Professional Documents
Culture Documents
2020 - Institutional and - Economic - Determinants of - Regional Public - Debt in Spain
2020 - Institutional and - Economic - Determinants of - Regional Public - Debt in Spain
Determinants of
Regional Public
Debt in Spain
Abstract
We analyze from an empirical point of view the evolution and determinants
of Spanish regional public debt. Spain offers an interesting case study because
of its high level of fiscal decentralization, implemented gradually during the
past four decades, the parallel entry into force of a number of national fiscal
rules in that period, and the heterogeneity of its regions, both in terms of
economic fundamentals and some institutional features. Our main findings
are the following: (i) on average, over the sample of study, regional govern-
ments’ fiscal policies reacted to public debt increase; (ii) fiscal rules played a
limited role in controlling debt surges; (iii) a higher degree of regional fiscal
coresponsibility tends to be linked to more subdued debt dynamics;
(iv) market disciple has encouraged some fiscal restraint at the regional level;
and (v) increases in public commercial debt have affected the standard debt.
Keywords
regional public debt, fiscal rules, fiscal federalism, market discipline
1
Banco de España, Madrid, Spain
Corresponding Author:
Javier J. Pérez, Banco de Espana, C/. Alcalá 48, Madrid 28014, Spain.
Email: javierperez@bde.es
Delgado-Téllez and Pérez 213
80 25 20
60 20 15
40 15 10
20 10 5
0 5 0
95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17
Figure 1. The evolution of public debt in Spain, by subsector. Source: Banco de España.
215
216 Public Finance Review 48(2)
In % of GDP
5 1,2
4 1,0
3 0,8
2 0,6
1 0,4
0 0,2
95 97 99 01 03 05 07 09 11 13 15 17
GDP at t and let Dt be the nominal value of government debt. The govern-
ment budget constraint accounts for how nominal interest rate it, net inflation
pt , net growth in real GDP gdpt , net-of-interest deficit as a percent of GDP
deft , and deficit-debt adjustment DDAt combine to determine the evolution of
the government debt-to-GDP ratio,
Dt 1 þ it Dt1 DDAt
¼ þ deft þ ; ð1Þ
Yt ð1 þ pt Þðgdpt Þ Yt1 Yt
where the nominal yield it and the real stock of debt Dt are averages of
pertinent objects across terms to maturity. Its linearized version, suitable for
accounting decomposition of the fundamental determinants of debt, takes
the standard form:
Dt Dt1 Dt1 DDAt
¼ ðit pt gdpt Þ þ þ deft þ : ð2Þ
Yt Yt1 Yt1 Yt
-2
-4
1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016
Primary Balance Interest Payments Real GDP growth GDP deflator growth Deficit-debt adjustment Change in debt
Figure 3. The determinants of changes in regional governments’ debt (changes as a percent of gross domestic
product) in the period 1995 to 2017. Source: Banco de España; Ministry of Finance.
217
218
Andalusia Aragón Asturias Baleares
In % of GDP In % of GDP In % of GDP In % of GDP
6 6 4 6
4 4 4
2
2 2 2
0
0 0 0
-2 -2 -2 -2
95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17
6 8
4
2 6
4
2 4
2
0 2
0
0 0
-2 -2 -2 -2
95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17
Figure 4. Autonomous Communities’ (regional governments) debt evolution in the period 1995 to 2017: Changes as a
percent of gross domestic product. Source: Banco de España.
Catalonia Valencia Extremadura Galicia
In % of GDP In % of GDP In % of GDP In % of GDP
6 10 4 4
8
4
6 2 2
2 4
2 0 0
0
0
-2 -2 -2 -2
95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17
6
4
2 2
4
2
2
0 0
0
0
-2 -2 -2 -2
95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17 95 97 99 01 03 05 07 09 11 13 15 17
La Rioja
In % of GDP
4
-2
95 97 99 01 03 05 07 09 11 13 15 17
Figure 4. (continued)
219
220 Public Finance Review 48(2)
Institutional Framework
The Process of Fiscal Decentralization in Spain
After a gradual transfer of responsibilities from the central government to
the regions since the 1980s, Spain has now become one of the most decen-
tralized countries in the European Union. Indeed, 50 percent of General
Government expenditure was carried out by subnational governments in
2017, with about 36 percent and 14 percent in the hands of regional gov-
ernments and local governments, respectively (see figure 5, second panel).
In particular, AC were responsible for more than 90 percent of health-care
and education public expenditures in 2017. The transfer of those expendi-
ture responsibilities has been asymmetric both in pace and intensity. The
main differences concern the time at which the various AC took over
education and health competencies.
In parallel to this devolution process, a financing system for the subna-
tional governments was also progressively developed (see figure 5, first
panel, on the extent of revenue decentralization). Again, the process was
not completely homogeneous across regions. In particular, a distinction
should be drawn between the ordinary-regime AC (all except the Basque
Country and Navarre), with partial fiscal autonomy, and the specific-status
AC (the Basque Country and Navarre), which have full fiscal autonomy
with the exception of customs tariffs.
