Political leaders' socioeconomic background and fiscal performance in Germany

https://doi.org/10.1016/j.ejpoleco.2014.01.009Get rights and content

Highlights

  • Impact of the socioeconomic status of the head of government on fiscal policy.

  • Sample encompasses the prime ministers of the German Laender.

  • Low-status prime ministers are associated with higher public spending and deficits.

  • Social mobility has an asymmetric influence on fiscal behaviour.

Abstract

This paper investigates whether the socioeconomic status of the head of government helps explain fiscal performance. Applying sociological research that attributes differences in people's ways of thinking and acting to their relative standing within society, we test whether the social status of German prime ministers can help explain differences in fiscal performance among the German Laender. Our empirical findings show that the tenures of prime ministers from a poorer socioeconomic background are associated with higher levels of public spending and debt financing. Social mobility has an asymmetric influence: social climbers adapt to their new class, whereas downwardly mobile prime ministers remain primarily influenced by their parents' upper-class status.

Introduction

Explaining variations in government size and public debt accumulation is an important issue for political economists, and one for which the motives of relevant political actors are thought to play a decisive role. The literature typically assumes that political decision-making is driven by one of three motivations. First, politicians behave in a purely opportunistic manner. For instance, political budget cycle (PBC) theorists argue that public spending and debt financing are connected to the legislation cycle (e.g., Rogoff and Sibert, 1988, Alesina et al., 1992)1: to enhance their re-election prospects, politicians are expected to raise the level of public expenditure and debt financing in election years, while fiscal consolidation is expected to occur in the aftermath of an election.2

A second branch of the literature views politicians as purely benevolent. Political actors design fiscal policy to optimise some sort of social welfare function. A well-known example is Barro's (1979) tax-smoothing hypothesis, in which it is assumed that governments choose tax rates with the aim of minimising the excess burden of taxation. As a consequence, the amount of public debt is expected to be linked to the business cycle, since governments are assumed to incur deficits during recessions and accumulate surpluses during booms (Alesina and Perotti, 1994).

Finally, there are political economists who link political performance to partisan ideology (see, e.g., Hibbs, 1977, Buchanan and Wagner, 1977). A common assumption in this body of literature is that the tenures of left-wing governments are associated with higher levels of public spending and debt financing than are the tenures of right-wing governments (e.g., Blais et al., 1993, Cusack, 1997).

However, empirical analyses cast serious doubt on all three theoretical approaches, since findings from different studies are often contradictory and the employed covariates have low explanatory power. In the case of PBC theory, for example, Shi and Svensson (2006) find robust evidence for PBCs in fiscal deficits for developing countries, but not for developed countries. Brender and Drazen (2005) provide similar results based on a differentiation between new and established democracies: PBCs are found in the former only.

Results from the two other strands are, at best, mixed, too. With respect to the partisan hypothesis, there is some evidence from OECD countries that the tenures of left-wing governments are indeed associated with an increase in public expenditure (e.g., De Haan and Sturm, 1994, De Haan and Sturm, 1997, Cusack, 1997), but the conclusion that leftist incumbents accumulate higher public debt becomes less compelling in light of the experience of many Western European countries between 1960 and 1990.

Finally, the tax-smoothing approach is not in line with public budgets in many OECD countries during the 1970s and 1980s (Roubini and Sachs, 1989, Alesina and Perotti, 1994). In particular, the absence of fiscal adjustments in some OECD countries after the recession of 1973–1974 and differences in the accumulation of public debt in the following decades cannot be convincingly explained by a tax-smoothing motive.

