead to pension reforms that drastically reduce public spending. On the other hand, deep crises create difficult economic conditions for introducing structural reforms. In fact, fewer resources are available to compensate the individuals who lose from these reforms, and these agents become even less keen on giving up their economic rents or benefits. For instance, in periods of high unemployment, greater labor market flexibility can prove unbearable for workers already dealing with adverse economic conditions (Bean, 1998) – and all the more so if no compensation package, such as more generous unemployment benefits, can be offered. Empirical analysis (see, among the others, IMF, 2004; Duval and Elmeskov, 2005, Høj et al., 2006; Galasso, 2014) supports the view that some reforms measures are introduced after economic crises. In the labor market, when the long-term unemployment has increased, unemployment benefits havebecome more generous, and more flexibility has been introduced in temporary contracts. Yet, the status of permanent workers in the labor market (the “insiders”) has barely changed. These reforms have been particularly strong in countries characterized by more rigid labour markets, such as Spain, which in 1984 was also experiencing a period of high and persistent (youth) unemployment. In this case, even the permanent contracts were made more flexible after a deep recession in 1994. Pension reforms have often taken place during financial crises. This pattern is particularly evident in Italy, where large pension reforms were introduced in 1992 and 2011 in order to limit public spending, and thereby stabilize financial markets in turmoil. Large economic crises have been shown to lead to deregulation in the product markets (Hoj et al., 2006), particularly in air transport and in postal services. In fact, crises create a demand for more economic efficiency, thereby leading to liberalization. Furthermore, if the response to a crisis includes an increase in public revenues and a reduction in public debt, privatization plans may also be pursued, as in Italy and Korea in

110

POLITICAL ECONOMICS

mid-90s when state-owned enterprises in the energy and telecommunication sectors were sold. 8.2.2 External Constraints A push for structural reforms may also come from external constraints, such as international agreements that bind countries to certain rules. Often, these rules need to be approved by national parliaments that effectively introduce structural reforms in particular markets. One of the main examples of an external constraint is the European Single Market Programme, which requires EU member countries to adopt share regulations in the product market, while leaving the labour-market largely unaffected. Empirical evidence (see Hoj et al., 2006) shows that the implementation of the Single Market Programme was conducive to product market liberalization, particularly in the air transport, road freight, telecommunications and, to some extent, energy sectors. On the other hand, the Single Market Programme did not induce any labour-market reforms (Alesina et al., 2010). Albeit less directly, the globalization can be considered as an external constraint on countries and their politicians. Countries open to international trade have experienced an increase in competitiveness, which increased pressure to reduce labour-market rigidities. Additionally, greater international competition in tradable goods and services compelled domestic producers to demand structural reforms in order to reduce the price of non-manufacturing intermediate inputs. As for the external constraints, empirical evidence suggests that globalization has been more effective at inducing reforms in the product rather than in the labor market. Another type of external constraint has been the introduction of the Euro. The potential for the Euro to act an external constraint was a heavily debate issue before its introduction. By entering the Euro zone, a country effectively relinquished control over monetary policy, which was instead delegated to an external authority – the European Central Bank. Hence, countries in the Euro zone could no longer use their monetary policy to accommodate negative shocks, and had to rely on structural reforms and market-based adjustments (Bean, 1998; Duval and Elmeskov, 2005; and Obstfeld, 1997). 2 Empirical evidence (Alesina et al. 2010) supports the view that adopting the euro had a positive effect on introducing structural reforms. This effect is particularly evident in the product market of Euro countries. Sectors, which were initially more regulated, such as energy and communication, experienced stronger reforms, during a period in which competitive devaluations became unavail2 Other researchers (Saint-Paul and Bentolila, 2000) have more pessimistic views about the effect of joining the euro on structural reforms. According to their argument, the typical up-front cost of reforming can no longer be absorbed by expansionary aggregate demand policies. Lacking this monetary policy instrument, structural reforms in the Euro zone were expected to be politically more difficult to adopt.

8.

Structural Reforms

111

able, and countries were thus forced to liberalize. As for the other external constraints, the adoption of the Euro had no effect on labor market reforms. 3 8.2.3 Political Factors Of course politics matters for reforms. Since politicians have to use their political capital to push forward structural reforms, electoral constraints and political institutions constitute the relevant framework. Strong governments with an electoral mandate to reform or governments with a strong (pro-market) political ideology are, at least in theory, more likely to implement reforms. Analogously, reforms should be easier to implement earlier in the political cycle, as the political cost of reforming immediately materializes, while the benefits occur later on. Empirical evidence mostly supports these views. When governments in power win elections on a pro-reform campaign, successful reforms follow, such as the pension reform in France in 2003, and the labour-market reform in Spain in 1997. Strong governments, with a large (relative) majority in the parliament (see IMF, 2004), are associated with more product market reforms. As majoritarian electoral systems are more likely to induce strong governments, countries with these systems, such as the UK, are found to reform more. Government ideology also matters. As expected, right-wing governments tend to liberalize the product market more, to reduce the generosity of the unemployment benefits for the long-term employed and to decrease implicit taxes on continuing to work (Høj et al., 2006). For instance, the significant reforms to the Spanish labour market in 1997, which introduced a “permanent employment promotion contract” to balance the use of temporary and permanent contracts, was promoted by the conservative Partido Popular (PP). Instead, opposition to (mostly labour-market) reforms is typically channelled through left-wing parties and unions, which are often able to mobilize the public in national strikes and social protests. 8.3 Designing Structural Reforms The picture portrayed so far is discouraging. How to promote structural reforms in absence of major crises or external constraints, and retain political power? Although some empirical evidence (see Buti et al., 2008) suggests that reforming governments do not pay an electoral cost, policy-makers are always cautious about risking their political capital. Although the theoretical literaInterestingly, wage moderation took place from 1993 to 1998 in the countries preparing to enter the Euro, with a substantial reduction in the adjustment of nominal wages to past inflation. After the adoption of the euro, this process continued in Germany, but much less so in Italy and France.