In essence, the Basque Country provincial authorities (Álava, Guipúzcoa
y Vizcaya) and Navarre’s regional government have the power to maintain,
establish, and regulate, inside their territory, the tax regime. However, there
are some coordinating provisions with the central government, which basi-
cally imply that the effective overall tax burden arising from their regula-
tory power must not be lower than the existing in the rest of the country.
70
60
50
40
30
20
10
0
AT
LV
FI
NL
MT
UK
IT
LT
DK
EL
CY
IE
HU
SI
LU
HR
EE
DE
PL
FR
SK
CZ
PT
BE
SE
RO
BG
ES
EU28
Regional Local
60
50
40
30
20
10
0
AT
LV
FI
MT
NL
UK
IT
LT
CY
DK
IE
EL
HU
SI
LU
HR
DE
PL
FR
EE
SK
BE
CZ
PT
SE
RO
BG
ES
EU28
Regional Local
previous section. The framework was reinforced recently on the wake of the
European sovereign debt crisis. In particular, in 2012, a more stringent and
comprehensive fiscal rules’ framework was approved by the Spanish Par-
liament (Budgetary Stability and Financial Sustainability law5). The main
elements of this legislation are (i) a debt limit for the General Government
as a whole (a debt-to-GDP ratio of 60 percent) and for each subsector (13
percent of GDP for AC), (ii) a (structural) balanced budget rule, (iii) a
public spending growth limit, and (iv) detailed rules to guarantee the timely
payment to government’s private sector providers.
Empirical Analysis
In this section, we analyze empirically the determinants of regional public
debt.
Fiscal rules. We use a measure of the strength of fiscal rules developed by the
EC, the so-called fiscal rules index, that is available for the whole period of
study.10 Public debt developments may be affected by the presence of
different types of fiscal rules insofar as they supposedly pose a permanent
constraint on fiscal policy. One of the main hypothesis that we intend to
contrast is whether fiscal rules have subdued the surge of regional debt, that
is, whether the rules have had the expected dissuasive effect. Thus, if fiscal
rules are effective, we would expect that an increase in strength of rules
should entail a moderation of debt. We also include the interaction of this
224 Public Finance Review 48(2)
index with three variables: lagged economic growth, the lag of the deviation
of budget balances with respect to the average for all AC, and the lag of the
deviation of the debt-to-GDP ratio to the average.11 The main idea behind
the selection of these variables is to test whether fiscal rules are more
effective depending on the economic and fiscal situation of the AC.
There are no previous insights about the effectiveness of fiscal rules
contingent on the strength of economic growth. Nonetheless, it would be
easier for the central government to be stricter in the application of rules to
those AC that present strong economic growth. Thus, we would expect a
negative impact for the interacted variable. A similar reasoning may be
applied to the other interacted variables. If the budget balance ratio is worse
for an AC in t 1, that is, it has a higher deficit or a smaller surplus, fiscal
rules would be effective if in the current year the debt level does not grow as
much. This would mean that the rules are effective for those AC with less
sound finances. In the case of the debt interaction variable, when an AC has
a higher debt, fiscal rules would be effective if the impact is negative,
meaning that the higher the debt, the less would it grow on the following
period. It may be noted that in the previous studies, such as Argimón and
Hernández de Cos (2012) and Delgado-Téllez, Lledó, and Pérez (2017),
fiscal rules do not seem to be very effective in deterring deficit noncom-
pliance. Therefore, we would not expect a high effectiveness of fiscal rules
in controlling debt evolution.
