Policy Research Working Paper 10942
Procyclical Fiscal Policy in Argentina
Drivers and Channels of a Long-Lasting Marriage
Julián Folgar
Francisco García Posleman
Macroeconomics, Trade and Investment A verified reproducibility package for this paper is
Global Practice available at http://reproducibility.worldbank.org,
October 2024 click here for direct access.
Policy Research Working Paper 10942
Abstract
Fiscal policy can play a critical role in mitigating business wages, setting Argentina apart from its peers. Furthermore,
cycle volatility. However, between 1990 and 2022, Argen- through an in-depth analysis of Argentina´s procyclical pen-
tina was one of the most procyclical countries globally. sions and public wage policies, the paper finds that both
This paper scrutinizes Argentina’s fiscal policy conduct have been influenced by the “price” effect, while the “quan-
throughout the business cycle, pinpointing procyclical tity” effect has been particularly relevant for the public wage
budget components and estimating the cyclically adjusted bill in recent decades. Additionally, Argentina’s unconven-
primary balance, incorporating transient and structural fac- tional taxes exacerbate procyclicality. These contributions
tors. The study reveals a significant decline in Argentina’s hold significant implications for policy makers, offering
cyclically adjusted primary balance. It shows that Argenti- valuable guidance in designing more effective policy levers
na’s fiscal procyclicality stems primarily from expenditure to achieve fiscal policy objectives, including macroeco-
policies, particularly those related to pensions and public nomic stabilization.
This paper is a product of the Macroeconomics, Trade and Investment Global Practice. It is part of a larger effort by the
World Bank to provide open access to its research and make a contribution to development policy discussions around the
world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may
be contacted at jfolgar@worldbank.org. A verified reproducibility package for this paper is available at http://reproducibility.
worldbank.org, click here for direct access.
RESEA
CY
LI R
CH
PO
TRANSPARENT
ANALYSIS
S
W
R
R E
O
KI P
NG PA
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
Produced by the Research Support Team
Procyclical Fiscal Policy in Argentina:
Drivers and Channels of a Long-Lasting Marriage1
Julián Folgar2
Francisco García Posleman
Key words: fiscal policy, procyclicality, Argentina, CAPB, pensions, wage bill, taxes.
JEL classification: E32, H29, H55, H59, H62, H63,
1
The authors are grateful to Sonia Araujo, Daniela Dborkin, and Jose-Daniel Reyes for their valuable comments.
2
Email: jfolgar@worldbank.org
1. Introduction
This paper investigates the influence of fiscal policy on macroeconomic stability and welfare in Argentina,
with a specific focus on the widely accepted economic consensus that underscores the role of fiscal policy
in stabilization. Both Keynesian and neoclassical economic paradigms concur on the potential detrimental
consequences of procyclical fiscal policies—those that expand during economic upswings and/or contract
during downturns—especially for low-income households. Such policies tend to exacerbate
macroeconomic volatility, accentuate the business cycle, suppress investment, curtail potential economic
growth, and ultimately affect the well-being of the poor (Braun, 2003; Fatas and Mihov, 2003, 2012;
Aghion and Marinescu, 2007; Woo, 2009; Izquierdo, et al. 2018;). Consequently, an in-depth analysis of
the cyclical attributes of fiscal policies emerges as a pivotal element in shaping effective policy responses
to fluctuations in the macroeconomic landscape.
It is well documented that developing countries tend to display higher volatility and lean towards more
procyclical fiscal policies when compared to advanced economies. Drawing upon prior research (Frankel,
Vegh, and Vuletin, 2011; Vegh et al., 2021), Figure 1 illustrates the correlation between the cyclical
component of public spending and real GDP, as well as the standard deviation of GDP growth across 74
countries during the period from 2000 to 2022. The findings are compelling: Most advanced economies
(represented by orange dots) tend to pursue acyclical or countercyclical fiscal policies and greater
economic stability. In contrast, developing economies (represented by blue dots) generally tend to exhibit
the opposite trend, albeit with varying degrees of heterogeneity.3
Figure 1. GDP Volatility and Fiscal Cyclicality 2000-2022
1.0
Procyclical
0.8
0.6
0.4
Fiscal Cyclicality
0.2
-
(0.2)
(0.4)
(0.6)
Countercyclical
(0.8)
(1.0)
0.010 0.020 0.030 0.040 0.050 0.060 0.070
Less volatile GDP Volatility More volatile
Source: OECD Stats, IMF, World Bank.
Note: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott filter. Public spending is
considered in real terms, deflated by CPI. Fiscal cyclicality is measured as the correlation between the cyclical component of real spending
and real GDP. GDP volatility is measured as the standard deviation of the annual growth rate for each country.
3
The focus, here, remains on establishing correlations rather than delving into causal relationships between fiscal procyclicality and GDP
volatility. For a more comprehensive exploration of causality links, see Ilzetzki & Vegh (2008).
Argentina has exceptionally high fiscal pro-cyclicality and GDP volatility. Over the past seven decades, the
nation has grappled with economic recessions for approximately 33 percent of the time, making it one of
the world's most volatile economies, surpassing even nations beset by conflicts in terms of political and
civil violence.4 Argentina's GDP volatility, as gauged by the standard deviation of economic growth rates,
ranks consistently high among the 74 countries in our sample spanning the last 22 years. 5 This
susceptibility to economic shocks has, over time, been associated with a decline in output growth. The
average annual growth rate over the past five decades stands at a disappointing 2.1 percent, notably
below the regional average of 3.3 percent.6
Argentina's fiscal policy has historically been highly procyclical, characterized by persistent fiscal deficits
that exacerbate economic fluctuations and hinder stable growth. The country ranks among the top four
nations, alongside Nicaragua, Latvia and Uruguay, in terms of the correlation between public spending
and GDP's cyclical component over the past 22 years. Moreover, as shown in Figure 2, this is a long-lasting
story. The cyclical component of public spending has been strongly linked to the business cycle, failing to
serve as a buffer against economic volatility. This fiscal approach has contributed to significant
macroeconomic imbalances, resulting in frequent credit events and magnified business cycles. From 2005
to 2016, public spending in Argentina rose significantly, reaching 41 percent of GDP (up from 25 percent
of GDP, similar to the historical average) due to favorable global conditions driven by the commodities
super-cycle. This spending increase, primarily in recurrent expenditures, outpaced revenue, leading to
additional fiscal imbalances, liquidity constraints, and debt restructuring.
Figure 2. Cyclical Components of GDP and Public Spending 1961-2022
30 60
20 40
cyclical component of Public Spending
10 20
Cyclical component of Real GDP
0 0
(10) (20)
(20) (40)
(30) (60)
(40) (80)
(50) (100)
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2019
2021
Source: OECD Stats, IMF, World Bank.
Note: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott filter. Public spending is
considered in real terms, deflated by CPI.
4
From a sample of 128 countries, Argentina stands out as being the country that has spent more time in recession in the past 70 years (33% of the
time), followed by the Democratic Republic of Congo (31% of the time).
5
Clearly, correlations do not provide us with direct information about causal effect that, in principle, could go both ways.
6
Based on the Conference Board Database. Regional average includes Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, Mexico, Paraguay,
Peru and Uruguay.
This paper investigates the cyclical aspects of Argentina's fiscal policy to identify the sources of fiscal
procyclicality. The analysis comprises two exercises. The first dissects the public budget to pinpoint
sources of procyclicality and assess how fiscal policy behaves over the business cycle. This aims to reveal
whether Argentina's fiscal instruments exhibit procyclical patterns and if this is driven by spending or
revenue policies. The second exercise focuses on estimating the cyclically adjusted primary balance (CAPB)
by accounting for transitory or structural factors. This estimation is particularly relevant in Argentina,
given the agriculture sector's significance and its potential impact on the fiscal stance. Recurrent shocks,
like droughts or commodity price fluctuations, affect economic activity, revenue collection, and
potentially public spending, leading to changes in the fiscal balance. Therefore, the CAPB estimate
specifically differentiates cyclical shocks related to commodities.