3

112

POLITICAL ECONOMICS

ture on the political economics of structural reforms has not provided a clear recipe for reforms, some best practices have emerged. Three main features can be used to ease the reforms: sequencing and bundling the reform measures, and providing information. 8.3.1 Sequencing Reforms Sequencing matters for reforming, since it may create positive spill-over across markets or sectors. For instance, liberalizing product markets reduces the economic rents available to workers and firms, and may reduce opposition to labour-market reforms (Blanchard and Giavazzi, 2002). Analogously, empirical evidence shows that, in the product market, the optimal sequence may be to liberalize final consumption goods first and then intermediate goods. In fact, liberalizing final goods markets – driven for instance by external constraints, such as the European Single Market, or globalization (see section 2.2 – will increase the demand for liberalization of intermediate goods by firms competing on the final goods market, in fact, these firms use intermediate goods as input, and seek to reduce their production cost. In the transportation sector, this sequence has proved successful. As trade liberalizations increased, competition in the final goods market, road freight - which represents a typical intermediate good - was also liberalized. Railroad and air transport also followed suit. Sequencing reforms may also obey a different logic: divide et impera. In this case, the timing of the reforms is used to disentangle the vested interests of different actors in an attempt to reduce the opposition to reforms. Empirical evidence (Hoj et al. 2006) suggests that this strategy has been successful in the product market (air transport and telecommunication). In these instances, the corporatization of public enterprises was pivotal in creating a constituency of workers with private contracts who were then less oppose to privatization. Sequencing may also work to reduce opposition in the labour market. In Spain, labor market reforms started in 1984 with the liberalization of the temporary contracts, which affected less powerful and less unionized workers. It was then extended in 1994 to include permanent contracts (Dolado and Jimeno, 2004). Yet, this type of sequencing has generally not worked outside of Spain. This logic of sequencing was used also in pension reforms in an attempt to concentrate costs on the (less politically powerful) younger generations, and to save the elderly from any loss from reforming. This has been achieved through long transition periods, which allowed senior workers and retirees to escape the cuts (Boeri et al., 2006). 8.3.2 Bundling Reform Measures Another instrument in creating politically sustainable structural reforms is to bundle different policy measures into a single reform programme; this allows

8.

Structural Reforms

113

to compensate potential losers from the main reform with benefits from other policies. This strategy is feasible when policy measures, which are close substitute for one another, are available. It is typically the case in the labor market, where UBs and employment protection legislation (EPL) have common features (Pissarides, 2001; Blanchard and Tirole, 2003), as discussed in chapter 7. This substitutability may, for instance, be exploited to move along the unemployment insurance – EPL trade-off. Similarly, the impact of less generous unemployment benefits may be compensated by more active labor market policies (ALMP). The are numerous examples of bundling reforms in the Danish labor market during 1994-97, and also during the 1993 French reform process, when unpopular retrenchment measures were limited to private workers only, while leaving highly unionised, state employees unaffected. 8.3.3 Providing Information Poor public information is responsible for a large degree of the uncertainty surrounding the cost and benefits of keeping or reforming the status quo. Reliable information is largely unavailable to individuals about these (complicated) public policies. Nevertheless, providing credible information is problematic, since individuals may be sceptical of information provided by the government due to possible ideological biases, rather than to actual economic factors. This may create a paradoxical effect, according to which parties typically opposed to structural reforms for ideological reasons are more credible in advocating them (Cukierman and Tommasi, 1998). International organizations and think-tank, such the World Bank, the IMF and the OECD, have played an important role in providing information on pension, product and labor market reforms to policymakers and the general public. The provision of information from these institutions has been crucial in the success of many pension reforms (Italy, 1995, France 2003). Yet, recent scepticism about the credibility of these international organizations, especially in Europe, may undermine their role as providers of reliable information. To some degree, information about the true cost of the status quo may be conveyed more effectively by sudden economic crises rather than government press releases.

9. The Politics of Anger

The last decade has been characterized by a spread sentiment of anger among voters in many countries. Several political analysts have attributed unexpected electoral defeats of incumbent or of (ex-ante) favorite candidates to “protest vote”. According to this view, those who voted “leave” in the Brexit referendum in June 2016, “no” in the referendum on the Italian constitution in December 2016, and for Donald Trump in the 2016 US Presidential election, were effectively expressing their discontent with the status quo and the incumbent political class. In this chapter, we examine factors that may have generated this discontent. In particular, we concentrate on the economic and cultural effects of migration on native populations. Is immigration the main reason for anger among voters in many OECD countries? Other recent phenomena may have produced similar economic effects as immigration. This is the case for skilled biased technological change and globalization, which induced similar changes in labor market for the native population. Yet, immigration is more visible and can easily be identified by the domestic population. To appreciate the relevance of this phenomenon, it is important to start with an assessment of the size of immigration flows, and its effects on the local population. Figure 1 shows the proportion of non-EU born individuals in the population of several EU countries. This measure provides a snapshot of the population’s composition, as immigrants ebb and flow through EU and non-EU across countries. The share of non-EU natives is above 5% in only six EU countries: Austria, Luxemburg, Greece, Spain, Italy and the Czech Republic. Several Eastern European countries, which have recently been strongly opposed to receiving refugees, are below 1%. Indeed, if we concentrate on the flow of refugees only (excluding so called economic immigrants), the annual influx of refugees to EU countries has always been below 0.5% of the EU population, even in the late 90s during the Balkan war.

