Why the financial aid of the EU's regional policy does not impact on the public investment level? The case of the Spanish regions

• Published date: 01 February 2022
• Author: José María Cantos‐Cantos,Daniel Balsalobre‐Lorente,Cristina García‐Nicolás
• Date: 01 February 2022
• DOI: http://doi.org/10.1002/pa.2295

ACADEMIC PAPER

Why the financial aid of the EU's regional policy does not

impact on the public investment level? The case of the

Spanish regions

José María Cantos-Cantos | Daniel Balsalobre-Lorente | Cristina García-Nicolás

Department of Political Economy and Public

Finance, Economic and Business Statistics, and

Economic Policy, University of Castilla-La

Mancha, Spain

Correspondence

Daniel Balsalobre-Lorente, Department of

Political Economy and Public Finance,

Economic and Business Statistics, and

Economic Policy, University of Castilla-La

Mancha, Ciudad Real, Spain.

Email: daniel.balsalobre@uclm.es

Given that an expansive impact of the European Regional Development Fund (ERDF)

aid on public investment is expected, two models have been estimated to determine,

respectively, the size of this impact and the relationship between the own financing

of regional public investment and the external financing received from ERDF in Span-

ish regions. Evidence has been obtained that the impact of transfers from the ERDF

on public investment is almost nil. In addition, for each euro that the Spanish regions

received from the ERDF, own funds for public investment allocated by regional gov-

ernments have been reduced 23 cents, showing an important relation of substitution

between the two sources of financing. Both models include data between 1994 and

2014 and their estimation has used traditional methods and methods based on the

cointegration of variables with a panel structure, such as FMOLS and DOLS. The

main results obtained with the two methodologies they are quite close, confirming

the initial hypothesis: in the presence of weak donor restrictions, the political process

in governments receiving ERDF transfers determine revealed preferences about pub-

lic investment, neutralizing conditions established by the donor. This study also

unequivocally demonstrates the presence of simultaneous causality between invest-

ment aid received from the ERDF and regional public investment and between said

aid and the own resources that regional governments dedicate to financing public

investment. These findings confirm the existence of a bias that seriously affects the

effectiveness of the European Union's regional policy.

JEL CLASSIFICATION

H70; H77; R58

1|INTRODUCTION,

INTERGOVERNMENTAL TRANSFERS, AND

PUBLIC SPENDING

The intervention of tax system in pr ivate economy is never neutral,

as neither is any fiscal or financial interventi on from top-level admin-

istration towards a lower level. While decision of an individual to

acquire an additional quantity of goods on t he private market also

involves providing the resources needed to prod uce them, in collec-

tive decision-making this simultanei ty is not given, allowing display a

distorted tax price of public service s that facilitates attracting

resources to the public sector and to the manager s themselves

(Wicksell, 1958).

Although a multilevel finance should produce a balanced distribu-

tion of financial capacity and spending among various levels of admin-

istration and government, in practice this balance does not usually

occur (Oates, 1999:1124). In fact, most of revenues come from the

highest levels of government, which causes incentives to alter level

and composition of expenditure at the lower government levels,

among other problems. As literature on fiscal federalism in many cases

shows (Bird, 2015; Bird & Staufer, 2001; McLure, 1993; Oates, 1972),

a vertical fiscal imbalance usually occurs during expansion process of

Received: 13 May 2020 Revised: 29 May 2020 Accepted: 12 July 2020

DOI: 10.1002/pa.2295

J Public Affairs. 2022;22:e2295. wileyonlinelibrary.com/journal/pa © 2020 John Wiley & Sons Ltd 1of24

https://doi.org/10.1002/pa.2295

public sector, generating greater financial capacity at the highest

levels of governments to detriment of those closest to citizens. Cor-

recting this imbalance can lead to a change of behavior from public

and private economic agents.

Bradford and Oates (1971a) wondered whether it would be the

same to give unconditional grants to governments and individuals—

financial equivalence's hypothesis.

1

Rejecting an approach until then

considered inappropriate, the authors adopted the premise that an

analytical framework based mainly on individuals cannot be applied to

another scenario where relations between groups of people are ana-

lyzed, without incorporating assumptions related to the underlying

political process in the community. As a feature of intergovernmental

grants is the ability of donor to establish conditions, Bradford and

Oates (1971b) also treat matching grants within analysis framework

that incorporates the political process.