So, what is driving the fiscal decisions of political actors? In this paper, we argue that a broader social science perspective may provide some important insights. Sociologists emphasise the strong connection between an individual's socioeconomic background—especially factors that determine an individual's relative position within society, called status—and preferences, attitudes, and habits. There is a considerable amount of political economics literature suggesting that factors related to individuals' status—such as occupation, income, and education—help explain differences in policy preferences and decision behaviour, but typically these variables are employed in a more or less ad hoc fashion. To the best of our knowledge, there is no empirical work that links theoretical research on status to the fiscal behaviour of leaders. We fill this gap by investigating the impact that a political actor's family status has on his or her fiscal preferences. To this end, we utilise observations on the prime ministers (Ministerpräsidenten) of the fiscally partially autonomous states making up the Federal Republic of Germany, the Laender (Bundesländer). We concentrate on the social status of the German Laender's prime ministers and study its influence on the level of public expenditure as well as on debt financing. We compile a unique dataset containing information about German prime ministers' parental and personal status, along with several additional sociodemographic variables. We cover the period from 1974 to 2009 for the West German Laender and from 1992 to 2009 for the East German Laender.

In our analysis, we find strong and theory-consistent evidence that a prime minister's socioeconomic background matters in terms of the incumbent government's fiscal performance: the higher a prime minister's status, the lower the public expenditures and debt financing. We also discover an asymmetric effect of social mobility: whereas there is no statistical difference between upwardly mobile prime ministers—i.e., prime ministers with low parental status, but high personal status—and upper-class prime ministers—i.e., prime ministers with high parental status and high personal status—with respect to the levels of public spending and deficit financing, downwardly mobile prime ministers appear to spend significantly less and incur fewer debts than their upper-class counterparts.

Methodologically, our empirical approach is based on the application of a consistent general-to-specific modelling strategy, which has several advantages compared to the commonly employed specific-to-general modelling approach (see Hendry, 2000). Several robustness checks reveal that the impact of leaders' socioeconomic background on the Laender's fiscal performance appears to be much more important than that of political factors proposed by ‘traditional’ political economic approaches. This supports our belief that economics may benefit substantially from integrating social science research.

Choosing the German Laender as objects of analysis has three advantages. First, they are highly homogeneous with regard to politics, culture, demography, and institutional as well as constitutional frameworks. Second, this homogeneity limits the number of necessary control variables. Third, the Laender are constitutionally endowed with a high degree of fiscal authority regarding budgetary matters. These advantages make the German Laender a popular object of research into the motivations of incumbent governments. However, neither the PBC theorem (Berger and Holler, 2007, Schneider, 2010), the tax-smoothing approach (Seitz, 2000), nor the partisan hypothesis (Seitz, 2000, Galli and Rossi, 2002, Jochimsen and Nuscheler, 2011, Schneider, 2010) is convincingly supported by the results of these empirical studies.

In the next section of this paper, we discuss related literature. In Section 3, we introduce the concept of status, provide empirical measures for it, and identify transmission channels that link a person's status to his or her preferences for public expenditure and debt financing. Then, in Section 4, we take a brief look at the German system of fiscal federalism and its political landscape. In Section 5, we motivate our empirical approach. Section 6 presents the results of our main analysis, in which we investigate the influence of German prime ministers' socioeconomic background on the fiscal performances of the German Laender. Section 7 concludes.

Section snippets

Related literature

By focussing on how the head of government influences economic outcomes, this paper contributes to an expanding branch of the literature. Starting with the work of Jones and Olken (2005), researchers have become increasingly concerned with the question of whether politicians exert an impact on economic performance. Jones and Olken (2005) provide evidence that leaders do influence the rate of economic growth in their country. They compare GDP growth rates before and after exogenous leader

The status concept, its measurement, and implications for individual behaviour

In this section, we clarify the concept of status, discuss its implications for motives and patterns of behaviour, and show how it can be measured. First, we provide a definition of status, discuss quantitative status indicators widely used in many areas of sociology, and briefly describe how status influences individual modes of thinking and acting. Second, we present theoretical and empirical evidence linking attributes of status with economic behaviour and fiscal preferences.