116

POLITICAL ECONOMICS

Figure 1: Percentage of non-EU nationals in the population

Figure 2 displays the inflow of refugees into the EU member countries from World War II to the present day. To summarize the overall size of the migration phenomenon, EU countries have admitted a total of 3.5 million permanent migrants per year between 2010 and 2014. In comparison, the US has admitted about 1.5 million migrants per year for the twenty years between 1990 and 2010. Are 1.5 or 3.5 million large numbers? As a reference for these magnitudes, it is worth noticing that the EU has a total population of about 500 million people, 350 million of whom are part of the labor force. Yet, due to population aging, more than one million individuals leave the labor market every year. Indeed, the net decline of the European labor force, which takes into account the retirement of elderly people and the employment of new workers, has been one million per year. Recent immigration may thus compensate for this declining labor force. Labor market opportunities are an important element in migration decisions. In fact, both push and pull factors interact when making the difficult decision to leave one’s home country. Push factors, which induce individuals to flee their home countries, include poor social and political conditions. Many refugees escape from civil wars or ethnic persecution, which threaten their lives. Poor economic conditions and extreme poverty are powerful push factors that induce individuals and families to seek better economic prospects abroad. These push factors create economic migrants, who, in the public debate, are

9.

The Politics of Anger

117

often considered separately from refugees escaping conflicts. The pull factors determine where migrants try to relocate themselves. Clearly, social and political freedoms as well as good economic conditions, particularly in the labor market, are important determinants in these relocation decisions. Figure 2: Percentage of refugees per capita in the EU (UN data)

9.1 The Effects of Migration on the Natives: Theoretical Aspects Regardless of their motivations, the presence of immigrants has an impact on the lives of the native population. Interaction between migrants and natives are common, and can have both positive and negative elements. The economic effects are mostly related to the impact of immigration on labor market outcomes – such as the employment or unemployment and wages – of the natives. Also the migrants may affect the natives’ access to welfare state programs and local public goods, such as unemployment benefits, health care, or kinder gardens. Their presence may overwhelm these services. Yet their contributions help finance these programs. Economic literature (Card, 2001, Borjas, 2003, Ottaviano and Peri, 2012, Dustmann et al. 2016) on the impact of new workers – the migrants – emphasizes two possible effects. The immigrants can be either substitute for or complement to the native workers. In the former case, which occurs when migrants and natives have similar labor market skills and abilities, the arrival of new immigrants will increase the labor supply of workers with these skills and abilities. Depending on the absence or existence of labor market rigidities,

118

POLITICAL ECONOMICS

economic theory predicts that respectively prices (i.e., wages) or quantity (namely, employment or unemployment rates) will adjust. Thus, if this substitutability among workers emerges, there will be lower wages or less employment (and/or more unemployment) for native workers. If the migrants complement native workers in the labor market due to differing skills or abilities, the inflow of immigrants favors the natives. In fact, by increasing the supply of workers with different skills, the inflow of immigrants effectively makes the native worker a less abundant factor (that is, type of worker) in the production function. As such, the typical jobs occupied by natives are not threatened by immigrants, who find employment in different occupations. Depending again on labor market rigidities, economic theory predicts that wages or employment rates for the natives will increase. Since immigrants are typically characterized by low skills (some immigrants possess high skills that are not recognized abroad), economic theory suggests that both patterns emerge yet for different types of native workers. Hence, low skilled natives will feel threatened by the arrival of low skilled immigrants, whereas high skilled natives will welcome them, as their labor market outcomes (wages and/or employment) improve. These effects on the labour market, however, are not only caused by immigration. Skill-biased technological change and globalization create similar changes. To appreciate this similarity, consider a production function with three factors of production: low-skilled labor (L), high-skilled labor (H), capital (K) and an exogenous productivity parameter (A). Low-skilled labor is a substitute factor with respect to capital, and a complementary factor with respect to high-skilled labor. Capital and high skilled-labor are complements. In other words, in this production function, physical capital – such as machines or computers – can replace low-skilled workers, but requires more high-skilled workers to be operated. In this production function, an inflow of (low ability) immigrants corresponds to an increase in the supply of low-skilled labor. In the absence of labor market frictions (i.e. if wages are fully flexible), the wages of the lowskilled workers will thus drop, but the wages of high-skilled workers and the return to capital will increase. Skill-biased technological change has a similar impact on the labor market. Due to technological progress, capital becomes more productive, and thus the accumulation of physical capital increases. More capital (more equipment and computers) increases the productivity – and thus the demand of high-skilled workers who earn higher wages. But it also reduces the need for (and thus the demand of) low-skilled workers, who are replaced by machines and thus earn lower wages. Finally, globalization may have a similar effect on the labor market. The impact of the globalization on the domestic production of industrialized countries is mostly due to the increasing competition from developing countries. In these emerging countries,

9.

The Politics of Anger

119

unskilled labor is cheap. Thus, low-value-added goods requiring cheap labor and little capital can be produced at low cost and exported abroad. Globalization thus increases competition in domestic markets for low-value-added goods. In industrialized countries, which also produce low-value-added goods, these sectors of production are characterized by a high number of low-skilled workers, a low number of high-skilled workers and low capital. Hence, in these countries, globalization lowered demand for low-value-added goods produced by domestic firms. However, these industrialized countries tend to also produce high-value-added goods. Globalization allows these goods to be exported and sold in larger foreign markets. Hence, globalization also stimulates demand of high-value-added goods, which are produced by domestic firms with capital-intensive technology and high-skilled workers. As a result of the globalization phenomenon, the demand of low-skilled workers will decrease as fewer domestic firms produce low-value-added goods. On the other hand, the demand of high-skilled workers increases, as more domestic firms produce high-value-added goods and export them. The wages of low-skilled workers thus decrease, while the wages of high-skilled workers increase. Therefore, migration, globalization and skill-biased technological change induce similar labor market effects on the native workers. However, the migration process carries a distinctive feature for the native population. Natives have to deal with having foreigners at home with whom share public resources, such as welfare state programs and local public goods. For both cultural and economic reasons, more homogenous societies may be less well equipped to deal with large-scale immigration. Let us begin by analysing the economic aspects related to the sharing of public resources. As discussed in the previous chapters, there are several explanations for the origin and the development of different welfare state programs. Whatever the explanation of their origin, if welfare state programs exist, the native population benefit and will oppose extending their coverage to others. Particularly when the others have a higher probability of using welfare programs, such as unemployment benefits, health care or kinder gardens. In fact, welfare programs provide insurance against certain risks: public health systems cover health risks, pension systems deal with the longevity risks, unemployment benefits ensure against job loss, and so on. Since all individuals finance welfare programs through taxes and contributions, if the probability of using these programs is similar across the population, each individual tends to pay a fair price (through their tax bill) for using them. This is typically the case when the population is relative homogenous in terms of health, life expectancy, unemployment risk, etc, as, for instance, in Scandinavian countries. If individuals in a population are instead heterogeneous in these risks, welfare states tend to induce redistribution, as individual with high risks (for instance of remaining unemployed) use the programs (the unemployment benefits) more often than