Political and institutional factors. The literature has proven the necessity of
including political and institutional factors to explain the persistence of
budget deficits and the accumulation of debt in advanced economies. In
our analysis, we include a number of political variables: (i) ideology, mea-
sured, first, by the percentage of left-wing MPs over the total seats of
regional parliaments and, second, by the weight of regionalist parties’ MPs
(regional parties that do not belong to a national party) over the total
number of seats of the regional parliament; the first variable aims at captur-
ing potential pro-spending biases depending on the ideological orientation
of the regional government, while the second could be instrumental for
testing the existence of different attitudes toward public debt accumulation
depending on the scope of the objective function of the regional govern-
ment; in both cases, there are no specific assumptions; (ii) a dummy for the
political concordance of national and AC governments that measures the
political alignment between both governments; (iii) electoral cycle: instead
of the standard election dummy that displays a value of 1 in an election year
and a 0 otherwise, we use a transformation of the original variable to
Delgado-Téllez and Pérez 225
Market discipline, structure of debt. Beyond the factors analyzed in the pre-
vious paragraphs, the ability to increase debt by a given level of adminis-
tration is fully determined by its ability to raise the necessary funds. In
addition to increasing taxes or decreasing expenditure, AC need external
financing (national or international). Thus, either investors buy the debt of
AC or banks grant loans to them. One possible conjecture is that market
pressure might be a key determinant on the change in public debt. The case
of Spain is not one in which there is full reliance on capital markets to
contain subnational borrowing as in the cases of Canada, Switzerland, and
the United States. The latter are cases in which the central government does
not set any limits on regional governments’ borrowing, so that these levels
of government are free to decide the form of borrowing, and may decide by
themselves to adopt a fiscal rule in an attempt to enhance their credit
standing in the market.17 Before 2012, Spanish regional governments were
constrained by upper-level rules while they were also subject to strict mar-
ket scrutiny. Nonetheless, in response to heightened fiscal stress levels due
to the European sovereign debt crisis, the central government constituted
the so-called Regional Government Liquidity Fund in 2012 which provided
funds directly to AC. Conditionality and financial control were reinforced
Delgado-Téllez and Pérez 227
Additional control variables: Pressure from units accounted for outside the
boundaries of the General Government sector. In particular, we consider the
two non-EDP debt measures described in Stylized Facts on Regional Public
Debt section: the debt of public corporations owned by a given region and
regional commercial debt. Indeed, the related literature would suggest that
(i) under tight budgetary rules, a government may try to circumvent its
constraints by cutting transfers to public corporations that, in turn, can
finance the same spending by issuing debt that is not computed by means
228 Public Finance Review 48(2)
Dit XN
D ¼ ai þ bj Ojit þ eit : ð3Þ
Yit j¼1
Under the proposed approach, the change in public debt of each regional
government, i, at time t, D DYitit , depends on a set of control variables, O,
encompassing the economic, political, institutional, market-induced, and
non-EDP factors mentioned above. Following the traditional fixed effects
model, ai in equation (3) aims at capturing all the unobservable AC effects
that are time-unvarying, while eit is an error term assumed to be white noise.
As for the estimation method, and in order to avoid any biases stemming
from the possible correlation between the individual effects and the regres-
sors, we estimate model (3) in first differences. Moreover, given the possible
simultaneity of some of the control variables and the dependent variable, the
estimation is carried out by the generalized method of moments (Arellano
and Bond 1991) using lagged regressors as instruments.
Although this type of models were first conceived for a sample with a
large number of subjects and a small number of periods, it has been grow-
ingly used in the literature for macroeconomic analysis purposes, where T is
relatively large and N is small. Nevertheless, when estimating this model for
a number of years (T) not small and number of subjects (N) not large (T ¼
23 and N ¼ 17), some problems may arise. First, we cannot use all lags
available as we would have a problem of proliferation of reduced form
coefficients (see Arellano 2016). Therefore, we have restricted the number
of lags to two.
Delgado-Téllez and Pérez 229
Results
The results are shown in tables 1 to 4. The two first tables include results
generated with two different samples, the whole sample (1995–2017) and a
sample excluding the recent crisis period (1995–2007), and two fixed
effects models’ estimation for robustness sake.
In table 1, we explore the role of the main economic variables and the
fiscal rules. In table 2, we focus on studying the institutional and political
factors. The effect of the different measures related with the market disci-
pline is shown in table 3 while in table 4, we consider the impact of the non-
EDP debt (public corporations and commercial debt) on the AC’s debt.
Beginning with the baseline model shown in table 1, there are six models
included in this table. The first two columns show the generalized method
of moments (GMM) for the whole sample, the second two the GMM with a
restricted sample from 1995 to 2007, and finally, the last two columns show
the fixed effects model for the whole sample. The dependent variable is the
change in debt-to-GDP ratio. The first conclusion we draw from the results
is that AC debt does not seem to follow an explosive path as both the lagged
dependent variable and the lagged level of debt show a negative effect on
the change of debt, which in the case of the level of debt is highly signif-
icant. This means that whenever the level of debt or the change in debt
increases in the previous year, the change in debt will be subdued in the
current one. This is probably due to a combination of factors, first, if AC
debt increases, the central government would control more thoroughly the
finances of the AC, and second, it would be harder for the AC to finance this
surge of debt at a reasonable cost, encouraging a stronger fiscal effort to
reduce the deficit. Interestingly, the stock of debt impact is stronger for the
shorter sample, probably indicating a looser fiscal discipline during the
crisis.
Table 1. The Determinants of Regional Governments’ Debt Changes (Changes as a Percentage of GDP): Baseline Models.
230
Dependent Variable: D EDP Debt GMM (1995–2017) GMM (1995–2007) FE (1995–2017)
Note: Instrument set in all models includes the second and third lag of the endogenous variable. EDP ¼ excessive debt procedure; GDP ¼ gross
domestic product; GMM ¼ generalized method of moments.
a
Hansen is the p value of the test of the overidentifying restrictions (see Hansen 1982), which is asymptotically distributed w2 under the null
hypothesis that these moment conditions are valid. A p value equal or higher than .05 indicates that the instrument set is valid, which is confirmed
under all models, m1 and m2 are the p values of serial correlation tests of order 1 and 2, respectively.
b
Hausman test p value under the null hypothesis indicates that the random effects model is preferred due to higher efficiency.