The study introduces innovative contributions to the analysis of procyclical fiscal policies in Argentina. It
employs a novel approach to decompose the sources of procyclicality in specific spending items, offering
potential avenues for further exploration. By scrutinizing the historical series of both the "price" and
"quantity" components of pension and public employment compensation spending, it disentangles the
procyclical effects, identifies their priority effects, and replicates this analysis for various pension
subsystems and government levels (where data is available).
The paper´s key findings show that Argentina’s high fiscal procyclicality is concentrated on the spending
side. Argentina ranks among the countries running the most procyclical public spending policies in recent
decades. Both automatic and non-automatic spending budget items appear to positively co-move with
the business cycle in a deterministic manner, with traditional automatic stabilizers having a minor role
and a weak overall effect. However, the paper highlights the need to explore the determinants of
cyclicality in the two most significant items of the budget: pensions and public wages. Although
conventional literature suggests that there are no objective reasons for these spending items to behave
countercyclically, there is a vast consensus that these items should be executed in a neutral way
concerning the business cycle.
Results indicate that Argentina has exhibited significant procyclical pension spending in recent decades,
contributing to overall fiscal procyclicality. The evidence reveals that public spending in major pension
schemes has been strongly influenced by "price" effects, which refer to the evolution of the real value of
benefits over time through indexation. Additionally, there is a partial influence of "quantity" effects.
Interestingly, while indexation rules typically prevent social security spending from being procyclical in
most countries, the Argentine case shows the opposite. The design and erratic implementation of
indexation rules have fueled procyclicality.
Furthermore, the analysis reveals that Argentina's public wage bill has displayed strong procyclicality,
driven by both "price" and "quantity" effects. In addition to having a larger public wage bill compared to
regional and global peers, Argentina's wage bill has exhibited significant procyclicality in recent decades,
even when compared to global and regional standards. The number of public servants, representing the
"quantity" effect, as well as their average wage, representing the "price" effect, have both shown a
pronounced procyclical behavior at both the national and subnational levels of government.
On the revenue side, the analysis suggests fairly acyclical tax policies for primary traditional taxes.
Nevertheless, unconventional taxes, with limited data for analysis, diminish the relevance of traditional
taxes, as a significant portion of tax revenue in Argentina comes from unconventional taxes. Evaluating
these unconventional taxes poses challenges due to the limited availability of detailed information.
Finally, the study highlights a significant deterioration in Argentina's structural fiscal balance (CAPB) over
the past decades, even prior to the conclusion of the commodities super-cycle. This deterioration
indicates the presence of procyclical discretionary policies. The paper also demonstrates that
discretionary policies tend to be more pronounced during economic booms. Additionally, given
Argentina's heavy reliance on agricultural commodities, the paper conducts a separate assessment of the
impact of commodity price shocks. The findings reveal the potential benefits of establishing a commodity-
related stabilization fund, which could have generated revenue equivalent to nearly 2 percent of GDP
between 2006 and 2008, a period when the country still had fiscal surpluses.
The paper is organized as follows. In Section 2, we define fiscal cyclicality and discuss its significance in
terms of its impact on economic welfare, drawing upon the main findings of the existing literature. Section
3 presents detailed estimates of the procyclicality of public spending and key tax revenues, providing a
breakdown of their sources of cyclicality to the best extent possible. Where applicable, international
comparisons are also included. Section 4 offers an overview of the fluctuations of the fiscal balance in
Argentina throughout the business cycle, utilizing innovative estimates of the CAPB for the country that
account for the influence of commodity prices. Finally, in Section 5, we summarize the main conclusions
of the paper.
2. Fiscal cyclicality: Theoretical framework and literature review
The adoption of macroeconomic policies aimed at minimizing business cycle volatility is a widely
recognized consensus in economics. Substantial research has demonstrated a negative correlation
between macroeconomic instability and medium- to long-term economic growth, household
consumption, and poverty reduction (Ramey and Ramey, 1995; Fatás, 2002; Acemoglu et al., 2003;
Hnatkovska and Loayza, 2005).
However, the adverse impacts of macroeconomic instability are notably more pronounced in developing
economies, attributable to several factors (Loayza, 2007). Firstly, these economies exhibit higher
vulnerability to external shocks, such as abrupt shifts in capital flows or significant fluctuations in
commodity prices, primarily because they are often reliant on commodity exports. 7 Secondly, emerging
economies confront a greater incidence of negative shocks stemming from domestic issues, including
political instability, highly procyclical policies, and less sophisticated economic institutions. Thirdly, these
economies typically possess smaller and weaker financial systems, constraining the capacity of both public
and private sector actors to mitigate the effects of business cycle fluctuations through credit channels.
Furthermore, less sophisticated institutional and regulatory policy frameworks, commonly found in
emerging economies, introduce inflexibilities into the economic landscape, impeding efficient resource
allocation in response to shocks (Caballero, 2004).
Therefore, the role of fiscal (and monetary) policies in stabilization has been a central subject of empirical
macroeconomic research in recent decades. Notably, the exploration and analysis of fiscal policy as a tool
for macroeconomic stabilization have gained greater prominence in recent years, primarily due to the
challenges faced by monetary policy, especially in advanced economies (Spilimbergo, et al. 2008).
The economic consensus advocates for governments to implement fiscal policies that do not exacerbate
(either positively or negatively) economic shocks. Fiscal policy is characterized as procyclical
(countercyclical) when it is expansionary (contractionary) during economic upswings and contractionary
(expansionary) during downturns. Alternatively, fiscal policies may exhibit a more neutral cyclical effect if
they display no significant biases across the business cycle. Procyclical fiscal policies are widely recognized
as detrimental to welfare, as they tend to amplify the business cycle, suppress investment, diminish
potential growth, and ultimately impact low-income segments (Woo, 2009; Aghion and Marinescu, 2007;
Fatas and Mihov, 2003, 2012; Izquierdo, et al 2018; Braun, 2003).
From a Keynesian perspective, the aggregate fiscal policy should follow a countercyclical approach to
smooth the business cycle as effectively as possible. On the other hand, the neoclassical tax-smoothing
theory (Barro, 1979) recommends implementing neutral (acyclical) fiscal policies by maintaining
consistent levels of spending and taxation over time. Regardless of the theoretical framework, the
overarching consensus underscores the importance of avoiding fiscal policies that positively co-move with
the business cycle. Furthermore, if excessive spending or tax cuts during economic booms are not fully
reversed during downturns, fiscal procyclicality can lead to a deficit bias, introducing additional
macroeconomic risks related to debt sustainability.
7
A “sudden stop” is an abrupt decline of capital flows into a nation's economy, which is often accompanied by
economic recessions and market corrections. Sudden stops may also be followed by a currency crisis, as foreigners
lose faith in a nation's economy (Calvo 1998).
Developing economies are particularly susceptible to procyclical policies due to their weaker institutions,
limited access to international credit markets, and vulnerability to external and domestic shocks (Frankel,
Vegh, Vuletin, 2011). The findings presented in the introduction are in alignment with existing literature.
While most advanced economies tend to implement countercyclical or acyclical fiscal policies, the
majority of developing countries typically adopt procyclical policies (Gavin and Perotti, 1997; Kaminsky,
Reinhart, and Vegh, 2004; Alesina, Campante, and Tabellini, 2008; Frankel, Vegh, and Vuletin, 2011;
Galeano et al., 2021).8
Developing countries often lack robust automatic fiscal stabilizers, essential for effective fiscal
stabilization. Automatic stabilizers comprise spending and revenue mechanisms that respond to economic
conditions with a slower and less discretionary nature compared to other support actions. They also
contribute to fiscal sustainability by naturally scaling back during expansions. Instruments like
unemployment insurance, targeted social protection programs, and well-structured income taxes
typically fulfill the role of automatic stabilizers. However, in low- and middle-income economies, these
mechanisms often have limited impact due to their low levels of generosity, design vulnerabilities, and
weaker labor markets marked by higher informality rates. Furthermore, some developing countries
exhibit procyclical spending patterns for remaining budget items, further undermining the stabilizing
capacity of fiscal policy. In these cases, recurrent and typically inflexible spending items tend to rise during
economic upswings, accentuating the procyclical stance and heightening fiscal sustainability risks. The
absence of robust automatic stabilizers during economic downturns exacerbates the procyclical effects of
fiscal policy initiated in the previous phase of the economic cycle.