120

POLITICAL ECONOMICS

others. Hence, redistribution takes place from the non-users, who still pay welfare contributions, to regular users. Immigrants typically have different characteristics from the native population. They tend to be younger, less educated, and predominantly male. They also differ in their behavior. For instance, they tend to have higher fertility rates than natives. As such, they constitute a different risk type from natives. Being less educated, they have higher unemployment rates – although migrants also have higher geographic mobility and move further than natives to seek employment. Depending on their age, they may have higher health risks. They may have a strong demand for childcare facilities, particularly since low wages induce both parents to work. All these aspects can encourage the natives to request preferential treatment in welfare programs. This aspect may be particularly strong in countries that have a very homogenous population and thereby perceive the effects of immigration as very disruptive to the composition of the pool of individuals insured by the welfare programs. However, being younger than the native population and lacking eligibility for pension benefits, legal immigrants also massively contribute to pension systems, while drawing little or no benefits. Finally, native populations have non-economic concerns regarding immigration, which relate to cultural assimilation, crime and terrorism. Clearly, the demographics of migrants make them a group that is more at risk of committing crimes, as immigrants are mostly young, male individuals. Thus, as migrants represent a selected group of individuals, migration may mechanically be associated with more crime. In other words, even if we isolated the crime rate of young, male individuals among the natives, we would find a higher than the average crime rate in society. Another element that may generate a higher crime rate among immigrants has an economic rational. Immigrants – particularly if illegal – may decide to engage in criminal activities, since they have a limited alternative options. In other words, an illegal immigrant without a regular job has little to lose from acting criminally. 9.2 The Economic Effects of Migration: Some Empirical Evidence A large volume of economic literature has analyzed the effects of migration on the labor market outcomes for the native population to validate empirically the theoretical predictions discussed in the previous section. This is not an easy task; it is difficult to identify the causal effect of migration on the labor market of the natives and to isolate this effect from other, competing effects, such as globalization and skill-biased technological change. Comparing the labor market outcomes between regions of a country that have had more or less migration may not be enough to disentangle the effect of immigration, as migrants may have chosen a location depending on pre-existing labor market

9.

The Politics of Anger

121

conditions. In this case, labor market conditions would affect migration decisions and not vice versa. To isolate the causal effect of migration on labor market outcomes for domestic workers, some empirical studies have relied on cases in which the migrants (most likely, refugees) were (almost randomly) assigned to different regions of a country. Other studies have instead used an instrumental variable approach, which allows us to take the immigrants’ location decision as exogenous. In both cases, the empirical analysis isolates the effect of migration on labor market outcomes (such as employment and wages) of native workers. Existing empirical evidence largely confirms the theoretical prediction that more immigrants have a positive effect on the wages of skilled workers. Evidence regarding the effects on the wages of unskilled native workers is less robust, but confirms a reduction in wages. However, an interesting research paper by Foged and Peri (2015) suggests that other dynamics may also take place, which are more favourable to the native workers. They analyzed the impact of the arrival of a large number of refugees on the labor market in Denmark. Between 1995 and 2008, 80,000 refugees entered the country fleeing from the Bosnian war, thereby tripling the share of the Danish population born outside the EU. The arrival of the Bosnian refugees created a ladder effect in the Danish labor market and wages, employment and occupational mobility of low-skill native Danes increased. This ladder effect occurred, since the arrival of the refugees stimulated the local economy and, more importantly, pushed low-skilled Danes into more complicated jobs, which complemented the refugees’ basic skills. An important element in this ladder effect was language. While able to perform basic tasks, refugees could not take jobs involving language skills, which remained available to low ability native workers. Evidence on the use of the welfare state by immigrants is more mixed. But anecdotal evidence suggests that large use of local public goods, such as emergency rooms and childcare facilities, by immigrants is an issue for some natives. Finally, there is little evidence of higher crime rates among migrants. Indeed, research by Pinotti et al (2016) that exploits a natural experiment occurred in Italy, where a given number of illegal immigrants could request to be legalized on a first-come-first-serve basis. The legalization procedure required illegal immigrants to register online starting at a specific time. Only a limited number of requests of legalization were available. Given the scheme popularity, it took only a few minutes to reach the maximum available. The analysis by Pinotti et al (2016) compares the behaviour of immigrants whose demand were just on time to be registered with the behaviour of immigrants whose demand just did not went through. Individuals in the first group were “lucky” and got legalized; individuals in the second group were instead unlucky and did not get legalized. The analysis of these two groups shows that the legalized immigrants committed fewer crimes than the non-legalized group. This evidence suggests

122

POLITICAL ECONOMICS

that providing an outside opportunity, such as a legal job, is a big deterrence to criminal behaviour. 9.3 Individual Preferences over Migration Policy The preferences of natives over immigration policies are clearly shaped by the economic effects of immigration described previously as well as by other noneconomic factors, such as political views and cultural elements. A large volume of literature has used survey data to examine how preferences related to immigration policies depend on individual characteristics, such demographics (age, gender, marital status), education, socio-economic status, and labor market status (employed, unemployed, out of the labor force). Figure 3: Individual preferences for migration

Analysis by Mayda (2006) provides a snapshot of these preferences in the midnineties, prior to the financial crisis and the large-scale immigration of the last decade. She uses individual answers to questions about the degree of support for a pro-immigration policy from two large surveys: the 1995 International Social Survey Programme (ISSP), which was run in 22 countries, and the 199597 World Value Survey (WVS) from 44 countries. Her empirical results from the ISSP data (see Figure 3) confirm the relevance of economic factors dis-

9.