*significant at 10 percent level.
**significant at 5 percent level.
***significant at 1 percent level.
231
232
Table 2. The Determinants of Regional Governments’ Debt Changes (Changes as a Percentage of GDP): Political and Fiscal
Decentralization Variables.
233
Table 3. The Determinants of Regional Governments’ Debt Changes (Changes as a Percentage of Gross Domestic Product):
234
Market Discipline.
235
Table 4. The Determinants of Regional Governments’ Debt Changes (Changes as a Percentage of Gross Domestic Product): Non-
EDP Components (Public Corporations Owned by Regional Governments and Commercial Debt).
236
Dependent Variable: D EDP Debt GMM (1995–2017) FE (1995–2017)
Note: Instrument set in all models includes the second and third lag of the endogenous variable. EDP ¼ excessive debt procedure; GMM ¼ generalized
method of moments.
a
Hansen is the p value of the test of the overidentifying restrictions (see Hansen 1982), which is asymptotically distributed w2 under the null
hypothesis that these moment conditions are valid. A p value equal or higher than .05 indicates that the instrument set is valid, which is confirmed
under all models, m1 and m2 are the p values of serial correlation tests of order 1 and 2, respectively.
b
Hausman test p value under the null hypothesis that the random effects model is preferred due to higher efficiency.
*significant at 10 percent level.
**significant at 5 percent level.
***significant at 1 percent level.
Delgado-Téllez and Pérez 237
The second main result from the estimations is that, as expected, the
business cycle affects significantly the evolution of debt.22 In turn, the
variable GDP per capita deviation has a negative but weak impact, signaling
that those AC that have a higher GDP per capita tend to have less accumu-
lation of debt. Neither consumer price inflation deviations nor the house
price inflation deviations seem to have any significant impact on debt
evolution.
As regards the variable measuring budgetary compliance, it displays a
puzzling outcome because it is only statistically significant in the fixed
effects model. If the sign of the coefficient is negative, as it is in the fixed
effects and the long sample GMM, it means that fiscal noncompliance in the
past does not restrict regional governments in the present. Regarding the
variable measuring the “Fund for Suppliers’ Payments,” it has a significant
and direct effect on debt evolution as expected because the funds received
since 2012 were directly recorded as standard debt.
Still in table 1, the last four variables are the ones related with fiscal
rules. First, the fiscal rules strength index in the previous period does not
have a significant impact on debt variation one year after. This is a very
striking result that is nonetheless in line with previous studies such as
Argimón and Hernández de Cos (2012) and Delgado-Téllez, Lledó, and
Pérez (2017): current fiscal rules are not very effective in coping with fiscal
noncompliance (deficit control). In Spain, most of the central government
control until recently was concentrated on deficit targets compliance as
there were no specific debt targets until 2012. Notwithstanding, there are
some significant results when interacting the fiscal rules index with GDP
growth, the budget balance, and the debt-to-GDP ratio in deviations from
the AC average. As it appears, fiscal rules would tend to be more effective
when economic growth is lower, when debt levels are higher, and when
budgetary balances are worse. Thus, the two last results are as expected, but
it is slightly puzzling that fiscal rules are more effective when economic
growth is more dimmed. This could be rationalized if one considers that in
bad economic times public finances can be at stress, and tight rules might be
necessary to cope with potential sustainability problems.
In table 2, we focus on political and institutional factors. Political factors
seem to have a weak effect. Curiously, the only political variable that has
some effect is the weight of regional parties’ parliamentarians but neither in
the short sample GMM nor in the fixed effects model (FE) it has any
significance. More interesting is the result of the institutional factors’
impact. Revenue autonomy plays a key role in debt evolution as expected.
When AC have a larger share of their revenues coming from taxes and other
238 Public Finance Review 48(2)
it in the table, the results for this variable are alike for the short sample
period 1995 to 2007.
Finally, table 5 shows some elevated correlations among specific vari-
ables which may lead to multicollinearity problems. In order to ensure that
multicollinearity is not affecting our estimates, we replicate all models
considering separately those variables that show a strong correlation in an
attempt to confirm that the results are robust. In addition, we relax the strict
exogeneity condition considering the lag of the stock of debt as a prede-
termined variable, obtaining similar results with the exception of the sta-
tistical significance of nondisciplinary fiscal rules.
Conclusions
In this article, we study the determinants of regional public debt accumula-
tion process over the period 1995 to 2017. We find that it depends mainly on
stabilizing factors such as the economic growth and also on the previous
stock of debt. Other factors affecting public debt are market discipline,
proxied by the financing cost, and “the fiscal decentralization design,” that
is, the distribution of revenues and expenditure competencies. On the con-
trary, political variables and fiscal rules strength do not seem to have a
noteworthy effect on AC debt. Finally, commercial debt has been a source
of standard debt expansions, especially during the recent economic crisis.