Argentina provides a prominent example of grappling with many of these challenges. Informality is
widespread in Argentina, contributing to high tax evasion rates. Despite improvements in the social safety
net, identifying impoverished households remains an administrative challenge for authorities. Fiscal
sustainability has consistently been a concern in Argentina, with recurring fiscal imbalances and financing
strategies significantly contributing to macroeconomic instability. The Treasury has frequently resorted to
Central Bank monetization of deficits, public debt issuance, and even the use of international reserves to
finance these imbalances. These events have left the economy vulnerable to various macroeconomic
crises. Consequently, a comprehensive examination of the cyclical bias of fiscal policy in Argentina
becomes imperative. Subsequent sections of this paper are dedicated to presenting empirical evidence in
this regard.
8Noteworthy, literature reflects that while still more procyclical than advanced economies, most emerging countries managed
to reduce their level of procyclicality in the last couple of decades (Vegh, et al 2017).
3. Breaking down fiscal cyclicality in Argentina
3.1 Public Spending
Substantial evidence from recent decades suggests a strong correlation between the cyclicality of public
spending and the income level of countries. The literature on this subject underscores that advanced
nations typically employ countercyclical or acyclical fiscal policies, while less developed countries tend to
exhibit procyclical policies (Galvin and Perotti, 1997; Kaminsky, Reinhart, and Vegh, 2004; Alesina,
Campante, and Tabanello, 2008; Frankel, Vegh, and Vuletin, 2011). It is worth noting that the majority of
these studies primarily focus on public spending patterns, as a comprehensive analysis of revenue
cyclicality necessitates a more detailed investigation, including the evaluation of fluctuations in tax rates.
A cross-country comparison of public spending reveals that Argentina ranks among the most procyclical
countries globally. Figure 3 illustrates the outcomes of an analysis spanning the last three decades (1990-
2022) across 74 economies, encompassing 30 advanced and 44 developing nations. The findings are quite
evident: developing countries tend to embrace procyclical public spending policies, with a statistically
significant average correlation of 0.279 for non-LAC (Latin American and Caribbean) countries and +0.416
for LAC countries, while the countercyclical policy category predominantly comprises advanced
economies, exhibiting a statistically significant correlation of -0.251. Notably, Argentina boasts the third-
highest correlation coefficient of +0.74, underscoring its substantial procyclical stance.
Figure 3. Public Spending Cyclicality 1990-2022
1.0
0.8
Correlation between real GDP and public spending cyclical component
Procyclical
0.6
0.4
0.2
-
(0.2)
(0.4) Countercyclical
(0.6)
(0.8)
(1.0)
Seychelles
Türkiye
Singapore
Czech Republic
Mexico
Ukraine
Taiwan
Finland
Sweden
Spain
Ireland
Canada
Vietnam
Norway
Moldova
China
Mongolia
Bolivia
Slovenia
Panama
Croatia
Alb ania
Brazil
Bulgaria
Luxembourg
Portugal
Fiji
Chile
Greece
Japan
France
Belgium
Germany
Peru
New Zealand
Bosnia & H.
Iceland
Belarus
Ecuad or
Montenegro
Uruguay
Austria
El Salvad or
Tonga
Myanmar
Romania
Latvia
Australia
Malaysia
Russia
Jamaica
Micronesia
Lithuania
Serbia
Colombia
Argentina
Guatemala
Switzerland
Denmark
Netherlands
Italy
Poland
Paraguay
Hungary
Costa Rica
Korea
Estonia
Israel
Philippines
Honduras
Slovak Republic
N. Macedonia
UK
USA
Source: IMF WEO, INDEC. Notes: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott filter.
Public spending is considered in real terms, deflated by CPI. *, ** and *** indicate statistical significance at the 10%, 5% and 1% levels,
respectively.
Additionally, Argentina distinguishes itself as one of the few nations that has consistently conducted
procyclical fiscal policies over the past three decades (1990-2022) and has not managed to reduce this
procyclicality over time. The period spanning from 2011 to 2022 holds particular relevance, marking the
conclusion of the global commodities boom, which had a substantial impact on developing economies,
including Argentina. Among the 74 countries examined in the sample (comprising 30 advanced and 44
developing countries), 51 countries exhibited a shift towards reduced procyclicality or a more
countercyclical fiscal policy approach over the past decade, compared to the preceding two decades
(1990-2010). Results show that Argentina continued to pursue fiscal policy as procyclical as in previous
decades. As depicted in Figure 4, there were minimal instances of countries transitioning from a
procyclical to a countercyclical stance or vice versa. Most nations remained within the same policy stance,
irrespective of whether they increased or decreased the correlation between their spending and the
business cycle, positioning themselves in either the upper-right or bottom-left segments of the chart.
Figure 4. Public Spending Cyclicality per country over time 1990-2010 vs 2011-2022
1,0
New procyclical Still Procyclical
0,8 NIC ARG
ECU
TÜR TAI COL
0,6 FIJ
RUS MIC
GRE URU
JAM BEL
MYA HUN
0,4
Corr (Spending, GDP) 2011-2022
VIE EST BOL SER
IRE
BRA BUL LAT
ROM
0,2 TON HON ICE
MOL ALB MON
MEX
CHI CHI
0,0 PHI
MAL CR POL LIT
NZ CRO
POR UKR BOS
-0,2 CR SR SEY
PAN
Still countercyclical SLO NM
ITA ES PAR SWE
-0,4 NET
LUX KOR SPA
FIN PER MON
US GUA
-0,6 GER
DEN
FRA NOR
BEL AUS ISR
-0,8 AUS
JAP SWI
UK New Countercyclical
CAN SIN
-1,0
-1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0
Corr (Spending, GDP) 1990-2010
Source: IMF WEO, OECD, INDEC. Notes: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott
filter. Public spending is considered in real terms, deflated by CPI.
3.1.1 Determinants of spending procyclicality: automatic vs non-automatic spending items
The public budget can be categorized into two main groups: automatic and non-automatic spending (as
illustrated in Figure 5). Automatic spending includes expenditures governed by specific legislation or
constitutional mandates, typically designed to benefit individuals who meet specific eligibility criteria.
These items are structured to limit interference by policymakers due to their normative guidance.
Generally, automatic spending includes unemployment insurance benefits, social programs (e.g., cash
transfer programs, family allowances, etc.), and the social security system, primarily centered around old-
age pension spending. Social security spending typically represents the largest component of the budget
in most countries. In contrast, non-automatic spending results from budgetary decisions made by policy
makers, subject to regular discussions, usually on an annual basis. The primary categories of non-
automatic spending are public consumption (mainly encompassing wages and the purchase of goods and
services) and public investment. Although it can be argued that personnel management policies (e.g.,
teachers, doctors, police officers, civil servants, etc.) inherently possess automatic features, the wage bill
remains subject to budgetary decisions that can and do experience policy changes. Public consumption,
alongside social security, is typically among the most substantial budget items in most countries.
In theory, neither non-automatic nor automatic spending is inherently expected to exhibit a
countercyclical behavior. As mentioned earlier (Section 2), there is a broad consensus across the economic
spectrum regarding the significance of governments avoiding the adoption of procyclical fiscal policies
that would intensify the impact of the business cycle on the real economy. However, not all fiscal policy
tools are intended to counteract the economic cycle, as certain items tend to respond to long-term
dynamics.