The Politics of Anger

123

cussed in the previous section. The more educated an individual the more likely they are to support a pro-immigration policy. If education is a proxy for labor market skills, the concerns of low and high-skilled individuals about the effects of immigration on the labor market are captured by these differences in preferences. Not surprisingly, individuals whose parents are foreign born are more in favour of migration, while individuals who have right wing political views are less in favour. Finally, there is less support for immigration in rural areas and among females. The empirical results from the WVS database confirm the relevance of education as a driving factor for a pro-immigration policy, together with relative income. Affiliation to right wing parties and a strong sense of national pride reduce support. A recent study by Moe Hansen and Legge (2016) provides a more contemporary assessment of these preferences in Europe and examines the interaction between preferences for immigration and welfare policies. The study uses data about immigration and welfare from the European Social Survey (ESS) conducted in 16 countries from 2002 to 2012. Figure 4: Changes in policy preferences between 2002 and 2012

Note: The figure shows changes in the share of survey participants selecting each possible answer to (a) the immigration and (b) the redistribution question in the ESS. The scale is such that, for example, the share of individuals who ‘agree strongly’ with redistribution rose by 6.7 percentage points. The sample is restricted to those sixteen countries which participated in each wave. We do not report changes in respondents who responded “Do not know” or “Refuse to answer”.

As shown in Figure 4, preferences over immigration policies largely polarized during this ten-year period. To the question “how many poor immigrants should be allowed?”, the share of “some” or “few” responses significantly

124

POLITICAL ECONOMICS

dropped, while the answers “many” and “none” strongly increased. Instead, preferences for redistribution unequivocally increased. The share of individuals disagreeing with the statement “the government should redistribute income” dropped, while the share of those strongly agreeing increased. Perhaps, the most interesting part of this study was its analysis of how individual preferences on redistribution and immigration depended on education. These preferences are displayed in Figure 5: the horizontal and vertical axes report the responses to the two previous questions respectively about immigration and redistribution, and the darker shades indicate the preferences of more educated individuals. It shows (in the top-left corner) that more educated individuals prefer less welfare state and more immigration, whereas (on the bottom-right corner) less educated individuals prefer more welfare and less immigration. These empirical results confirm the predictions of our theoretical models. High skilled (more educated) individuals are net contributors to the welfare state, but benefit from immigration. Low skilled (less educated) individuals benefit from the welfare state, but feel economically threatened by immigration. Figure 5: Policy preferences by education

Note: The figure shows the fraction of individuals with high education for each combination of an answer to the question on immigration (x axis) and income redistribution (y axis). We use the ESS data from those sixteen countries that participated in each wave between 2002-2012 and pool the data.

But are preferences over immigration policies exclusively driven by economic factors? Not really! Recent research by Poutvaara and Steinhardt (2016) sug-

9.

The Politics of Anger

125

gests that psychological factors may also play a role. The authors use survey data from the 2005 and 2010 waves of the German Socio-Economic Panel to identify the characteristics of individuals who are more or less concerned about migration. Besides analyzing economic factors, this study considers whether individuals may have different preferences depending on how successful they are in life. The survey, in particular, asks individuals to give an answer to the following statement: “Compared to other people, I have not achieved what I deserve” on a scale from to 1 (totally agree) to 7 (totally disagree). By combining these results with answers from a question about how worried individuals are about immigration to Germany, Figure 6 shows that people who feel that “they have not got what they deserve in life” are more likely to be worried about immigration. Interestingly, this result is stronger for males than for females. Hence, non-economic reasons play a major role in driving opposition to immigration in bitter or unfulfilled people. Figure 6: Bitterness and migration preferences Germany, 2010. Wording of the question on worries about immigration in the SOEP questionnaire: “How is it with the following topic – immigration to Germany – do you have worries about it?” Possible anwers: “Big worries” “Some worries” “No worries”. Wording of the question on bitterness in the SOEP: “Compared to other people, I have not achieved what I deserve.” Answers are on a 7-point scalte, 1 being the lowest value, and 7 denoting total agreement with the statement. n= 10,844.

126

POLITICAL ECONOMICS

9.4 Parties Preferences on Immigration Policy If individuals have strong preferences about immigration policies, it can affect their voting decisions at elections. Hence, parties need to address immigration and come to a clear position. Political science literature analysing party preferences relating to immigration policies has argued that parties no longer have homogenous views on this issue, particularly in the US. For instance, Watanabe and Becerra (2006) suggest that the Republican Party is split. Some leaders want tougher restrictions, including building a wall on the Mexican border as announced by Donald Trump. Other leaders are less opposed to immigration policies, as they fear alienating the Latino vote and business owners who need laborers to fill blue collar jobs in construction, cleaning, gardening and other industries. Also discontent has emerged within the usually proimmigration Democratic Party, with labor unions expressing concerns about the increase in foreign workers damaging their members’ labor market perspectives. To better understand the position of US representatives in Congress from both political parties, a study by Facchini and Steinhardt (2011) has systematically analyzed the drivers of their voting behavior on immigration policy in the period 1970 to 2006. In line with the discussion above, this research aimed to assess the role of economic factors that vary at the district level in influencing the votes of US representatives. Their empirical results show that representatives from districts where higher skilled labor is more abundant, are more likely to support an open immigration policy towards unskilled immigrants; whereas the opposite is true for representatives from districts with an abundance of less skilled labor. Thus, the economic factors underlined in the previous sections also appear to be important drivers at the political level. 9.5 Only Migration Causes Anger? Concerns about immigration seem to be driven mostly by economic factors. Yet, the economic effects of immigration are similar to those induced by globalization and the skill-biased technological change. How can we be sure that the recent “protest votes” and rise of populist movements are a response to immigration challenges only? A recent study by Colantone & Stanig (2016) analyzes the impact of globalization on protectionist and nationalist voting in Europe at a local level from 1980 to 2007. To measure the impact of globalization in the different local regions, they consider the type of manufacturing industry existing at the electoral district level. Some industries are more open to competition from developing countries, as they produce low-value-added goods that rely more intensively on (low-skilled) labor. As the size of Chinese exports strongly increased at the end of the twentieth century (see Figure 7), these industries faced

9.

