Sovereign Risk, Public Investment and the Fiscal Policy Stance

https://doi.org/10.1016/j.jmacro.2020.103263Get rights and content

Abstract

Of the different types of government outlays, since the 2000s public investment has been the main variable of adjustment during recessions in advanced and emerging economies. These contractions (expansions) have been associated with relatively medium-high (low) sovereign spreads, especially in advanced economies. To rationalize these issues, we develop a model of fiscal policy and sovereign default, with corporate default risk. Policymakers must decide between the provision of an unproductive public good and public investment, weighting their respective net benefits in terms of short-term stabilization and debt sustainability. In our model, investment follows a countercyclical stance only in the case of low levels of debt and moderate negative shocks, and otherwise contracts during recessions. The policy stance, along with the mix between different outlays, is determined by how sovereign risk responds to adverse economic shocks.

Introduction

In the context of financial crises, the burden of sovereign debt affects macroeconomic outcomes and policy responses. Romer and Romer (2019) show that OECD countries with lower debt-to-GDP ratios present smaller recessions than countries with higher leverage ratios, thanks to a more aggresive countercyclical fiscal response.1 In this paper, we analyze the behavior of different types of government expenditures around the time of economic downturns, since the begining of the century. In advanced and emerging economies, public investment is the main variable of adjustment, while other types of outlays do not conform to the same procyclical pattern. Then, we provide evidence of public investment contractions when sovereign spreads are relatively high, and of expansions under the opposite scenario, especially in advanced economies. To shed light on these issues, we use a fiscal policy model of sovereign debt and default in which credit to firms is risky, and study how the response of public investment during difficult times depends on the size of the debt-to-GDP ratio and the severity of the shock. In our model, sovereign risk plays a key role in determining a procyclical or countercyclical public investment stance.

In the first section of the paper we look at the evolution of different categories of public expenditures (over output) around recession windows (centered on output troughs). The sample includes a group of advanced and emerging markets, with data availability for public investment and sovereign spreads. Our exploration of recession windows shows that at the output trough, on average, the ratio of public investment over output falls by 0.32 percentage points; resources allocated to social benefits and subsidies rise by 1.35 percentage points; and we do not find conclusive results in the case of compensation to employees. These three categories account for approximately two-thirds of public outlays.

The contraction in public investment occurs between one year before the output trough and at the trough itself. On this basis, we calculate the change in government outlays between two points in time: two years before the trough, and at the trough. Then we analyze whether these variations were correlated with a proxy of the initial fiscal space (we use five-year sovereign CDS spreads at two periods preceding the trough). We find that before economies hit the bottom of the recession: (i) government spending cuts (as a percentage of output) are implemented mainly via public investment (reductions in the ratios of social benefits-subsidies and wages are unusual events); (ii) public investment cuts are associated with relatively high sovereign risk (limited or no fiscal space), while public investment expansions are implemented when sovereign risk is low (which indicates the existance of fiscal space); and (iii) in advanced economies with little fiscal space, investment cuts give room for the expansion of social benefits-subsidies. Our results for advanced economies are clearly driven by the stimulus experience during the Great Recession, while the contractionary public investment episodes correspond to recessions postdating 2009. This change of policy stance has been widely discussed; see for example Alesina et al. (2015) or Callegari et al. (2017).

In the following sections, we build a model of fiscal policy and sovereign default to account for these facts. A first key element of the model is that sovereign spreads are endogenous, since governments can choose to repay or default on their previous obligations and international investors price long-term sovereign bonds based on their expectations about repayment (Eaton, Gersovitz, 1981, Aguiar, Gopinath, 2006, Arellano, 2008, Chatterjee, Eyigungor, 2012). In fiscal policy models of sovereign default (Cuadra, Sanchez, Sapriza, 2010, Hatchondo, Martinez, Roch, 2015, Arellano, Bai, 2017, Roch, Uhlig, 2018) governments pursue austerity when things turn bad, via cuts in the provision of a public good. The second key element, and the main difference between our model and the literature on sovereign default, is the introduction of public capital and financial frictions in the private sector of the economy.

The inclusion of public capital, in addition to the standard element of public consumption, affects the set of goals that policymakers might pursue. Now, governments face the tradeoff between the provision of an unproductive public good (short-term welfare stabilization) and productive public spending (with potential medium run benefits in terms of debt stabilization). Since public capital is an input of production for private firms, investing more today raises future private production and future tax collection, thus increasing the chances of repayment. However, the decision to implement a countercyclical fiscal policy must account for the negative effects of the recession, which, in addition to the fact that more investment might be debt-financed, increases the burden of future repayment obligations, reducing the chances of repayment. Hence, in the model, the reaction of sovereign spreads reveals whether the potential debt stabilization benefits outweight the associated costs of the increase in public investment. Asonuma and Joo (2020) also include public capital in a fiscal policy model of default with debt restructurings, but they do not aim to understand the heterogeneity of the policy stance, or its relation with spreads.

