Subnational borrowing and bailouts: When the federal government looks at the votes (differently) and its borrowing matters

• Published date: 01 June 2022
• Author: Diego Martínez‐López
• Date: 01 June 2022
• DOI: http://doi.org/10.1111/jpet.12559

Received: 27 March 2021

|

Accepted: 17 November 2021

DOI: 10.1111/jpet.12559

ORIGINAL ARTICLE

Subnational borrowing and bailouts: When the

federal government looks at the votes

(differently) and its borrowing matters

Diego Martínez‐López

Department of Economics, Universidad

Pablo de Olavide and GEN, Seville, Spain

Correspondence

Diego Martínez‐López, Department of

Economics, Universidad Pablo de

Olavide, Ctra. Utrera Km. 1, 41013

Seville, Spain.

Email: dmarlop1@upo.es

Abstract

Sometimes it is difficult to find a rationale for episodes of

bailouts in which the political motivations of the federal

government are not clear cut. In other cases, the soft‐

budget constraint of subnational governments (SNGs)

seems to be based on unlimited federal resources. This

paper aims to shed some light on both issues taking as a

reference the workhorse model developed by Goodspeed.

The principal change in its basic assumptions lies in al-

lowing the federal government to borrow to finance its

grants to SNGs. The results indicate that the way in

which the federal government translates voters' pre-

ferences into electoral probabilities is crucial to determine

grants. Moreover, the SNGs do not borrow excessively.

When the model is extended to consider risk premium

spreads in the SNGs debt and the option of transferring

part of their borrowing to the upper level, the main

outcomes remain.

1|INTRODUCTION

Episodes of bailouts can adopt multiple forms and motivations. So many that sometimes it is

difficult to find a rationale. Let us take as an example the recent measures implemented by the

federal government in some countries to support subnational governments (SNGs) during the

J Public Econ Theory. 2022;24:609–633. wileyonlinelibrary.com/journal/jpet

|

609

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial License, which permits use,

distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

© 2021 The Authors. Journal of Public Economic Theory published by Wiley Periodicals LLC

Covid‐19 pandemic (OECD, 2020). Australia, Canada, Brazil, and the United States have

provided a variety of financial assistance to SNGs through the central bank. Norway, Italy,

Poland, and Spain have offered extra funds to their regions and municipalities, with the latter

two postponing the fall of regular equalization funds as well.

Strictly speaking, these discretionary measures were certainly not motivated by the pressure

of highly indebted SNGs or closely related to necessary stabilization functions (in the hands of

federal governments). Consequently, they cannot be easily classified in principle as bailouts.

But, as we will show later, this type of policy might imply a kind of preventive bailouts, in an

effort to anticipate future overborrowing problems.

Anyway, a quick explanation of these discretionary grants would be: The federal concern re-

garding the financial impact of Covid‐19 on SNGs leads to the instrumentation of additional grants

aimed at correcting major negative vertical imbalances. This is indeed the traditional approach from

the first‐generation fiscal federalism models, based on the assumption of benevolent social planners.

However, from Oates (2005) onwards, strategic interactions in federal countries between political

officials, voters, and fiscal institutions go beyond the maximization of social welfare (Weingast, 2009).

In this renewed context, the literature on bailouts provides interesting insights.

The usual approach to bailouts in hierarchical structures of government has indeed been

analyzed from different perspectives (Goodspeed, 2017; Kornai, 2014): in the presence of tax

competition, focusing on public investment, stressing the effect of equalization with different

types of grants, and so forth. But all of them share a common starting point: The existence of

soft‐budget constraints within SNGs. Local/regional politicians have reasons to build bailout

expectations once the upper government is not able to credibly commit to a hard budget

constraint policy with no scope for discretionary grants. In fact, the implementation of bailouts

usually takes the form of ex post grants in favor of lower tiers of government but, sometimes,

directly involves taking subnational debt over or lending to them under favorable conditions.

The workhorse model developed by Goodspeed (2002) shows the main factors entailed in

bailouts. The framework consists of a sequential game in which SNGs move first, knowing the

subsequent reaction of the federal government as second mover. As long as the SNGs are

dealing with a politically motivated federal government, they take advantage of that and lead

their finances to overborrowing. Having arrived at this point, the unsustainability of public

finances, even linked to some difficulties providing socially important expenditures, results in

additional and discretionary (but expected) resources from the federal government to the SNGs.

However, further findings obtained by the empirical literature or features taken from the world of

policy‐making, such as those mentioned above on the Covid‐19 bailouts, cast some doubts on the

appropriateness of this canonical model to capture new insight. In this paper, we modify slightly the

basic model provided by Goodspeed (2002) to see the extent to which the results from the original

approach change. The idea is to keep the model as close as possible to this seminal approach because

we are interested in highlighting how new assumptions lead to different outcomes and, at the same

time, in retaining the ability of Goodspeed (2002) to arrive at simple and stylized findings.

The two novelties introduced here are related to the relevance of politics in deciding on grants

and the federal intertemporal budget constraint when supporting other tiers of government. In the

first case, Baskaran (2013) provides a good illustration of what we aim to highlight about the

political economy of bailouts. He questions the direct effect of electoral considerations to guide

federal decisions on discretionary grants to SNGs. Using a panel of municipalities in the German

State of Hesse, he finds that bailouts did not affect political support for the governing parties at the

state level. So, what about the political motivations of the incumbent government when deciding on

discretionary grants, as the theoretical models predict?

610

|

MARTÍNEZ‐LÓPEZ