Fiscal federalism and political landscape in Germany

The German federal system consists of four parliamentary governmental levels—federal, state, district, and community—each with its own fiscal competences and responsibilities, which are specified by the German Constitution (the Grundgesetz).10 At the state level, there

Data and empirical approach

Our research hypothesis is that the tenures of low-status German prime ministers are associated with (i) higher public expenditure and (ii) larger public deficits. We rely on two empirical strategies to test this hypothesis. First, we apply a two-step approach. In Step 1, we estimate the following dynamic panel data (DPD) model:yi,t=αi+μt+ρyi,t1+βeconomicvariablesi,t+γpoliticalvariablesi,t+δdemographicvariables+εleaderdummiesi,t+ζi,t.

We use two dependent variables in our analysis: (i)

Main specifications

We commence our analysis with the results of the two-step approach. To economise on space, we report only the results of Step 2, i.e., Eq. (2). The estimates of the regressions linking characteristics of German prime ministers to public spending in the German Laender are presented in Table 2, those for net public borrowing in Table 3. In both specifications, we start by estimating a general model containing all the prime minister characteristics described in Section 5. Then, we eliminate

Conclusion

Political economy theories that view politicians' motives as decisive in fiscal performance often have little explanatory power in empirical research. Neither opportunism, benevolence, nor partisan ideology satisfactorily explains politicians' decision-making. In light of this, some economics researchers have begun focussing on politicians' backgrounds instead, taking into account age, experience, or certain educational and occupational characteristics. However, yet again the results are

References (89)

  • J.F. Kiviet

    On bias, inconsistency, and efficiency of various estimators in dynamic panel data models

    J. Econ.

    (1995)
  • S.E.G. Lea et al.

    The economic psychology of consumer debt

    J. Econ. Psychol.

    (1993)
  • S.E.G. Lea et al.

    Psychological factors in consumer debt: money management, economic socialization, and credit use

    J. Econ. Psychol.

    (1995)
  • S.M. Livingstone et al.

    Predicting personal debt and debt repayment: psychological, social and economic determinants

    J. Econ. Psychol.

    (1992)
  • C.A. Shelton

    The size and composition of government expenditure

    J. Public Econ.

    (2007)
  • M. Shi et al.

    Political business cycles: do they differ across countries and why?

    J. Public Econ.

    (2006)
  • C.M. Walker

    Financial management, coping and debt in households under financial strain

    J. Econ. Psychol.

    (1996)
  • A. Agarwal et al.

    Social roots and psychological implications of time preference

    Int. J. Psychol.

    (1983)
  • A. Alesina et al.

    Preferences for redistribution

  • A. Alesina et al.

    The political economy of budget deficits

  • A. Alesina et al.

    Macroeconomic policy and elections in OECD democracies

    Econ. Polit.

    (1992)
  • G.-M. Angeletos et al.

    The hyperbolic consumption model: calibration, simulation, and empirical evaluation

    J. Econ. Perspect.

    (2001)
  • M. Arellano et al.

    Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations

    Rev. Econ. Stud.

    (1991)
  • R. Barro

    On the determination of public debt

    J. Polit. Econ.

    (1979)
  • G.S. Becker et al.

    The endogenous determination of time preference

    Q. J. Econ.

    (1997)
  • H. Berger et al.

    What determines fiscal policy? Evidence from German states

  • P.L. Berger et al.

    The social construction of reality: a treatise in the sociology of knowledge

    (1966)
  • T. Besley et al.

    Do educated leaders matter?

    Econ. J.

    (2011)
  • A. Blais et al.

    Do parties make a difference? Parties and the size of government in liberal democracies

    Am. J. Polit. Sci.

    (1993)
  • A.S. Blinder et al.

    Public opinion and the balanced budget

    Am. Econ. Rev. Pap. Proc.

    (1984)
  • A.H. Börsch-Supan et al.