These results highlight the need for some policy actions to ensure
regional debt sustainability. First, the lack of effectiveness of fiscal rules
might be read in relation to the economic literature emphasizing that a set of
features are crucial to achieve a certain incidence of any fiscal rule on the
behavior of governments. In this regard, issues such as transparency, the
credibility of penalties for fiscal noncompliance, and the existence of inde-
pendent institutions responsible for monitoring compliance appear as deter-
minants of the success of the fiscal rules. Second, as argued by Oates
(2008), fiscal discipline may be enhanced when subnational governments’
spending competences are funded to a larger extent with their own reven-
ues. Reviewing the current design of Spanish fiscal decentralization system
considering ways of improving efficiency and transparency in the govern-
ance could improve AC fiscal responsibility.
In addition, market discipline may be fundamental in enhancing regional
governments’ fiscal prudence. The recent bailout has deeply damaged this
source of discipline, rendering essential to recover investors’ trust by
improving their solvency image in order to resume financing their activity
with private funding. Finally, regulation has been adapted to prevent the
240
Table 5. Correlation Matrix.
Variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)
1. Debt change 1
2. Economic cycle .69*** 1
3. Inflation deviation .03 .02 1
4. GDP per capita deviation .07 .06 .28*** 1
5. House inflation deviation .05 .06 .26*** .09 1
6. Fiscal rules .43*** .41*** 1
7. Revenue autonomy .03 .02 .09 .46*** .01 .19*** 1
8. Vertical fiscal imbalances .14** .16** .08 .57*** .01 .03 .80*** 1
9. Health expenditure .14* .16** .26*** .80*** .09 .07 .47*** .65*** 1
10. Education expenditure .34*** .36*** .25*** .77*** .12* .39*** .31*** .57*** .80*** 1
11. Investment expenditure .09 .09 .44*** .07 .29*** .20*** .35*** .40*** .41*** 1
12. Electoral date distance .04 .08 .01 .01 .01 .02 .06 .04 0 .09 1
13. Concordance .16** .11* .09 .13** .01 .01 .11* .14** .06 .10* .14** .05 1
14. Percentage of left-wing MPs .14** .18*** .01 .22*** .03 .10* .09 .13** .11* .13** .16** .01 .05
15. Percentage of regionalist .06 .01 .17*** .56*** .05 0 .31*** .31*** .31*** .35*** .17*** .02 .38***
16. Short-/long-term debt .08 .02 .05 .02 .09 .24*** .16** .04 .05 .09 .07 0 .01
17. Securities/loans debt .19*** .17*** .06 .31*** .06 .16** .26*** .26*** .16** .34*** .12* .02 .16**
18. Nonresident/resident debt .02 .02 .14** .15** .08 .07 .09 .16** .04 .08 .12* .01 .1
19. Implicit interest .32*** .31*** .01 .04 .03 .73*** .09 .04 .07 .36*** .02 .01 .02
20. Issuance interest .13* .11* 0 0 0 .66*** .17*** .06 .06 .30*** .08 .04 .02
21. Rating .38*** .20*** .03 .16** .03 .74*** .01 .03 .06 .30*** .52*** .08 .05
22. Budget compliance .55*** .35*** .07 .05 0 .33*** .01 .11* .08 .20*** .08 .02 .20***
23. Debt/GDP .49*** .27*** .04 .08 0 .72*** .10* .01 .08 .32*** .41*** .03 .03
24. Fund for suppliers’ payments .67*** .35*** .02 .15** .02 .43*** .01 .04 .12* .25*** .23*** .08 .07
25. Public companies’ debt .50*** .64*** 0 0 0 .37*** .12* .11* .11* .33*** .21*** .09 .12*
26. Commercial debt .19*** .13* .07 .29*** .05 .05 .23*** .19*** .27*** .19*** 0 .06 .02
(continued)
Table 5. (continued)
Variable (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25)
1. Debt change
2. Economic cycle
3. Inflation deviation
4. GDP per capita deviation
5. House inflation deviation
6. Fiscal rules
7. Revenue autonomy
8. Vertical fiscal imbalances
9. Health expenditure
10. Education expenditure
11. Investment expenditure
12. Electoral date distance
13. Concordance
14. Percentage of left-wing MPs 1
15. Percentage of regionalist .30*** 1
16. Short-/long-term debt .04 .14** 1
17. Securities/loans debt .03 .41*** .13** 1
18. Nonresident/resident debt .09 0 .15** .28*** 1
19. Implicit interest .01 .06 .24*** .16** .17*** 1
20. Issuance interest .02 .01 .27*** .06 .1 .77*** 1
21. Rating .05 .14* .06 .44*** .06 .40*** .34*** 1
22. Budget compliance .04 .13** .20*** .11* .09 .27*** .14** .32*** 1
23. Debt/GDP 0 .01 .03 .21*** .1 .46*** .44*** .88*** .42*** 1
24. Fund for suppliers’ payments .16** .1 .06 .21*** .05 .21*** .05 .54*** .42*** .48*** 1
25. Public companies’ debt .12* .02 .05 .02 .13** .43*** .25*** .18** .31*** .06 .07 1
26. Commercial debt .04 .27*** .12* .25*** .12* .16** .09 .21*** .23*** .01 .06 .33***
241
242
Table 6. Summary Statistics.
