Fiscal Transparency, Political Parties, Abd Debt in OECD Countries

Abstract

Many believe and argue that fiscal, or budgetary, transparency has large, positive effects on fiscal performance. However, the evidence linking transparency and fiscal policy outcomes is less compelling. To analyze the effects of fiscal transparency on public debt accumulation, we present a career-concerns model with political parties. This allows us to integrate as implications of a single model three hitherto-separate results in the literature on deficit and debt accumulation: that transparency decreases debt accumulation (at least by reducing an electoral cycle in deficits), that right-wing governments (at least for strategic reasons) tend to have higher deficits than left- wing governments, and that increasing political polarization increases debt accumulation. To test the predictions of the model, we construct a replicable index of fiscal transparency on 19-country OECD data. Simultaneous estimates of debt and transparency strongly confirm that a higher degree of fiscal transparency is associated with lower public debt and deficits, independent of controls for explanatory variables from other approaches.

1 We thank David Skilling for participating in the early stages of this project and Alberto Alesina, Morten Bennedsen, Tim Besley, Carles Boix, Bob Inman, Casey Mulligan, Jonathan Rodden, Christian Schultz, Ken Shepsle, David Stromberg, Guido Tabellini, and participants in the Fiscal Federalism Conference at CES, Harvard, the Political Economy Workshop at Harvard, the 97th Annual Meeting of the American Political Science Association (San Francisco), ECPR (Canterbury), and the Danish Public Choice Workshop III for very useful comments and suggestions. This work was begun when Lassen was visiting Harvard. He would like to thank the Economic Policy Research Unit, the Danish Research Academy and a Sasakawa International Fellowship for funding, and Harvard for its hospitality. The work was completed under an EPRU-network grant. * jalt@latte.harvard.edu ** david.dreyer.lassen@econ.ku.dk

Fiscal Transparency and Fiscal Policy Outcomes

1. Introduction

Many believe that fiscal transparency has large and positive effects on fiscal performance. According to the IMF, “transparency in government operations is widely regarded as an important precondition for macroeconomic fiscal sustainability, good governance, and overall fiscal rectitude” (Kopits and Craig 1998: 1). Both the IMF and the OECD have recently developed Codes of Best Practice for Fiscal Transparency. Fiscal adjustment programs (like some aimed at satisfying Maastricht Treaty criteria) can employ or produce creative accounting practices. Transparency can affect the probability that such budgetary tricks are revealed. In ways like that, more transparency leads to lower budget deficits and makes fiscal discipline and control of spending easier to achieve. However, while such asserted effects are common, there is not much empirical evidence about institutional transparency and fiscal policy outcomes. Some links appear between fiscal transparency and fiscal performance in European countries, and between indirect measures of transparency and fiscal performance in Latin American countries.Fiscal Transparency and Fiscal Policy Outcomes 1. Introduction Many believe that fiscal transparency has large and positive effects on fiscal performance. According to the IMF, “transparency in government operations is widely regarded as an important precondition for macroeconomic fiscal sustainability, good governance, and overall fiscal rectitude” (Kopits and Craig 1998: 1). Both the IMF and the OECD have recently developed Codes of Best Practice for Fiscal Transparency. Fiscal adjustment programs (like some aimed at satisfying Maastricht Treaty criteria) can employ or produce creative accounting practices. Transparency can affect the probability that such budgetary tricks are revealed. In ways like that, more transparency leads to lower budget deficits and makes fiscal discipline and control of spending easier to achieve. However, while such asserted effects are common, there is not much empirical evidence about institutional transparency and fiscal policy outcomes. Some links appear between fiscal transparency and fiscal performance in European countries, and between indirect measures of transparency and fiscal performance in Latin American countries.

Many remain convinced of the importance of fiscal transparency, however.2 The purpose of this paper is to investigate whether a higher degree of fiscal transparency is in fact associated with lower public debt, other things equal. To do this, we extend the career- concerns model of public debt developed by Persson and Tabellini (2000) and, in particular, Shi and Svensson (2002) to include political parties with preferences over public spending. We show that this allows us to integrate as implications of a single model three hitherto-separate results in the literature on deficit and debt accumulation: that transparency decreases debt accumulation, at least partly through an effect on the electoral cycle (Shi and Svensson, 2002), that increasing political polarization increases debt accumulation (see, for instance, Alesina and Tabellini, 1990), and that right-wing governments, at least for strategic reasons, tend to have higher deficits than left-wing governments (Persson and Svensson, 1989). We also develop measures of fiscal transparency and show that recent evidence from OECD countries is consistent with these implications, even after controlling for the effects of explanatory variables from other political- economic models of debt and deficits, and accounting for the potential endogeneity of transparency.

2 On Europe, see von Hagen (1992), de Haan et al. (1999), Milesi-Ferretti (forthcoming), and Hallerberg et al. (2001). On Latin America, see Alesina et al. (1999).

The theoretical literature on the causes and consequences of fiscal, or budgetary, transparency is not large.3 In the Shi and Svensson (2002) political agency model that we extend below, voters want more competent politicians in office, as they can provide more public goods for given levels of taxation and private consumption. However, this creates incentives for incumbents to try to “appear competent” by issuing debt, providing more public goods by ‘buying now and paying later’. In the model, the degree of fiscal or budget transparency determines when and how far voters can observe debt, and thus the extent to which an incumbent can use debt to appear competent. Milesi-Ferretti (forthcoming), on the other hand, considers in a reduced-form model the effect of transparency on government debt and deficits in a regime characterized by fiscal rules, allowing for creative accounting practices like those arising in connection with the Maastricht Treaty (Easterly 1999).