For instance, automatic spending items such as unemployment insurance and specific social programs are
designed to absorb negative shocks, rendering them countercyclical. These are commonly referred to as
automatic stabilizers (Figure 5) since they automatically activate and deactivate based on specific criteria,
aiming to assist eligible beneficiaries. Unemployment insurance serves as a textbook example of an
automatic stabilizer, aiming to mitigate output volatility and stabilize household income and consumption
by providing income support to workers who have lost their jobs for a certain period (depending on the
institutional design of each country). It is worth noting that unemployment insurance is not prevalent in
developing countries due to their low generosity, design vulnerabilities, and, most importantly, weaker
labor markets with high informality levels, which hinder its effective operation (Duval and Loungani, 2019;
Asenjo and Pignatti, 2019). While social programs can also act as automatic stabilizers, their specific
programs and effectiveness vary significantly across countries.9
Conversely, automatic expenditures, such as old-age spending, are not expected to be consistently linked
to the economic cycle but rather to demographic trends. The amount of public funds allocated to pension
systems is influenced by both "quantity" and "price" effects, reflecting the number of beneficiaries and
the real value of benefits, respectively. As the number of beneficiaries should primarily be shaped by slow-
moving demographic trends,10 any pension system should aim to maintain benefits at a stable level in real
terms rather than react to the business cycle. This is particularly crucial in countries experiencing high
inflationary pressures, as benefits must keep pace with rising costs to ensure income stability for the
elderly. In theory, neither price nor quantity effect should co-move with the economic cycle, making the
case for old-age spending to be acyclical. Indeed, most studies examining the fiscal aspects of social
9 This usually contains all sort of programs such as: unconditional and conditional cash transfer programs to households, food
transfers programs, family allowances, etc. However, as shown in Figure 5, the labeling of this item is more debatable, as there
are lots of different programs that could potentially be categorized as “social program”. Arguably, specific social programs, such
as cash transfers aimed to protect the most vulnerable households could be expected to have a neutral link with the cycle
(acyclical), as they are oriented to assure the provision of a basic set of goods and services (or its equivalent in money) and should
avoid fluctuations. On the contrary, if we think of emergency social programs created during a crisis, these should be part of the
set of discretionary spending items. Nonetheless, we decided to schematically include “social programs” as part of the automatic
stabilizers items, thinking about automatic social benefit programs that are activated in crisis (social benefit for informal or
unemployed workers), usually imperfectly replacing a broad-based unemployment insurance.
10 For instance, in terms of coverage (“quantity” effect), even though it is possible to consider the potential effects of having more
retirements during crisis periods, these are usually long-term decisions involving both employers and employees and
materializing with much more lag than other budgetary items (i.e. automatic stabilizers).
security systems tend to concentrate on potential demographic effects on fiscal sustainability rather than
their cyclical behavior.
Similarly, while public investment and public consumption are both non-automatic in nature, they could
theoretically exhibit distinct cyclical characteristics. Capital spending is often linked with the Keynesian
perspective on the public sector, potentially serving as a countercyclical tool. However, it is not
categorized as an automatic stabilizer since it is influenced by specific policy and budgetary planning and
may involve implementation delays. On the other hand, public consumption, primarily encompassing
salaries for professions like education, health, police, and defense, is expected to have a neutral cyclical
impact because the number of employees (e.g. teachers, doctors, policemen, and military personnel) and
their wages should remain stable throughout the economic cycle (i.e., acyclical). Efforts to employ the
wage bill as a countercyclical tool may encounter issues due to the lag between a negative shock and the
implementation of an expansion in the public sector workforce, which may lead to unintended
consequences (i.e., potentially becoming an automatic destabilizing policy). Moreover, while increasing
public spending is relatively straightforward, phasing it out may generate resistance, contributing to
inflexible spending items and fiscal sustainability challenges.
Figure 5. Classification of main budgetary spending items
Public
Consumption
Non-
automatic Public
Spending Investment
Primary
Automatic Stabilizers
Spending
Unemployment
insurance
Automatic Social
Counter-Cyclical
Spending Programs
A-cyclical Social Security
System
Source: World Bank based on Vegh et al (2021). Note: automatic spending are expenditure categories that follow a specific
pre-existing legislation (or even a constitutional mandate) and tend to benfit individual meeting certain eligibility criteria. Non-
automatic spending are the result of policy makers budgetary decisions.
The literature typically concentrates on analyzing the cyclical behavior of aggregate non-automatic
spending categories, including public consumption and public investment, alongside the customary
automatic stabilizers. Empirical evidence suggests that public investment tends to exhibit countercyclical
patterns in advanced economies, aligning with the Keynesian rationale, but displays strong procyclicality
in developing countries. This phenomenon can be attributed to the fact that reducing government
investment is often the least politically challenging fiscal adjustment during economic downturns,
particularly in nations with inflexible budgetary structures (Ardanaz and Izquierdo, 2017). Public
consumption, on the other hand, tends to be procyclical in the developing world, while findings in
advanced countries are less conclusive.11
In addition to these non-automatic spending categories, automatic stabilizers like unemployment
insurance and targeted social programs have proven to be effective countercyclical tools. Their
effectiveness in smoothing household income and consumption hinges on institutional design and
budgetary size, with a more prominent role in advanced economies compared to developing nations, as
previously mentioned.
Recent studies have also delved into the role of social security spending, yielding mixed results. Darby and
Melitz (2008) discovered that retirement and pension transfers to households have a stabilizing effect in
a sample of OECD countries. Wibbels (2006) revealed that social security spending, using a broader
definition encompassing pension, has no direct link to the business cycle and is therefore acyclical, both
in OECD and Latin American countries. Vegh et al (2021) found that social security spending demonstrates
mild countercyclicality or acyclicity in advanced economies while being strongly procyclical in developing
countries. In summary, the cyclical attributes of social security spending appear to vary depending on the
country and the specific program under consideration.
3.1.2 Breaking down spending cyclicality in Argentina
In Argentina, government consumption, social security spending, and public investment constitute the
primary components of the overall budget. Based on official data from 2022, which aggregates the central
government and the 24 sub-national administrations' public accounts, public primary spending amounted
to an estimated 37 percent of GDP, with these three categories—government consumption, public
investment, and social security—accounting for almost three-quarters of this expenditure ( Figure 6).
Public consumption represents the largest portion, constituting 13.5 percent of GDP (with the wage bill
accounting for nearly 11 percent of GDP).12 Social security spending reached 10 percent of GDP, primarily
driven by the various pension sub-systems. Meanwhile, public investment comprised 3.8 percent of GDP
in 2022. The remaining 9.5 percent of GDP in spending is largely attributed to various types of transfers
to households, private and public enterprises, including energy and transport subsidies. Remarkably, the
unemployment insurance program is insignificant in budgetary matters and is equivalent to only 0.02
percent of GDP in 2022, pushing other -typically more rigid- items to replace the role of unemployment
insurance.
11
For instance, Lamo, Perez and Schuknecht (2007) show evidence that public consumption items, namely the wage bill, co-moves
with the business cycle in a procyclical way with a 1–2-year lag, for most of the euro area countries. Moreover, the authors prove
that wage bill procyclicality appears to derive mainly from wage-setting behavior as opposed to employment decisions. In a
similar vein, Holm-Hadulla et al (2010) build on previous works by showing evidence of strong co-movements between public
and private wages, suggesting that governments should be cautious so that wage-setting and employment measures do not put
macroeconomic balances at risk. Moreover, opting for a different methodology, Eckardt and Mills (2014) show that public sector
wage bill spending across Europe and Central Asia tends to behave pro-cyclically, especially in transition economies. At last,
according to Ilzetzki and Vegh (2008), public consumption is mildly procyclical in high income countries, though far less procyclical
than in developing countries.