243
244 Public Finance Review 48(2)
Authors’ Note
The motivation of this article draws significantly from Hernández de Cos and Pérez
(2013). The authors take responsibility for all the remaining mistakes.
Acknowledgments
We thank, in particular, Pablo Hernández de Cos for his comments. We also thank
the editor; the associate editor; the three anonymous reviewers; Guillem López-
Casasnovas; Alessandro Turrini; Marcello Sartarelli; Vicente Rı́os; and participants
at the ECOMOD conference 2017, the AEDE conference 2017, and the Encuentro
de Economa Pública 2018 for helpful comments and discussions and Rocı́o Prieto
and Ligia Topán for their research assistance.
Funding
The author(s) received no financial support for the research, authorship, and/or
publication of this article.
ORCID iD
Mar Delgado-Téllez https://orcid.org/0000-0002-8259-2737
Notes
1. For a broad perspective on the role of deficit-debt adjustment, see Campos,
Jaimovich, and Panizza (2006) for an international perspective on this issue.
2. The institutional determinants of local governments’ indebtedness have been
more widely analyzed in the literature, mainly from a less aggregated macro
perspective than the standard in papers looking at the determinants of Auton-
omous Communities’ (AC) debt. See, for instance, Cabasés, Pascual, and Vallés
(2007), Bastida, Beyaert, and Benito (2013), and the references quoted therein.
3. EDP stands for excessive debt procedure. It is the relevant concept of debt for
the purposes of the ceilings set in the European Stability and Growth Pact.
4. Commercial debt consists of financial claims which are created as a counterpart
of a financial or a nonfinancial transaction with the private sector in cases where
there is delay in the payment.
Delgado-Téllez and Pérez 245
5. Organic Law 2/2012, April 27, Organic law on budget stability and financial
sustainability.
6. Some authors point out, however, that the higher revenues in economic boom
periods may generally entail pressure on the growth of public spending in such a
way that the relationship between the economic cycle and the budget balance
may be altered or, at least, evidence asymmetrical behavior over the course of
the cycle. See Morris and Schuknecht (2007) on related grounds.
7. We measure the budgetary compliance as the difference between the actual
budget in cash terms and the initial budget. Thus, a positive value would entail
compliance as the budget balance would be better than expected in the initial
budget. A possible explanation for a better-than-first-expected budgetary beha-
vior would be the willingness of the AC government to comply with a medium-
term target.
8. This deviation is measured as the GDPpc divided by the average GDPpc multi-
plied by 100.
9. Apart from the impact on nominal gross domestic product (GDP; the denomi-
nator of the debt ratio), higher inflation may increase the budget deficit through
higher nominal interest rates and higher real cost of purchases of goods and
services or investment and, in general, of those items of public spending that can
be indexed (e.g., pensions and wages). In the presence of nonindexed taxes,
inflation may also generate higher revenues if, for instance, the tax rates are
progressive (see Hernández de Cos et al. 2016).
10. See https://ec.europa.eu/info/publications/fiscal-rules-database_en. The new
index is a normalized one, with a starting value of 0.45 to an ending value
of 2.27. We have rescaled the variable so that there are no negative values that
could hinder the estimation sign interpretation
11. Budgetary balances are obtained from the national accounts (NA) data pub-
lished by the Finance Ministry. Because these data are only available at an
annual basis from 2000 onward, the beginning of the sample is completed using
the public accounts’ budgetary data (cash based). This strategy will be used for
all NA variables. We use the lag of the budget balance over GDP to prevent
endogeneity problems.
12. On electoral cycles and budgetary outcomes, see, for example, Von Hagen
(2010) or Mink and de Haan (2005).
13. Thus, if there are four years between elections, the variable would take 0.25,
0.5, 0.75, and 1.
14. Concept first developed by Amilcare Puviani (Puviani and Volpi 1973) and
broadly analyzed in the literature (see, for instance, Oates 1979; West and
Winer 1980).
246 Public Finance Review 48(2)
15. The period 1995 to 1999 is obtained as a linear extrapolation with the execution
data as the NA only cover the period 2000 to 2017. All NA data come from the
General Intervention Board of the State Administration website.