    Household saving in Germany: results of the first SAVE study

  • P. Bourdieu

    Outline of a Theory of Practice

    (1977)
  • P. Bourdieu

    Distinction: A Social Critique of the Judgement of Taste

    (1984)
  • P. Bourdieu

    The forms of capital

  • P. Bourdieu et al.

    An Invitation to Reflexive Sociology

    (1992)
  • R.H. Bradley et al.

    Socioeconomic status and child development

    Annu. Rev. Psychol.

    (2002)
  • T. Brambor et al.

    Understanding interaction models: improving empirical analyses

    Polit. Anal.

    (2006)
  • R. Breen

    Risk, recommodification and stratification

    Sociology

    (1997)
  • G.S.F. Bruno

    Estimation and inference in dynamic unbalanced panel-data models with a small number of individuals

    Stata J.

    (2005)
  • J. Buchanan et al.

    Democracy in Deficit

    (1977)
  • C. Büchner et al.

    Influences on socio-economic status in Germany: reproduction of social structure or modes of individual acting?

  • C.D. Carroll et al.

    On the concavity of the consumption function

    Econometrica

    (1996)
  • A. Cukierman et al.

    A political theory of government debt and deficits in a neo-Ricardian framework

    Am. Econ. Rev.

    (1989)
  • T. Cusack

    Partisan politics and public finance: changes in public spending in the industrialized democracies 1955–1989

    Public Choice

    (1997)
  • Cited by (51)

    • Political budget cycles revisited: Testing the signalling process

      2021, European Journal of Political Economy
      Citation Excerpt :

      He reports that low-status individuals have a higher probability of re-electing incumbents who run deficits, whereas he finds no effect for high-status individuals. Hayo and Neumeier (2014) provide evidence that low-status (high-status) prime ministers of German states tend to run relatively higher (lower) deficits. Thus, an alternative explanation could be that low-status voters generally prefer low-status leaders.

    • National or sub-national parties: Does party geographic scope matter?

      2020, Journal of Development Economics
      Citation Excerpt :

      Similarly, Brollo and Troiano (2016) find that, in Brazil, female mayors tend to be less corrupt, while Gilardi (2015) finds that the election of female politicians increased future female representation in Switzerland. Other studies have examined personal attributes such as age (Alesina et al., 2019), business experience (Beach and Jones, 2016; Szakonyi, 2018), ethnicity (Franck and Rainer, 2012) and race (Broockman, 2013; Logan, 2018), socioeconomic background (Hayo and Neumeier, 2014), tenure (Coviello and Gagliarducci, 2017), and children's gender (Washington, 2008). Our paper also relates to the literature (mostly from political science) examining the process of ‘denationalization’ of electoral politics (i.e., the raise of sub-national parties), and, more broadly, the importance of taking into account the ‘territorial dimension’ of electoral politics (Hopkin, 2003; Caramani, 1996; Jones and Mainwaring, 2003).

    • Educated dictators attract more foreign direct investment

      2020, Journal of Comparative Economics
    • Empowering knowledge: Political leaders, education, and economic liberalization

      2020, European Journal of Political Economy
      Citation Excerpt :

      Brender and Allan (2013) report a large medium-term effect of national leadership turnovers on change in the composition of public expenditure for democratic countries. Hayo and Neumeier (2014) focus on Germany and find that the pattern of fiscal spending is correlated with the prime ministers’ initial social status. For subnational politicians, Neumeier (2018) is in accordance with Dreher et al. (2009) in showing that leaders with business experience enhance economic performance.

    View all citing articles on Scopus

    Thanks to Edith Neuenkirch, Matthias Neuenkirch, Britta Niehof, Gunnar Otte, Matthias Uhl, participants of a research seminar at the Helmut-Schmidt University Hamburg and the annual Scottish Economic Society Conference in Perth as well as two anonymous referees for their helpful comments on earlier versions of the paper. We are also obliged to Beate Jochimsen and Sebastian Thomasius for sharing their data with us. The usual disclaimer applies.

    View full text