In conclusion, under certain assumptions, authors find financial

equivalence rule applicable to lump sum grants it is also valid for

matching grants. Moreover, it is not true that any individual will

always prefer increased public spending under a matching grant than

under a lump sum grant, although the first has a favorable price effect

that adds to the income effect—the resulting median budget being

higher. Specifically, the authors only compare the expansionary

effects on public expenditure of two types of intergovernmental

transfers, exploring the conditions under which matching grants are

not more expansive than lump sum grants. However, nothing indi-

cates whether, under some assumptions, expect matching grant has

contractive effects on recipient government public spending. Since

then, some empirical researches have tried to check whether

matching grants result in an expansion of public spending at recipient

government level, and in what proportion. After all, if main motive for

giving a matching grant is to increase public spending on a good or

service in target jurisdiction, it is interesting to see if this happens and

to what extent.

Contrary to the hypothesis of financial equivalence, the first

empirical applications obtained evidence that both unconditional and

conditional grants produced an expansion of public spending.

Thereby, Gramlich and Galper (1973) analyze the impact on state and

local public spending from different types of intergovernmental trans-

fers in the United States, and they find that specific revenue sharing

and lump-sum grants have a similar impact on the public spending of

the recipient government level estimated between $0.25 and $0.43

of expenditure for each dollar of grant. Meanwhile, this amount is

between $0.65 and $0.90, for the case of categorical grants, and

$0.80 for matching grants. On his part, Winer (1983) applies a model

based on the decisive voter to see whether central government trans-

fers to the Canadian provinces causes an increase on public spending

of the latter, obtaining evidence that by being reduced tax price to the

receiver, it produces an expansive effect on public spending that

decreases over time but does not completely disappear.

Other research, such as Case, Rosen, and Hines (1993), whose ini-

tial aim was to estimate the flypaper effect of federal transfers on

States public spending, have been widely questioned due to the

model specification and the data used (see Becker, 1996). Indeed,

since the 1980s, many papers have turned to the corroboration of the

hypothesis known as "flypaper effect," an empirical regularity

detected by Arthur Okun and formulated in response to the work of

Courant, Gramlich, and Rubinfeld (1979), based in lump sum grants.

2

Many empirical methods, variables and data sources have been used

in several countries to find out what truth there is in the flypaper

hypothesis. Gramlich (1987) suggests that crowd-in effect sometimes

observed is due to political factors, since government programs take

root and generate some patronage, so they are politically difficult to

avoid, even in the case of a reduction in the intergovernmental grants

received.

3

However, since the study of Besley and Case (2000), several

empirical studies demonstrate that intergovernmental transfers do

not always contribute to the expansion of public spending.

4

Con-

versely, they can propel counterclockwise to move the local expendi-

ture financing either to other budget items, or in the form of a

reduction of the local government budget, producing a crowd-out.

While the latter is not part of a strict set of conclusions at Bradford

and Oates (1971b) for matching grants, it is one of implications of his

theoretical work, as after we discuss. The contribution of Besley and

Case (2000) affects one of the problems that may have influenced the

validity of some empirical models, as is the failure to consider the pos-

sible endogeneity of grants regarding political decision-making, moti-

vated by omission of negotiating framework of the grants themselves,

and other political process aspects in the recipient government head-

quarters. It would be a matter of seeing if the political process of

determining grants could be considered as an unobservable variable

correlated with the grants themselves, being determined through

some legislative negotiations or optimizing political reelection models.

In these cases, the models that include matching grants as pre-

determined variable may be biased against the local government

spending crowd-out.

5

In Knight (2002), an instrumental variables' methodology (IV) is

used to study the U.S. federal program to support State programs

roads, where it is expected that the expenditure of each State for this

purpose will increase, at least in the amount received from federal

grant. Once chosen instrumental variables and applying the method of

estimation in two stages, Knight demonstrates that federal transfers

for this purpose eject local spending in 88% of the existing level in

each state, that is, for every dollar received for matching grant, the

State public investment in highways is reduced by 88 cents.

Against most of the empirical results obtained until then which

stated the presence of a flypaper effect on educational grants, Gordon

(2004) gets a result that in the short term seems to support the idea

that the federal grant also increases local and state government

spending on education. However, when the longer term behavior is

analyzed (using a delay of 3 years), a local spending crowd-out effect

is observed, whose fall would probably be accompanied by a tax

reduction. That is, education spending accommodates the rhythm of

the federal grant received. Nevertheless, Pallesen (2006) analyzes the

impact on Danish municipalities derived from change matching grants

subsidies by lump-sum grants, finding that there was not significant

impact on local public spending.

2of24 CANTOS-CANTOS ET AL.