16. Yearly disaggregated data are available separately in the Classification of the
Functions of Government publication available at the Finance Ministry, being
2016 the last year available.
17. In Switzerland, after an episode of soaring debt over the nineties, the govern-
ment introduced a debt brake rule, which is a highly strict rule, according to
which the deficit of one year must be compensated in the next years, as
explained by Bodmer (2006).
18. The Regional Government Liquidity Fund is a financial facility created by the
central government to provide funding to the AC. The central government
directly paid the AC maturities and recorded those payment as loans granted
to the AC. The entrance to this program implied a reinforced conditionality with
a stronger supervision on the AC budgetary execution.
19. Measured as the ratio of interest payments over the stock of debt.
20. Some papers have found short-term debt to be an indicator of vulnerability to
international financial crises: Borensztein et al. (2004), Rodrik and Velasco
(1999), and Bussière and Mulder (1999).
21. On a discussion about the role of public sector enterprises in Spain, see Fernán-
dez Llera and Garcı́a Valiñas (2013).
22. We use the lag of the GDP growth to avoid possible endogeneity problems.
23. In the generalized method of moments, the contemporaneous lag of implicit
interest rate is included in the estimation so that endogeneity problem may be
solved. On the contrary, we considered that the issuance interest rate does not
have the problem of endogeneity as it is an average variable of all AC, and the
data are common for all of them. Therefore, the increase in debt in one AC
should not have a very significant effect on the average issuance rate.
24. This variable is only available for the aggregate of AC.
References
Arellano, M. 2016. “Modelling Optimal Instrumental Variables for Dynamic Panel
Data Models.” Research in Economics 70:238–61.
Arellano, M., and S. Bond. 1991. “Some Tests of Specification for Panel Data:
Monte Carlo Evidence and an Application to Employment Equations.” The
Review of Economic Studies 58:277–97.
Argimón, I., and P. Hernández de Cos. 2012. “Fiscal Rules and Federalism as
Determinants of Budget Performance: An Empirical Investigation for the Span-
ish Case.” Public Finance Review 40:30–65.
Delgado-Téllez and Pérez 247
Bastida, F., A. Beyaert, and B. Benito. 2013. “Electoral Cycles and Local Govern-
ment Debt Management.” Local Government Studies 39:107–32.
Bodmer, F. 2006. “The Swiss Debt Brake: How It Works and What Can Go Wrong.”
Swiss Journal of Economics and Statistics (SJES) 142:307–30.
Borensztein, E., M. Chamon, O. Jeanne, P. Mauro, and J. Zettelmeyer. 2004.
“Sovereign Debt Structure for Debt Prevention.” IMF Occasional Paper, 237,
International Monetary Fund, Washington, DC.
Broner, F. A., D. Clancy, A. Erce, and A. Martin. 2019. “Fiscal Multipliers and
Foreign Holdings of Public Debt.” ECB Working paper, 2255, European Central
Bank, Frankfurt.
Brunnermeier, M. K., L. Garicano, P. R. Lane, M. Pagano, R. Reis, T. Santos, D.
Thesmar, S. Van Nieuwerburgh, and D. Vayanos. 2016. “The Sovereign-bank
Diabolic Loop and Esbies.” American Economic Review 106:508–12.
Bussière, M., and C. Mulder. 1999. “External Vulnerability in Emerging Market
Economies: How High Liquidity Can Offset Weak Fundamentals and the
Effects of Contagion.” IMF Working Paper, 88, International Monetary Fund,
Washington, DC.
Cabasés, F., P. Pascual, and J. Vallés. 2007. “The Effectiveness of Institutional
Borrowing Restrictions: Empirical Evidence from Spanish Municipalities.” Pub-
lic Choice 131:293–313.
Campos, C. F., D. Jaimovich, and U. Panizza. 2006. “The Unexplained Part of
Public Debt.” Emerging Markets Review 7:228–43.
Canuto, O., and L. Liu. 2010. “Subnational Debt Finance and the Global Financial
Crisis.” World Bank - Economic Premise, The World Bank (13):1–7.
Delgado-Téllez, M., V. D. Lledó, and J. J. Pérez. 2017. “On the Determinants of
Fiscal Noncompliance: An Empirical Analysis of Spanish Regions.” In Fiscal
Politics, Chap. 14, edited by V. Gaspar, S. Gupta, and C. Mulas-Granados,
345–76. Washington, DC: IMF.
European Commission. 2012. Workshop “Fiscal relations across government levels
in times of crisis -making compatible fiscal decentralisation and budgetary dis-
cipline.” 1–16. Belgium: European Commission, Directorate-General for Eco-
nomic and Financial Affairs (DG ECFIN).
Eyraud, L., and L. Lusinyan. 2013. “Vertical Fiscal Imbalances and Fiscal
Performance in Advanced Economies.” Journal of Monetary Economics
60:571–87.