The Politics of Anger

127

a tough competition from low-cost producers. Clearly, these industries were located in different countries and regions. Hence, within each country, particular regions experienced strong competition from Chinese competitions, while other regions were not so adversely affected. Figure 8 shows a map of Europe with the more affected regions displayed in darker colors. To analyze the political impact of these negative globalization shocks, the authors used the Comparative Manifesto Project to assess the policy position of each party at every election in Europe from 1980 to 2007. Their empirical results show that parties have more protectionist stance in locations where globalization has had a stronger effect and have better electoral outcomes. Their results thus suggest that immigration is not alone in creating anger among low-skilled natives, and that globalization is also responsible. Figure 7: Globalization: imports from China and other countries

This argument is analyzed also by Guiso et al. (2017), who argue that shocks to economic security, which can be originated by several factors – including globalization and skill-biased-technological change – tend to create sharp changes in trust and attitudes towards migration. In particular, they study the recent populist movements, by considering simultaneously the demand and supply of populism. Demand of populism emerges from fears and enthusiasms of people, to which politicians pander. Supply of populism is met by those pol-

128

POLITICAL ECONOMICS

iticians, who claim to be on the side of people and against the elites. Guiso et al. (2017) suggest that the demand and supply of populism lead to short-term protection policies, which have long term costs. In fact, on one hand, facing economic insecurity, (some) people request short-term protection. On the other hand, (some) political parties are willing to promise short-term protection policies, as they endorse a political agenda based on the dichotomy “people vs elites” – and long-termism is considered to be in the interest of the elites. Hence, the politics of anger breeds populism. Figure 8: Globalization: location of the imports shocks

Bibliography

• • • • • • • • • • • •

Aaron, Henry (1966), «The social insurance paradox», Canadian Journal of Economics and Political Science, 32(3), 371–374. Abel, Andrew B. (1989), «Assessing dynamic efficiency: theory and evidence», Review of Economic Studies, 56(1), 1 – 19. Alesina, Alberto, Silvia Ardagna and Vincenzo Galasso, (2010), “The Euro and Structural Reforms”, in A. Alesina and F Giavazzi (eds.) Europe and the Euro, University of Chicago Press and NBER Arrow, Kenneth (1951), Social Choice and Individual Values, New York: John Wiley & Sons. Azariadis, Costas and Galasso, Vincenzo (2002), «Fiscal constitutions», Journal of Economic Theory, 103(2), 255-281. Barr, Nicholas, and Diamond, Peter (2010), «Pension Reform: A Short Guide. » Oxford University Press. Barron, David, and Ferejohn, John (1989), «Bargaining in Legislature», American Political Science Review, 83(4), 1181-1206. Bassanini, A. and E. Ernst (2002), «Labour Market Institutions, Product Market Regulation, and Innovation: Cross-Country Evidence», OECD Economics Department Working Papers No. 316. Bean, C. (1998), «The Interaction of Aggregate-Demand Policies and Labour-Market Reform» Swedish Economic Policy Review 5:2. Becker, Gary S. (1981), «Treatise on the Family», Chicago: University of Chicago Press Becker, Gary S. (1983), «A theory of competition among pressure groups for political influence», The Quarterly Journal of Economics, 98(3), 371-400. Becker, Gary S. (1985), «Public policies, pressure groups, and dead weight costs», Journal of Public economics, 28(3), 329-347.

130

POLITICAL ECONOMICS

• • • • • • • • • • • • • • •

Bentolila, Samuel, and Bertola Giuseppe (1990), «Firing costs and labour demand: how bad is eurosclerosis?» The Review of Economic Studies, 57(3), 381-402. Bertola, Giuseppe, Boeri Tito and Cazes Sandrine (2000), «Employment Protection in Industrialized Countries: the case for New Indicators», International Labour Review, 139(1), 57-72. Besley, Timothy, and Coate, Stephen (1997), «An economic model of representative democracy», The Quarterly Journal of Economics 112(1), 85-114. Blake, David (2006). Pension Economics. New York: John Wiley & Sons. Blanchard, Olivier and Francesco Giavazzi (2003), «Macroeconomic Effects of Regulations and Deregulation in Goods and Labour Markets», The Quarterly Journal of Economics, 118:3, August. Blanchard, Olivier and Jean Tirole (2003), «Contours of Employment Protection Reform», MIT Dept of Economics Working Paper No. 0335. Boadway, Robin W. and Wildasin, David E. (1989), «A median voter model of social security», International Economic Review, 30(2), 307– 328. Boeri, Tito, Micael Castanheira, Riccardo Faini and Vincenzo Galasso (2006), Structural Reforms without Prejudices, Oxford: Oxford University Press. Boeri, Tito, Börsch-Supan Axel, and Tabellini Guido (2001), «Would you like to shrink the welfare state? A survey of European citizens», Economic policy, 16(32), 8-50. Boldrin, Michele and Rustichini, Aldo (2000), «Political equilibria with social security», Review of Economic Dynamics, 3(1), 41–78. Borjas, George J. (2003), «The Labor Demand Curve is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market» Quarterly Journal of Economics, 118(4), 1335-74. Boskin, Michael J, Kotlikoff, Laurence J, Puffert, Douglas J, Shoven, John B. (1987), «Social security: a financial appraisal across and within generations», National Tax Journal, 40, 19– 34. Brennan, Geoffrey, and James M. Buchanan (1980). The power to tax: Analytic foundations of a fiscal constitution. Cambridge: Cambridge University Press. Breyer, Friedrich (1994), «The political economy of intergenerational redistribution», European Journal of Political Economy 10(1), 61–84. Browning, Edgar K. (1975), «Why the social insurance budget is too large in a democracy», Economic Inquiry, 13(3), 373–388.