Transparency, Milesi-Ferretti argues, affects the probability that such practices are revealed, resulting in a penalty for not meeting the formal budget rule requirement. Thus, transparency determines the scope for creative accounting vs. “true” fiscal adjustment, and matters only because of the existence of fiscal rules. Finally, Ferejohn (1999) examines an agency model in which fiscal transparency affects voter trust in government and thus the size of government. In all these cases, transparency increases the probability or accuracy of observations of incumbents’ performance. The empirical literature on transparency is also limited, in part by measurement problems.4 So, in addition to providing a model that synthesizes and unifies several disparate results on debt, we construct a direct, replicable index variable measuring the transparency of budget processes of OECD countries. The index contains variables comparable though not identical to ones collected and analyzed in the American states (Alt, Lassen, and Skilling 2002). Further, we use this index to investigate empirically the hypotheses of the model. We find that fiscal transparency is, indeed, robustly associated with lower public debt and deficits, even after allowing for the effects of partisanship and polarization. However, we also recognize that fiscal institutions are subject to change, and investigate the reasons why governments change fiscal 3 Asymmetric information models of fiscal policy have been studied by, e.g., Rogoff (1990), but a direct focus on the role of fiscal transparency is rare. Besley and Prat (2001) touch on transparency in their investigation of the role of the media in communicating information about the government to voters. The effects of transparency about monetary policy objectives has begun to be studied as well; see, e.g., Faust and Svensson (2001), Jensen (2002), and Stasavage (2003).

4 Alesina and Perotti (1996) note that the “results on transparency probably say more about the difficulty of measuring it, than about its effect on fiscal discipline”, a point echoed in Alesina and Perotti (1999) and Tanzi and Schuknecht (2000). 2 Fiscal Transparency and Fiscal Policy Outcomes transparency. This allows us to correct transparency for possible endogeneity in the empirical analysis. Estimating the simultaneous empirical model leaves the main results unaltered. Finally, many other analyses of public debt focus on the number of actors involved in the budget process. Two conjectures predominate: either there is a “common pool problem” so that actors do not internalize the full cost of their spending or there is a “fragmentation problem” so that they cannot coordinate, for instance on a response to negative shocks. “Actors” include the number of spending ministers, parties in a governing coalition, decentralized units in a federal system, or veto players.5 Our model does not make specific predictions about these other variables, so in the empirical analysis we control for as many alternative approaches as possible. The paper proceeds as follows. Section 2 defines fiscal transparency. Section 3 summarizes the career-concerns model of fiscal transparency with competing political parties, deriving results for the effects of transparency, and partisanship and polarization from the “strategic debt” literature, to guide the empirical analysis. (We present the full model in an appendix.) Section 4 describes the construction of the transparency index, as well as other data, used in the empirical work. Section 5 examines the effects of fiscal transparency on fiscal performance, the causes of variation in fiscal transparency across the OECD, and the possible endogeneity of transparent institutions with respect to debt. Section 6 concludes.

2. Defining Fiscal Transparency

Greater transparency eases the task of attributing outcomes to the acts of particular politicians. It makes observers more able to distinguish effort from opportunistic behavior or stochastic factors “primarily by providing actors with greater or lesser degrees of certainty about the present and future behavior of other actors” (Hall and Taylor 1996, p. 939).6 With respect to the budgetary process, a comprehensive definition of fiscal transparency is the following: “Fiscal transparency is defined … as openness toward the public at large about government structure and functions, fiscal policy intentions, public sector accounts, and 5 Proposed remedies include delegation to a strong central Ministry of Finance (when there is no problem of ideological heterogeneity) and a form of commitment among coalition partners when there is (Hallerberg et al. 2001). On the effects of number, possibly conditional on decentralization and heterogeneity, of ministers and parties see Kontopoulos and Perotti (1999) and Volkerink and de Haan (2001); on decentralized units in a federal system see Rodden and Wibbels (2002); and on veto players see Tsebelis (2002).

Outcomes projections. It involves ready access to reliable, comprehensive, timely, understandable, and internationally comparable information on government activities … so that the electorate and financial markets can accurately assess the government’s financial position and the true costs and benefits of government activities, including their present and future economic and social implications” (Kopits and Craig 1998: 1).7 The literature also provides specific examples of transparent budget reporting procedures: “A transparent budget process is one that provides clear information on all aspects of government fiscal policy. Budgets that include numerous special accounts and that fail to consolidate all fiscal activity into a single ‘bottom line’ measure are not transparent. Budgets that are easily available to the public and to participants in the policymaking process, and that do present consolidated information, are transparent” (Poterba and von Hagen 1999: 3-4). As features of non-transparent financial reporting, Alesina and Perotti (1996) identify optimistic predictions on key economic variables and forecasts of the effects of new policies, and creative and strategic use of what is kept on or off budget, budget projections, and multi-year budgeting. We believe more transparent procedures have four distinct characteristics. First, more transparent procedures should process more information, and, other things equal, do so in fewer documents. This speaks to openness and ease of access and monitoring. Second, the possibility of independent verification, which has been shown experimentally to be a key feature in making communication persuasive and/or credible, increases transparency. Third, there should be a commitment to non-arbitrary language: words and classifications should have clear, shared, unequivocal meanings. The use of generally accepted accounting principles in some of the American states is a good example of this. Finally, the presence of more justification increases transparency, reducing the optimism and strategic creativity referred to above. Below we operationalize multiple indicators of these characteristics into an index of budget transparency.

Cite this page