Fernández Llera, R., and M. Garca Valiñas. 2013. “The Role of Regional Public
Enterprises in Spain: Room for a Shadow Government?” Hacienda Pública
Española/Review of Public Economics 205:9–31.
248 Public Finance Review 48(2)
Franzese, R. J., Jr. 2000. “Electoral and Partisan Manipulation of Public Debt in
Developed Democracies, 1956–90.” In Institutions, Politics and Fiscal Policy,
edited by R. R. Strauch and J. von Hagen, 61–83. Boston, MA: Springer.
Franzese, R. J., Jr. 2002. “Electoral and Partisan Cycles in Economic Policies and
Outcomes.” Annual Review of Political Science 5:369–421.
Governatori, M., and D. Yim. 2012. “Fiscal Decentralisation and Fiscal Outcomes.”
Directorate General Economic and Financial Affairs (DG ECFIN), European
Commission, Economic Notes.
Hansen, L. P. 1982. “Large Sample Properties of Generalized Method of Moments
Estimators.” Econometrica: Journal of the Econometric Society 1029–54.
Hernández de Cos, P., S. Hurtado, F. Martı́, and J. J. Pérez. 2016. “Public Finances
and Inflation: The Case of Spain.” Occasional paper, 1606, Banco de España,
Madrid, Spain.
Hernández de Cos, P., and J. J. Pérez. 2013. “Sub-national Public Debt in Spain
Political Economy Issues and the Role of Fiscal Rules and Decentralization.” In
Estado del bienestar: sostenibilidad y reformas. In XX Encuentro Economa
Pública, Universidad de Sevilla.
Köppl-Turyna, M., and H. Pitlik. 2018. “Do Equalization Payments Affect Subna-
tional Borrowing? Evidence from Regression Discontinuity.” European Journal
of Political Economy 53:84–108.
Kornai, J., E. Maskin, and G. Roland. 2003. “Understanding the Soft Budget Con-
straint.” Journal of Economic Literature 41:1095–136.
Mink, M., and J. de Haan. 2005. “Has the Stability and Growth Pact Impeded
Political Budget Cycles in the European Union?” CESifo Working Paper Series
1532, CESifo Group Munich, Munich, Germany.
Morris, R., and L. Schuknecht. 2007. “Structural Balances and Revenue Windfalls:
The Role of Asset Prices Revisited.” ECB Working Paper 737, ECB, Frankfurt,
Germany.
Musgrave, R. A. 1988. “Public Debt and Intergeneration Equity.” In The Economics
of Public Debt, 133–48. London: Palgrave Macmillan.
Mussons Olivella, J. M. 2017. “Fiscal Decentralization and the Cycle in Spain: An
Empirical Analysis of Fiscal Policy Responses.” Politica Economica 33:305–44.
Oates, W. E. 1979. “Lump-sum Intergovernmental Grants Have Price Effects.” In
Fiscal Federalism and Grants-in-aid, edited by Peter Mieszkowski and William
H. Oakland, 23–30. Washington, DC: The Urban Institute.
Oates, W. E. 2008. “On the Theory and Practice of Fiscal Decentralization.” Institu-
tional Foundations of Public Finance: Economic and Legal Perspectives 165:176.
Puviani, A., and F. Volpi. 1973. Teoria della illusione finanziaria. Milano, Italy:
Isedi.
Delgado-Téllez and Pérez 249
Qian, Y., and G. Roland. 1998. “Federalism and the Soft Budget Constraint.” Amer-
ican Economic Review 88:1143–62.
Rodrik, D., and A. Velasco. 1999. “Short-term Capital Flows.” Technical Report,
National Bureau of Economic Research Working Paper 7364, National Bureau
of Economic Research, Cambridge, MA.
Simon-Cosano, P., S. Lago-Peñas, and A. Vaquero. 2013. “On the Political Deter-
minants of Intergovernmental Grants in Decentralized Countries: The Case of
Spain.” Publius: The Journal of Federalism 44:135–56.
Sorribas-Navarro, P. 2012. “Determinants of Subcentral Governments’ Bailouts:
Evidence for the Spanish Regions. European Journal of Political Economy 27:
154–70.
Von Hagen, J. 2010. “Sticking to Fiscal Plans: The Role of Institutions.” Public
Choice 144:487–503.
West, E. G., and S. L. Winer. 1980. “Optimal Fiscal Illusion and the Size of
Government.” Public Choice 35:607–22.
Author Biographies
Mar Delgado-Téllez is a senior economist in the Research Department of the Bank
of Spain (Eurosystem). She received her PhD from Complutense University of
Madrid, Spain. Her research interests include fiscal policy, regionalism, and
macroeconomics.
Javier J. Perez is the director of International Economics in the Research Depart-
ment of the Bank of Spain (Eurosystem). He received his PhD from Complutense
University of Madrid, Spain. His research interests include macroeconomics, fiscal
policies, European integration, and textual analysis.