Bibliography

• • • • •

• • • • • • •

• •

131

Buti, M., A. Turrini, P. van den Noord and P. Biroli (2008), «Defying the Juncker Curse: Can Reformist Governments Be Re-Elected?», European Economy Economic Papers, No. 324, May. Card, David (2001): «Immigrant inflows, native outflows, and the local labor market impacts of higher immigration», Journal of Labor Economics, 19(1), 2-64. Casamatta, Georges, Cremer, Helmuth, Pestieau, Pierre (2000), «The political economy of social security», The Scandinavian Journal of Economics, 102(3), 503-522. Coe, D. and D. Snower (1997), “Policy Complementarities: The Case for Fundamental Labour Market Reform”, IMF Staff Papers Vol.44 Conway, P. and G. Nicoletti (2006) “Product Market Regulation in the Non-Manufacturing Sectors of OECD Countries: Measurement and Highlights,” OECD Economics Department Working Paper No. 530. Cooley, Thomas F. and Soares, Jorge, (1999), «A positive theory of social security based on reputation», Journal of Political Economy, 107(1), 135– 160. Corneo, Giacomo and Grüner, Hans Peter (2002), «Individual preferences for political redistribution», Journal of Public Economics, 83(1), 83-107. Cukierman, Alex and Meltzer, Allan H. (1989), «A political theory of government debt and deficits in a Neo-Ricardian framework», American Economic Review, 79(4), 713– 732. Cukierman, A. and M. Tommasi (1998), “When Does It Take a Nixon to Go to China?”, American Economic Review 88:1. Diamond, Peter (1996), «Proposals to restructure social security», Journal of Economic Perspectives, 10(3), 67– 88. Dixit, Avinash and Londregan John (1996), «The Determinants of Success of Special Interests in Redistributive Politics», Journal of Politics, 58(4), 1132-1155. Dolado, Juan Josè. and Juan Francisco Jimeno (2004), « Contratacion Temporal y costes de despido en España: Lecciones para el futuro desde la perpectiva del pasado », Documento de Trabajo 48, Fundacion Alternativas. Drazen, Alan (2000), Political Economy in Macroeconomics, Princeton: University Press, Princeton. Dustmann, Christian, Uta Schnberg, and Jan Stuhler (2016): “The Impact of Immigration: Why Do Studies Reach Such Different Results?”, Journal of Economic Perspectives, 30(4): 31-56.

132

POLITICAL ECONOMICS

• • • • • • • • • • • • • • • •

Duval, R. (2004), “Retirement Behaviour in OECD Countries: Impact of Old-Age Pension Schemes and Other Social Transfer Programmes”, OECD Economic Studies, No.37. Duval, R. and J. Elmeskov (2005), “The Effects of EMU on Structural Reforms in Labour and Product Markets”, OECD Economics Department Working Papers No. 438 Fernández, Raquel and Dani Rodrik (1991), “Resistance to Reform: Status Quo Bias in the Presence of Individual Specific Uncertainty”, American Economic Review 81:5 Galasso, Vincenzo (2001), «The US social security: a financial appraisal for the median voter», Social Security Bulletin, 64(2), 57-65. Galasso, Vincenzo (2003), «Redistribution and fairness: a note», European Journal of Political Economy, 19(4), 885-892. Galasso, Vincenzo (2006), The Political Future of Social Security in Aging Societies, MIT Press. Galasso, Vincenzo (2014), “The role of political partisanship during economic crises”, Public Choice, 158, pp 143-165. Galasso, Vincenzo, and Profeta Paola (2004), «Lessons for an ageing society: the political sustainability of social security systems. » Economic Policy, 19(38), 64-115. Gramlich, Edward M. (1996), «Different approaches for dealing with social security», Journal of Economic Perspectives, 10(3), 55–66. Grossman, Gene M. and Helpman, Elhanan (1994), «Protection for Sale», The American Economic Review, 84(4), 833-850. Grossman, Gene M. and Helpman Elhanan (2001), Special Interest Politics, Cambridge, Mass.: The MIT Press. Gruber, Jonathan and Wise, David (eds.) (1999), Social Security and Retirement around the World, Chicago: University of Chicago Press. Gruber, Jonathan and Wise, David (eds.), (2004), Social Security Programs and Retirement around the World: Micro Estimation, Chicago: University of Chicago Press. Guiso, Luigi, Helios Herrera, Massimo Morelli and Tommaso Sonno (2017) “Demand and Supply of Populism” CEPR Discussion Paper 11871 Haggard S. and R. R. Kaufman (1992), The Politics of Economic Adjustment: International Constraints, Distributive Conflicts and the State, Princeton: Princeton University Press. Hammond, Peter (1975), «Charity: Altruism or cooperative egoism», In: Phelps, E.S. (Ed.), Altruism, Morality and Economic Theory. New York: Russell Sage Foundation, , pp. 115– 131.