12 The data for overall public consumption is an estimate since official figures exclude municipal spending. However, they do
include the transfers made by the provinces to the municipalities. It is assumed that 70% of these transfers to municipalities are
spent on consumption spending (given that consumption spending represents that proportion of total municipal spending and
that provincial transfers finance two-thirds of total municipal spending).
Notably, all the major budget items in Argentina exhibit a growing and pronounced procyclical tendency.
According to official data spanning from 1961 to 2022, this study reveals that the correlation between the
cyclical components of total public spending and real GDP in Argentina stands at 0.72 (statistically
significant). When dissecting public spending, the data shows that, throughout the entire period, the wage
bill, social security, and public investment demonstrated strong procyclical tendencies (correlations of
0.61, 0.56, and 0.59, all statistically significant). Intriguingly, public consumption and government
investment have exhibited an even stronger correlation with the business cycle over the past two decades
compared to the preceding 40 years (Figure 7).
The following sections will delve into the identification of the sources of procyclicality in the two primary
expenditure items, namely the public wage bill and pension spending.
Figure 6. ARG Public Spending as percent of GDP Figure 7. Breakdown of Argentina Public
2021 Spending Cyclicality per item (1961-2021)
40 0.9
37 % of GDP
Correlation between the cyclical components of real
0.8
35 0.72 ***
Others 0.7
30 spending items and real GDP 0.61 ***
0.59 ***
0.6 0.56 ***
25 Public Investment 0.5
Social Security 0.4
20
0.3
15
0.2
Public
10 Consumption 0.1
-
5
wage bill Social Security Capex Total exp
0 1961-2022 1961-1999 2000-2022
2022
Source: Ministry of Economy. Note: Includes both Federal and Source: Ministry of Economy, INDEC. Notes: the cyclical components
provincial administrations (excluding municipalities). of GDP and public spending have been estimated using the Hodrick-
Prescott filter. Public spending is considered in real terms, deflated
by CPI. *, ** and *** indicate statistical significance at the 15%, 10%
and 5% levels, respectively.
3.1.3 Public Wage bill
Argentina stands out for its substantial public wage bill compared to its regional and global counterparts.
In 2022, Argentina allocated nearly 11 percent of GDP to public employee compensation, with subnational
administrations absorbing most of these expenditures. Specifically, provinces accounted for 7 percent of
GDP, municipalities for 1.5 percent of GDP, and the central government for 2.4 percent of GDP in public
wage spending. According to data from a sample of 66 advanced and developing countries, Argentina
ranks within the top 25 percent of nations with the highest allocation to public wages (See the Annex).
Furthermore, Argentina's public wage bill has displayed significant procyclicality over the past few
decades, even when compared to international standards. Among a sample of 70 advanced and
developing countries, Argentina's wage bill policy ranked as the fifth most procyclical over the last 30
years (Figure 8). Particularly, the correlation coefficient between the cyclical components of real public
spending on wages and GDP in Argentina stands at 0.65 (statistically significant). Unlike other spending
items, such as unemployment insurance, there appears to be no clear connection between income levels
and the cyclicality of the public wage bill across countries. Over half of the advanced economies in the
sample (15 out of 27) adopted counter- or acyclical wage bill policies (i.e., a negative correlation or a
positive correlation below 0.2), compared to 49 percent of developing economies.
To comprehend the drivers behind Argentina's wage bill procyclicality, it is imperative to analyze both the
"price" and "quantity" effects. The "quantity" effect is influenced by whether a government expands or
contracts its workforce during different phases of the business cycle. Conversely, the "price" effect hinges
on whether a government decides to increase or decrease the compensation per employee during
economic booms or recessions, irrespective of the number of staff. Therefore, examining historical series
of public employment staff and their compensation measures is essential. In Argentina's case, given the
significant fragmentation of public employment across various levels of government, it is crucial to
examine the wage bill policies of subnational and central authorities.13
In recent decades, the size of the civil service ("quantities") and the average wage ("price") have both
exhibited strong procyclical patterns, both at the provincial and central government levels. The data
reveals that at the provincial level, both the "price" and "quantities" effects have contributed to overall
provincial wage bill procyclicality, with statistically significant correlation coefficients of 0.69 and 0.65 for
staff and average wage, respectively. It is noteworthy that this trend encompasses diverse realities within
each of the 24 provinces (See the Annex). As for the central administration, the results show a positive
but weaker "quantities" effect (0.39, not statistically significant), indicating that real wage policies
(positive correlation of +0.66) have played a more critical role in terms of cyclicality than personnel
recruitment policies (Figure 9).
One reason for this trend is the significant increase in public employment during the period 2002-2015,
which almost doubled the number of public servants across all levels of government. While there were
around 2 million public employees on average during the 90s, in 2015, the figure peaked at 3.7 million,
with an average yearly growth rate of 3.7 percent, almost three to four times the population growth rate.
The most significant expansion occurred at the subnational level, with the number of public servants
increasing from 1.3 to 2.2 million from the late-90s to 2015 (See the Annex). Notably, this surge in the
public sector workforce coincided with an overall expansion of primary public spending, which increased
from
24.8 percent of GDP in 2005 to 40.3 percent of GDP in 2015, partially reverting to an estimated 37 percent
of GDP in 2022.
Furthermore, the absence of automatic indexation mechanisms has contributed to the significant
fluctuations in the real value of wages during this period. Inflationary pressures have been a constant
feature of the Argentine economy, and any increase in the rate of inflation (usually accompanied by a
depreciation of the peso and a slowdown in economic activity) has had an immediate negative impact on
real wages, given the annual or semiannual frequency of wage negotiations with unions. As a result, the
execution of wage bill policies has been subject to significant shocks, which have further contributed to
their procyclical behavior.
13
There is available data for the number of staff both for provincial administrations and the National Administration. Note that
this is a subset of the overall Argentine public sector. Namely, municipalities, SOE, Trust Funds and other decentralized public
entities of the central government are not included, due to lack of available data. Regarding compensation per employee, this is
measured by the average wage bill spending per employee on each administration. Naturally, this hides a large set of
heterogeneities among public employees, which are not being captured.
Creating greater fiscal space to guarantee wage stability during crisis situations and adopting institutional
and regulatory innovations to guide personnel hiring policies would greatly benefit the state and help
reduce procyclicality. It is important to avoid large swings in the real wage of public servants (i.e. teachers,
doctors, police officers) without objective reasons. A more accurate regulatory framework to govern
hiring policies would also help prevent substantial increases in staff (typically occurring during economic
upturns) that are challenging to reverse. However, in countries such as Argentina, where the wage bill is
mainly decentralized across subnational administrations, coordinating these policies may be more
challenging. Therefore, achieving more stable wage policies is contingent on establishing coordination
mechanisms among different levels of government and ensuring the effective implementation of
regulatory frameworks.
Figure 8. Wage bill cyclicality per country 1992-2022 (or closest years)
1
EMERGING COUNTRIES ARG:
0,8
+0,65
0,6 ADVANED ECONOMIES
0,4
0,2
0
-0,2
-0,4
-0,6
-0,8
-1
Italy
Chile
France
Belgium
Switzerland
Finland
Spain
Singapore
Mexico
Peru
Azerbaijan
Egypt
Moldova
Germany
Denmark
Turkey
Romania
Colombia
Norway
Belarus
Korea
Indonesia
Canada
Albania
Australia
Slovenia
Malta
Serbia
Bulgaria
Luxembourg
New Zealand
Kazakhstan
Kosovo
Poland
Israel
Brazil
Iceland
Greece
Cabo Verde
United Kingdom
Congo
Palau
Japan
Ukraine
Thailand
Paraguay
Cyprus
United States
Austria
Netherlands
Mongolia
Lithuania
UA Emirates
Bolivia
Costa Rica
Portugal
El Salvador
Bosnia Her
Ireland
Sweden
Afghanistan
Russian Federation
Slovak Republic
Czech Republic
Croatia
Estonia
Argentina
Armenia
South Africa
Source: Ministry of Economy, INDEC, GFS IMF, OECD. Notes: the cyclical components of GDP and public spending have been estimated using
the Hodrick-Prescott filter. Public spending is considered in real terms, deflated by CPI.