The aforementioned government tradeoff must also account for the potential effects on the private sector. In our model, the private sector side takes into account not only public capital, but also the possibility that heterogeneous firms borrow in international markets to finance the wage bill. Firms differ because they are subject to an idiosyncratic productivity shock. Following the current literature on sovereign and corporate debt, a group of firms with very low shock realizations default on their obligations. Hence, the model allows for a negative feedback loop between sovereign and corporate risk, as in de Ferra (2018) and Kaas et al. (2020). In these two papers the link between spreads is attributable to taxation (with an increase in tax rates during downturns), while in our setup it is public investment that plays the major role.2

The results of the model show that governments can implement a countercyclical fiscal policy if the burden of debt is not high and negative shocks are not that strong. Because initial debt levels are low, there is fiscal space to increase government debt and finance more investment and the public good as a response to a negative shock, without increasing sovereign spreads to prohibited levels. More capital tomorrow yields more production and tax collection, and this effect outweighs the burden of higher repayment obligations tomorrow. A different scenario arises if initial debt levels are higher or the recession is too deep; when this occurs, a deterioration in economic conditions triggers a strong increase in spreads. The optimal government response is to aggresively cut public investment, in order to smooth out the drop in public consumption. These two scenarios provide a rationale for the behavior of public investment and social benefit-subsidy expenditures in advanced economies during the recession of 2009 (with fiscal space) and the subsequent economic contractions (around 2013 and with less fiscal space).

The third scenario in the model corresponds to an economy with relatively large repayment obligations subject to a significant unexpected negative shock. As in any model of sovereign default, the government ends up defaulting on its obligations. Under any of the previous scenarios, the model yields a positive association between sovereign spreads and corporate risk premium. This is consistent with the evidence related to both types of yields presented by Agca and Celasun (2012), Klein and Stellner (2014), and Augustin et al. (2018), among others. Moreover, as Kaas et al. (2020) shows, the feedback mechanism brings the model statistics closer to the data.

A question worth highlighting is why the proposed mechanism is not present in most of the fiscal policy models of sovereign default (Cuadra, Sanchez, Sapriza, 2010, Hatchondo, Martinez, Roch, 2015, Arellano, Bai, 2017, Roch, Uhlig, 2018). Intuitively, this is simply because increasing public consumption in this setup does not yield important benefits in terms of debt stabilization (see Bianchi et al. (2019) for a model in which this statement does not hold); or public consumption mainly affects the expenditure side, but not the revenue side, of the government’s budget constraint. This point is relevant because the tradeoff between short-term stabilization and debt sustainability has been present in the recent fiscal policy discussions for the Euro area (E.Commission, 2016, ECB, 2016, Ferdinandusse, Checherita-Westphal, Attinasi, Lalouette, Bańkowski, Palaiodimos, Campos, 2017, IMF, 2017, IMF, 2019, Campos, Checherita-Westphal, Jacquinot, Burriel, Caprioli, 2019). Our contribution takes one step toward understanding this tradeoff, since we abstract from an important closely related channel: the effect on private investment and growth.

We contribute to the sovereign debt literature by showing the debt dependence of the fiscal policy stance, via different public investment behaviors. Our work is along similar lines to two preceding papers. Bianchi et al. (2019) rationalize the debt dependence of the fiscal policy stance through a sovereign default model with heterogenous households, downward nominal wage rigidities, a fixed exchange rate, and optimal fiscal policy. They show that for low debt, a debt-financed stimulus via public consumption yields positive aggregate demand effects (reduction in unemployment and inequality) that outweigh the costs in terms of sovereign risk. One difference between Bianchi et al. (2019) and our paper, is that public consumption stabilization benefits capture the textbook Keynesian mechanism, while we aim to model a more direct channel in terms of debt stabilization. Anzoategui (2019) does not model optimal fiscal policy, and estimates fiscal rules using historical data in the context of a sovereign default model. The author establish a set of conditions under which fiscal austerity policies can worsen recessions.

Moreover, our paper follows the contribution of Gordon and Guerron-Quintana (2018), who develop a sovereign default model for a small open economy with aggregate capital (using a planner’s version). In their setup, fiscal policy is absent and the main focus is on the business cycle properties. In addition, they discuss how aggregate capital accumulation yields counteracting forces that affect the default decision. In a paper closely related to ours, Asonuma and Joo (2020) study the evolution of public investment and consumption around debt settlements. Their model captures the sovereign debt overhang in terms of public investment, showing its severe decline and slow recovery up to the point of restructurings. When it comes to their empirical facts, they observe that recessions associated with default episodes present deeper and more protracted declines in public investment than non-crisis recessions—although they do not present evidence on sovereign spreads.