Bibliography

• • • • • • • • • • • •

• • •

133

Hansson, Ingemar and Stuart, Charles (1989), «Social security as trade among living generations», American Economic Review, 79(5), 549– 559. Høj, J., V. Galasso, G. Nicoletti and T. Dang (2006), “The Political Economy of Structural Reform: Empirical Evidence from OECD Countries”, OECD Economics Department Working Papers No. 501. Homburg, Stefan (1991), «Interest and Growth in an Economy with Land.» Canadian Journal of Economics, 24(2), 450-459. Hu, Sheng Cheng (1982), «Social security, majority-voting equilibrium and dynamic efficiency», International Economic Review 23(2), 269– 287. IMF (2004), “Fostering Structural Reforms in Industrial Countries”, chapter III of World Economic Outlook: Advancing Structural Reforms, Washington, DC: International Monetary Fund, , April. Krusell, Per and Rios-Rull Jose-Victor (1999), «On the size of US government: political economy in the neoclassical growth model», American Economic Review, 89(5), 1156-1181. Imrohoroglu, Ayşe, Imrohoroglu, Selahattin and Joines, Douglas H. (1995), «A life cycle analysis of social security», Economic Theory, 6(1), 83–114. Lindbeck, Assar and Weibull Jörgen W. (1987), «Balanced-budget redistribution as the outcome of political competition», Public choice, 52(3), 273-297. Marx, Karl (1867), Capital: Critique of Political Economy, Hamburg: Verlag von Otto Meissner. Meltzer, Allan H. and Richard, Scott F. (1981), «A rational theory of the size of government», Journal of Political Economy, 89(5), 914-927. Mulligan, Casey B. and Sala-i-Martin, Xavier (1999a), «Social security in theory and practice (I): facts and political theories» NBER working paper #7118, Cambridge, Mass. Mulligan, Casey B. and Sala-i-Martin, Xavier (1999b), «Social security in theory and practice (II): efficiency theories, narrative theories, and implications for reform», NBER working paper #7119, Cambridge, Mass. Obstfeld, M. (1997), “Europe’s Gamble”, Brookings Papers on Economic Activity, Washington, DC: The Brookings Institution. OECD (2006), OECD Employment Outlook 2006, Paris: OECD. Olson, M. (1965), The logic of collective action, Cambridge, Mass: Harvard University Press.

134

POLITICAL ECONOMICS

• • • • • • • • • • • • • • •

•

Osborne, Martin J. and Slivinski Al. (1996), «A model of political competition with citizen-candidates», The Quarterly Journal of Economics, 111(1), 65-96. Ottaviano, G.I.P., and Peri, G. (2012). “Rethinking the effect of immigration on wages.” Journal of the European Economic Association, 10(1), 152-197. Persson, Torsten and Tabellini Guido (2000), Political Economics, Cambridge, Mass.: The MIT Press. Piketty, Thomas (2014), Capital in the Twenty-First Century, Cambridge Mass.: Belknap Press. Pissarides, C. (2001), “Employment Protection”, Labour Economics 8:131-59. Roberts, Kevin W.S. (1977), «Voting over income tax schedules», Journal of Public Economics, 8(3), 329-340. Romer, Thomas (1975), «Individual welfare, majority voting and the properties of a linear income tax», Journal of Public Economics, 4(2), 163-185. Rosen, Harvey S. (1991), Public Finance, Richard D. Irwin, Inc. Saint-Paul, Gilles (1996), «Exploring the Political Economy of Labor Market Institutions», Economic Policy, 23, 265-315. Saint-Paul, Gilles (1999a), «The Political Economy of Employment Protection», CEPR Discussion Paper, No. 2109. Saint-Paul, Gilles (1999b), «Assessing the Political viability of Labor Market Reform: the case of Employment Protection», CEPR Discussion Paper, No. 2136. Saint-Paul, Gilles (2000), «The Political Economy of Labor Market Institutions», Oxford: Oxford University Press. Saint-Paul, G. and S. Bentolila (2000), “Will EMU Increase Eurosclerosis?”, CEPR Discussion Paper No. 2423. Samuelson, Paul A. (1958), «An exact consumption-loan model of interest with or without the social contrivance of money», Journal of Political Economy 66(6), 467– 482. Scarpetta, S., P. Hemmings, T. Tressel and J. Woo (2002), “The Role of Policy and Institutions for Productivity and Firm dynamics: Evidence from Micro and Industry Data”, OECD Economics Department Working Paper No. 329, April. Schiantarelli, F. (2005), “Product Market Regulation and Macroeconomic Performance: A Review of Cross Country Evidence”, IZA Discussion Paper No. 1791.

Bibliography

• • • • •

135

Shepsle, Kenneth A. (1979), «Institutional Arrangements and Equilibrium in Multidimensional Voting Models», American Journal of Political Science, 23(1), 27-59. Sjoblom, Kriss (1985), «Voting for social security», Public Choice 45(3), 225– 240. Tabellini, Guido (2000). «A positive theory of social security», Scandinavian Journal of Economics, 102(3), 523– 545. Veall, Michael R. (1986), «Public pensions as optimal social contracts», Journal of Public Economics 31(2), 237– 251. Wright, Randall (1986), «The Redistributive roles of Unemployment Insurance and the Dynamics of Voting», Journal of Public Economics, 31(3), 377-399.

MyBook Political Economics Vincenzo Galasso

T

his book is written for undergraduate students in Economics and in Political Science who want to learn about the political economics of redistributive policies. It provides a positive analysis of the political process behind the design and implementation of redistributive policies, by using the minimum level of mathematical formalization required, in an attempt to be accessible to second- and third-year undergraduate students. The book does not span the entire domain of political economics but concentrates on redistributive policies. This includes monetary transfers, such as pensions and unemployment benefits, as well as regulations, in the labor and product market, which allow to redistribute economic rents. Although the book is mostly self-contained, readers are expected to have a basic knowledge of microeconomics.

Political Economics Vincenzo Galasso

Vincenzo Galasso is Professor of Economics at Bocconi University, where he is the Director of the BSc in International Politics and Government, Research Fellow at the Center for Economic Policy Research (CEPR), Member of the Council of Economic Advisors to the Italian Prime Minister and Editor of the European Journal of Political Economy. He has widely published in top academic journals. He received his PhD in Economics from the University of California at Los Angeles.

A Global Economic Redistributive Policies History

A Global Economic History

Political Economics.indd 1

7 mm

05/09/17 10:31