Figure 9. Wage bill procyclicality: staff, average wage and level of government (1992-2022)
1,0
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0,0
National Administration Provinces
Source: Ministry of Economy, INDEC. Notes: the cyclical components of GDP and public spending have been estimated using the Hodrick-
Prescott filter. Public spending is considered in real terms, deflated by CPI. *, ** and *** indicate statistical significance at the 15%, 10% and
5% levels, respectively.
3.1.4 Pensions Spending
The public pension system in Argentina is fragmented into various sub-systems and constitutes the second
most significant expense, ranking among the highest globally. Among a sample of 33 advanced, emerging,
and regional peers, Argentina's estimated pension spending in 2022, accounting for 10 percent of GDP,
ranks within the top five countries (see the Annex). At the regional level, Argentina performs similarly to
Uruguay and Brazil, surpassing the rest of its neighboring countries.
Argentina has conducted its pension spending in one of the most procyclical manners during the past
decades (
Figure 10), and this has significantly contributed to overall fiscal procyclicality. During the period of 1995-
2021, nearly 80 percent of the countries in the sample pursued either counter-cyclical or acyclical pension
spending policies. In contrast, the correlation between the cyclical component of pension spending and
real GDP in Argentina stood at +0.53, which is statistically significant (at 5 percent level) and the highest
in the region, significantly surpassing OECD figures (-0.422, statistically significant). Therefore, instead of
remaining neutral to the economic cycle, as theory suggests, the Argentine public spending on pensions
has reinforced economic fluctuations.
Figure 10. Pension spending cyclicality (1995-2021 or closest year)
1,0
OECD: -0,422*
0,8
Latin America: +0,037
Correlation between cyclical components of GDP and pensions
Argentina: +0,53**
Procyclical
0,6
0,4
0,2
spending
0,0
-0,2
Countercyclical
-0,4
-0,6
-0,8
-1,0
Peru
Austria
Sweden
Slovenia
Italy
Estonia
Spain
Denmark
Korea
Lithuania
Norway
Poland
Hungary
Germany
Iceland
Chile
Australia
Türkiye
Greece
Slovak Republic
Israel
Portugal
Canada
Mexico
Japan
Netherlands
Brasil
New Zealand
Costa Rica
United Kingdom
Ireland
Latvia
Belgium
Colombia
Finland
United States
Luxembourg
Czechia
France
Uruguay
OECD Latin America
Source: OECD, CEPAL, Ministry of Economy, INDEC, ANSES. Notes: the cyclical components of GDP and public spending have been estimated
using the Hodrick-Prescott filter. Public spending is considered in real terms, deflated by CPI. *, ** and *** indicate statistical significance at
the 15%, 10% and 5% levels, respectively.
Figure 11. Breaking down ARG pension spending cyclicality: number and value of benefits per sub-
system (2000-2022)
0,8
Correlation between cyclical components of each variable and real GDP
0,6
0,4
0,2
0,0
-0,2
-0,4
-0,6
-0,8
(1)NCP stands for Non-Contributory pensions, including disability pensions
Notes: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott filter. Public spending is
considered in real terms, deflated by CPI. *, ** and *** indicate statistical significance at the 15%, 10% and 5% levels, respectively.
Source: Ministry of Economy, INDEC, ANSES.
What drives the cyclicality in pension spending? Similar to the wage bill, the dynamics of pension spending
are determined by both "quantities" and "price" effects - the number of benefits/beneficiaries and the
evolution of the real value of benefits. Theoretically, the number of benefits granted by a pension's
retirement scheme should be determined by demographic trends (i.e. aging) as well as structural labor
market dynamics, with little, if any, short-term determinants. Hence, an aging population or economies
with a steady reduction in potential growth, which limits formal employment generation, will face higher
fiscal pressure to support pension systems. Therefore, in the absence of structural reforms, the number
of benefits should behave neutrally with respect to the cycle. Similarly, there is no social security rationale
to see the real value of benefits co-moving with the cycle. Moreover, evidence shows that countries
lacking automatic pension indexation mechanisms tend to exhibit more procyclical behavior (Galeano et
al. 2021). Ideally, benefits should follow a stable path, ensuring income stability for pensioners.
Paradoxically, Argentina does not fit well with this finding, as it has chosen inherently procyclical
indexation mechanisms, as we will demonstrate later.
Interestingly, in the case of Argentina, the quantity effect - determined by the number of benefits - has
been slightly procyclical, with two pension schemes standing out in particular: the semi-contributory
scheme (moratorium) and the non-contributory pensions (i.e. disability and welfare pensions) ( Figure 11).
Data suggests that the procyclical performance of almost all the different pension schemes is not driven
by the number of benefits, aligning with theoretical expectations. However, the evolution of benefits
granted by the “moratorium” program does reflect a more procyclical behavior, with a cyclical correlation
coefficient of +0.28, statistically significant at a 15% level. This program -that took place between 2006-
2009 and a second phase since 2015- involved incorporating nearly 3.5 million new beneficiaries (almost
doubling the overall number of pensioners) into the regular pension system who had not contributed as
formal workers according to the normative (i.e. more than 30 years).
Moreover, in the 2000s, Argentina witnessed a significant rise in the number of disability and welfare
pension beneficiaries. The number of beneficiaries surged from an average of 0.33 million to 1.5 million
between 2004 and 2015, without any discernible objective justification for the increase, and this level has
persisted through 2022. Accordingly, when confronting this (cyclical) trend with the business cycle, results
show that the number of disability and welfare pension benefits exhibited a mildly procyclical behavior,
with a statistically significant correlation coefficient of +0.38, at a 15% significance level. Given the
substantial size of both the moratorium and, to a lesser extent, the disability pension scheme, which
together account for 58 percent of all pension benefits under the national purview, the overall "quantity"
effect appears to be slightly procyclical, with a correlation coefficient of +0.31, also significant at a 16%
level.
Furthermore, the 'price' effect (i.e. the real value of benefits), exhibited an even stronger procyclicality in
recent decades, being the primary contributor to overall retirement pension spending cyclicality, while
showing similar results for the disability and welfare scheme. The data reveals a statistically significant
cyclical correlation of +0.45 (statistically significant) between the real value of benefits and GDP for the
overall retirement pensions schemes under the central government umbrella. In a similar vein, for the
disability and welfare scheme the cyclical correlation is +0.44, statistically significant ( Figure 11).
Considering the fragmented nature of the pension system, we can further dissect these figures. Using a
newly compiled database for this study, the statistics indicate that the 'price' effect was positive in almost
all identifiable schemes, although some of them show small correlations, thus pointing towards a more
acyclical behavior. Notably, the general contributory regime managed by the central government (ANSES)
exhibited a cyclical correlation of +0.39 (statistically significant), while the semi-contributory scheme
(moratorium) also showed a positive correlation of +0.40 with statistical significance.
The “price” effect procyclicality is directly influenced by the various indexation policies in place during the
past decades. Since the early 2000s, the indexation mechanism for schemes managed by ANSES (the Social
Security Agency) has undergone several erratic reforms. In fact, this period can be divided into at least
five distinct phases with different indexation policies:
2000-2008: During this period, there was no automatic indexation. Following the 2002 economic
crisis, the government began implementing discretionary increases to pension benefits between 2003
and 2008. Initially, these adjustments were focused on minimum benefits but were later extended to
all beneficiaries. As a result, minimum benefits increased substantially in real terms, while other
benefit scales experienced mixed evolutions.