In addition to sovereign default papers, our work is closely related to the literature on fiscal procyclicality in emerging markets (Gavin, Perotti, 1997, Talvi, Vegh, 2005, Kaminsky, Reinhart, Végh, 2004, Ilzetzki, Vegh, 2008), providing a mechanism that explains the heterogenity of fiscal responses during recent decades (Frankel et al. 2013). Naturally, there is also a connection with the empirical literature on fiscal multipliers (see Ramey (2019) for a survey), and in particular with those recent contributions that stress the heterogeneity of government outlays and public investment (for example, see Alloza et al. (2019) for the Euro Area case and Izquierdo et al. (2019) for emerging markets).

This paper is organized as follows. In Section 2 we discuss a set of empirical regularities about the evolution of public outlays during recessions and the initial fiscal space. In Section 3 we describe the model. In Section 4 we present the quantitative analysis of the model, as well as evaluating the main trade-offs and debt dependence of fiscal policy. In Section 5 we conclude.

Section snippets

Government Outlays Around Recessions

To analyze government expenditures, we use the information provided by the Government Finance Statistics Dataset (GFS) of the International Monetary Fund (IMF). Using the same source, Michaud and Rothert (2018) study the cyclical properties of most government outlay categories between 1990 and 2015, but they do not factor public investment into their analysis.3

The Model

We consider a dynamic, stochastic and general equilibrium model for a small open economy (SOE) and international investors. Time is discrete and indexed by t. The domestic economy is made up of three types of agents: a representative household, a continuum of heterogeneous firms and a domestic government.

Households have preferences over public and private consumption, and leisure. Households work and own firms in the economy. On the other hand, firms use labor supplied by households and public

Quantitative Analysis

This section presents the calibration exercise and the quantitative results (policy functions and simulations) of the model.

Conclusions

We develop a fiscal policy model of sovereign debt and default to contribute to the debate around the tradeoff between economic stabilization and debt sustainability. Government spreads are an endogenous object that captures the probability of repayment of government obligations. The link between fiscal policy and the private sector is public capital, in a setup in which these interactions are affected by sovereign and corporate risk. In the first part of the paper we provide evidence that the

CRediT authorship contribution statement

Antonio Cusato Novelli: Conceptualization, Methodology, Formal analysis, Supervision. Giancarlo Barcia: Conceptualization, Methodology, Formal analysis.

References (62)

  • E.M. Leeper et al.

    Government investment and fiscal stimulus

    Journal of Monetary Economics

    (2010)
  • P. Medas et al.

    Fiscal crises

    Journal of International Money and Finance

    (2018)
  • A. Michaud et al.

    Redistributive fiscal policies and business cycles in emerging economies

    Journal of International Economics

    (2018)
  • P.A. Neumeyer et al.

    Business cycles in emerging economies: The role of interest rates

    Journal of Monetary Economics

    (2005)
  • F. Roch et al.

    The dynamics of sovereign debt crises and bailouts

    Journal of International Economics

    (2018)
  • A.N. Sy

    Rating the rating agencies: Anticipating currency crises or debt crises?

    Journal of Banking & Finance

    (2004)
  • E. Talvi et al.

    Tax base variability and procyclical fiscal policy in developing countries

    Journal of Development economics

    (2005)
  • M. Aguiar et al.

    Sovereign Debt

    (2014)
  • M. Aguiar et al.

    Investment cycles and sovereign debt overhang

    Review of Economic Studies

    (2009)
  • M. Aguiar et al.

    Quantitative Models of Sovereign Debt Crises

    (2016)
  • A. Alesina et al.

    Austerity in 2009-2013

    Working Paper

    (2015)
  • D. Anzoategui

    Sovereign spreads and the effects of fiscal austerity

    (2019)
  • M. Ardanaz et al.

    Current expenditure upswings in good times and capital expenditure downswings in bad times?

    (2017)
  • C. Arellano

    Default risk and income fluctuations in emerging economies

    American Economic Review

    (2008)
  • C. Arellano et al.

    Fiscal austerity during debt crises

    Economic Theory

    (2017)
  • C. Arellano et al.

    Sovereign risk and firm heterogeneity

    Working Paper No.

    (2019)
  • C. Arellano et al.

    Partial Default

    Working Paper

    (2019)
  • T. Asonuma et al.

    Sovereign debt overhang, expenditure composition and debt restructurings

    Mimeo

    (2020)
  • P. Augustin et al.

    Sovereign to corporate risk spillovers

    Journal of Money, Credit and Banking

    (2018)
  • R. Baldwin et al.

    The Eurozone Crisis: A Consensus View of the Causes and a Few Possible Solutions

    Technical Report

    (2015)
  • J. Bianchi et al.

    Fiscal stimulus under sovereign risk

    Technical Report

    (2019)
  • Cited by (0)

    View full text