2009-2017: At the end of 2008, the National Congress passed the Social Security Mobility Law
(26,417), introducing a new automatic indexation mechanism for pension benefits. This mechanism
involved increasing benefits nominally every six months based on the evolution of wages and tax
collection per beneficiary. Therefore, increases in tax collection translated into higher benefits,
subject to a formula-imposed ceiling. This use of tax collection as part of the indexation formula was
unconventional, with only 5 percent of countries using similar approaches, while 95 percent
predominantly employed inflation and/or wage evolution (Rofman 2020). During the first half of this
period, benefits increased substantially in real terms, aligning with the expanding economy and tax
revenue.
2018-2019: In late 2017, the Social Security Mobility Law underwent reforms, adopting a new
indexation formula more in line with international standards. This formula involved quarterly nominal
indexation linked to past inflation (70 percent) and formal wage increases (30 percent), with a lag of
six to nine months. However, the three-month transition between the previous and new formulas
created issues and eroded the purchasing power of benefits.
2020: In late 2019, the incoming administration suspended the indexation formula, replacing
automatic adjustments with discretionary quarterly increases. Throughout 2020, minimum benefits
received increases relatively in line with inflation, while higher benefits levels experienced significant
declines in real terms.
2021-2023: At the end of 2020, the Congress approved a new indexation formula, resembling the one
in place during the 2009-2017 period. Consequently, the real value of social security benefits remains
strongly linked to the economic cycle, as it depends on the evolution of tax collection and formal wage
dynamics.14
The current indexation formula is inherently procyclical by design. Substantial evidence demonstrates a
strong correlation between tax collection and the business cycle, as illustrated in Figure 12.15 Therefore,
when the predominant variable for adjusting pension benefits is tax collection (as in the current
indexation formula), it implies that the indexation mechanism will broadly move in tandem with the
business cycle. The degree of cyclicality is primarily influenced by the duration of the cycle and the lag in
implementing the indexation.16
It is noteworthy that even if the indexation were based on CPI inflation, a significant implementation lag
could still produce a procyclical effect, stemming from abrupt changes in the rate of inflation. Similarly,
depending on the phase of the economic cycle, the introduction of a new (procyclical) indexation
mechanism can have lasting effects. If this kind of indexation is implemented during the end of a severe
crisis or at the onset of a positive phase of the cycle, it becomes evident that benefits would increase
rapidly as the economy and tax revenue begin to recover.
Interestingly, while indexation rules typically help prevent social security spending from becoming
procyclical in most countries, the Argentine case appears to demonstrate the opposite. Recent literature
(Vegh, et al 2021) suggests that the absence of automatic indexation can exacerbate cyclical fluctuations
in public spending. However, Argentina's uniqueness lies in its adoption of strongly procyclical indexation
rules (linked to tax collection) for 10 out of the past 20 years, along with a history of unstable indexation
rules (Figure 13). Consequently, it is natural to expect a more pronounced procyclical behavior in the
Argentine pension system.
14 At the moment of finalizing this paper, by December 2023, the new presidential administration sent a new bill to the congress
aiming to suspend again the current indexation formula, replacing it with discretionary adjustment until a new formula is
discussed in the Congress.
15 The cyclical correlation between the main tax receipts and the real GDP is very high and statistically significant in all cases. VAT,
Income Tax and selected Excise Taxes show correlation coefficients of +0.83, +0.59 and +0.60, respectively.
16 However, it is hard to predict, as it depends on the duration of the cycle and the parameters of the formula. Indeed, the only
way in which the current formula could end up having a countercyclical effect is that the lag with which the formula kicks-in is
negatively equivalent to the size of the cycle. Certainly, this is very unlikely given the current indexation lag is between 3-to-6
months.
Figure 12. Tax collection cyclicality (2000-2022): Figure 13. Real value of Minimum pension
correlation between cyclical component of real benefit throughout different indexation
GDP and tax sources. mechanisms (2001-2022)
350
0,9
0,83 *** 0,83 ***
Correlation betwween the cyclical component of tax revenue and the real
Tax collection+wages
0,8 0,68 ***
0,65 ***
300
0,7
0,6 0,39 * CPI+wages
GDP
250
0,32 *
Constant AR$
0,5
0,4
200
Tax
Collection+wages
0,3 0,10
DISCRETIONARY AUTOMATIC
0,2 150
0,1
- 100
Dec-01
Dec-03
Dec-05
Dec-07
Dec-09
Dec-11
Dec-13
Dec-15
Dec-17
Dec-19
Dec-21
Dec-23
Aug-02
Apr-03
Aug-04
Apr-05
Aug-06
Apr-07
Aug-08
Apr-09
Aug-10
Apr-11
Aug-12
Apr-13
Aug-14
Apr-15
Aug-16
Apr-17
Aug-18
Apr-19
Aug-20
Apr-21
Aug-22
Apr-23
VAT Income Tax Financial Self-Employed Fuel Taxes Cigarretes Total
Transactions
Source: Min. Economy, INDEC. Notes: the cyclical components of GDP Source: Ministry of Economy
and public spending have been estimated using the Hodrick-Prescott
filter. Public spending is considered in real terms, deflated by CPI. *, **
and *** indicate statistical significance at the 15%, 10% and 5% levels,
respectively.
In summary, public spending across major pension schemes in Argentina has consistently shown
procyclicality, driven mainly by the "price" effect and, to a lesser extent, the "quantities" effect. The real
value of benefits has exhibited a procyclical trend, influenced by fluctuating indexation rules and variables
tied to the economic cycle, such as tax collection. Only the semi-contributory and non-contributory
schemes demonstrated a procyclical "quantities" effect. Accordingly, in terms of policy implications, it
becomes vital to protect the real value of benefits and reduce reliance on exceptional measures, like
moratoriums. The implementation of a well-designed indexation formula can mitigate procyclicality and
reduce the need for frequent reforms. Additionally, broader systemic reforms, including a gradual
expansion of horizontal coverage and reducing fragmentation in a sustainable manner, can enhance the
pension system's resilience in both favorable and adverse labor market conditions, ultimately leading to
a more equitable, predictable, and robust pension system in Argentina.
3.2. Tax Policy
The literature suggests analyzing the cyclical behavior of tax policy by focusing on tracking the cyclical
patterns of tax rates rather than tax revenue. This approach is recommended due to concerns related to
endogeneity. Tax revenue fluctuates with the business cycle and non-policy factors, making it an
unreliable measure for policy analysis. The tax base's cyclical movements often dominate any effects from
changes in tax rates (Vegh, et al 2017). Therefore, examining tax rate trends is more insightful for
understanding tax policy's cyclical behavior in each country.
Recent literature has primarily focused on income taxes (personal and corporate -PIT and CIT,
respectively) and value-added tax (VAT) rates due to their significant contribution to tax revenue in most
countries (Vegh and Vuletin 2012; Vegh, et al 2017). These studies have made valuable contributions by
constructing a comprehensive cross-country database that tracks the historical series of the top statutory
tax rates for CIT, PIT, and VAT. Building upon these contributions and using this database, our analysis will
concentrate on a sample of 33 countries, including 18 advanced economies and 15 developing economies,
spanning a period of 40 years from 1980 to 2020. Additionally, we will extend the analysis to include other
taxes, specifically focusing on the case of Argentina and updating the data whenever possible. However,
it is crucial to interpret these results cautiously within the specific context of Argentina.
Argentina's tax structure deviates from international standards and theoretical recommendations, relying
heavily on unconventional tax components that are more regressive and distortive. By 2021, Argentina's
tax burden accounted for 29.1 percent of GDP (Figure 14), with nearly one-third (around 9.1 percent of
GDP) coming from unconventional taxes, including financial transaction taxes, turnover taxes, taxes on
foreign exchange purchases, and export duties.17 In contrast to developing and advanced economies
where VAT, CIT, and PIT make up a substantial portion of total tax revenue, in Argentina, they represent
only 40 percent of the total tax revenue. Therefore, conducting standard tax rate cyclical tests on PIT, CIT,
and VAT in Argentina may not provide as representative insights as in most other countries.
Figure 14. Tax burden (as percent of GDP) Figure 15. Selected tax rates (1990-2022)
40
34.1 40
35
1.4
30
29.1 Unconventional 35
Taxes
9.5 30
25
9.1 Social Security
25
20.6
20 20
2.5
9.8 VAT & Excise
5.2
15
4.3 15
8.5 Income & 10
10 8.0 Property
13.4 5
5
6.3 5.8 0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
0
OECD ARG LAC VAT CIT PIT Export Duties Financial Transaction
Source: Ministry of Economy and OECD stats. Note: LAC Source: Ministry of Economy. Note: Grey areas denote
excludes Argentina. Unconventional taxes in Argentina are economic recessions.
represented by the turnover tax (ingresos brutos), financial
transaction tax, export duties, tax on FX purchase, among
others. Data refers to 2021 or closest.
In recent decades, Argentina has experienced limited changes in its top statutory income and value-added
tax rates, while "emergency" taxes have become more prevalent and unstable. While VAT, CIT, and PIT
top statutory rates have remained relatively stable since the 1990s ( Figure 15), it is important to note that
the current tax rate database may not fully capture all policy changes. For instance, modifications in
income brackets, income tax floors, and lower marginal tax rates are not reflected in the database, making
represent overall tax collection, regardless of the final destiny of those resources (central or subnational
17 These figures
governments). Thus, these figures are different from those in the next section.
it a weak source of information for Argentina. Additionally, export duties, which are among the most
distortive taxes, have exhibited significant fluctuations (Figure 15).
To measure the cyclicality of tax rate policy, it is crucial to consider that tax rates tend to vary very little
or remain unchanged compared to government spending and other macroeconomic variables. While
government spending is a continuous process, changes in tax rates typically result from parliamentary
discussions, leading to a more rigid dynamic. Consistent with recent findings (Vegh Vuletin, 2015) and
based on the database used (i.e., CIT, PIT, and VAT top rates), the average frequency of tax rate changes
per year ranges from 0.12 to 0.20 globally. Moreover, the frequency of changes in Argentina is even lower,
approximately 0.10. This means that top-statutory tax rates of the three main taxes change, on average,
once every 10 years. Therefore, while the usual approach to measure cyclicality involves using the cyclical
component of a fiscal variable (e.g., government consumption), it is more appropriate to utilize the
percentage change in tax rates for analyzing the cyclicality of tax policy. A positive (negative) correlation
between the change in tax rates and real GDP indicates a countercyclical (procyclical) tax policy stance.
Furthermore, to capture the overall cyclical behavior of taxes in each country, a tax index is constructed,
which comprises a weighted average of the three tax rates (weighting on their respective share of tax
revenue).
������i = ������������������i ∗ ������������������������ℎ������CIT + ������������������i ∗ ������������������������ℎ������PIT + ������������������i ∗ ������������������������ℎ������VAT
When analyzing the three main taxes included in the tax index, it appears that Argentina has pursued a
relatively acyclical tax policy. The correlation between changes in the tax rate index and the real GDP cycle
for Argentina is +0.27, although not significantly different from zero. This suggests a tax policy that is
neutral in terms of its cyclical impact, at least for the subset of taxes analyzed (CIT, PIT, and VAT).
Furthermore, in a cross-country comparison, unlike what has been observed on the spending side, the
cyclicality of tax policy appears to be less correlated with the country's income level ( Figure 16).18 The
analysis of each individual tax yields similar results for Argentina (see Annex C). In all three cases,
Argentina is positioned towards the right end of the charts, indicating a positive correlation with GDP.
However, these correlations are not significantly different from zero, suggesting that, at least regarding
the top statutory rate of these taxes, Argentina has pursued a relatively acyclical tax policy in recent
decades. In relative terms, these results are more favorable compared to those of its peers.
Figure 16. Correlation between the change in the Tax Index and cyclical real GDP (1990-2020)
18
See the figures in the Annex for individual cyclical correlation between PIT, CIT and VAT rates with real GDP.
1
0,8
0,6
0,4
0,2
0
-0,2
-0,4
-0,6
-0,8
-1
Korea
Peru
South Africa
Austria
Mexico
Costa Rica
Malta
Spain
Denmark
Latvia
Canada
Japan
Lithuania
Greece
Sweden
Colombia
Australia
Germany
Bolivia
Italy
France
Norway
Uruguay
Philippines
Brazil
Argentina
Belgium
Portugal
Honduras
New Zealand
United States
Chile
United Kingdom
Source: Min. Economy, INDEC. Notes: the cyclical components of GDP and tax rates have been estimated using the Hodrick-Prescott filter.
A positive (negative) correlation indicates a countercyclical (procyclical) tax policy.
However, as previously mentioned, traditional taxation in Argentina is less representative due to the
significant role played by a wide range of heterodox taxes, as depicted in Figure 14. Unusual sources of
revenue, which are typically more distortive, contribute to almost one-third of taxation in Argentina.
Unfortunately, the availability of data poses a constraint when attempting to expand the analysis to
include all sources of taxation, including these heterodox taxes. Specifically, the turnover tax presents
extreme complexity as it is collected in different ways by each of the 24 provinces. Furthermore, it varies
across economic sectors, and historical data availability is limited. Consequently, it becomes challenging
to establish a single historical series to measure the effect of cyclicality. However, for export duties and
financial transaction taxes, both collected at the federal level, readily available information is accessible.
Export duties and financial transaction taxes in Argentina exhibit a negative correlation with the economic
cycle. Historically, export duties were initially implemented as emergency measures to address the
inflationary impact of significant peso depreciation, typically during balance of payment crises. In line with
this, export duties were reintroduced in 2002 following a sharp devaluation of the peso. However, unlike
previous crisis episodes, export duties have remained a regular source of tax revenue even after the
recovery from the 2002 crisis, persisting for more than two decades. The structure of export duties is
complex, with varying tax rates for each product or sub-product. Notably, export duties differ across
sectors and even within agricultural products. For the purpose of this study, an export tax rate index was
constructed, assigning weights to each product based on its contribution to total exports. Figure 15
illustrates significant changes in export duties over the years. These fluctuations in tax rates demonstrate
a negative correlation with real GDP (-0.26), suggesting a procyclical behavior ( Figure 17). However, the
correlation is weak and not statistically significant. Similarly, the financial transactions tax was initially
introduced as an emergency tax during the 2001-2002 crisis but has since become a regular tax with a
constant rate of 1.2 percent. With minimal changes since its implementation, the correlation with the
economic cycle is very weak, albeit negative (-0.12), as it was introduced during the economic recession
of 1999-2002.
Figure 17. Correlation between change in tax rates and cyclical component of real GDP in
Argentina (1990-2020)
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
VAT CIT PIT Financial Export Duties
Transaction
Source: Ministry of Economy, INDEC. Notes: the cyclical components of GDP and tax rates have been estimated using the
Hodrick-Prescott filter. A positive (negative) correlation indicates a countercyclical (procyclical) tax policy. Note: results are
not statistically significant.
In summary, the analysis indicates that Argentina has pursued an acyclical tax policy, with traditional tax
sources such as CIT, PIT, and VAT being less representative of overall tax revenue due to the dominance
of unconventional tax sources. While correlations between traditional tax sources and real GDP are
positive at both the aggregate and individual levels, they lack statistical significance, suggesting an
absence of a clear relationship. Conversely, correlations between changes in tax rates for unconventional
taxes like financial transactions and export duties and real GDP are negative, indicating an acyclical
behavior. However, the study has limitations. It only covers the three traditional taxes, overlooking other
significant sources of taxation, which is particularly relevant for countries like Argentina with a prevalence
of non-traditional taxation. Additionally, using a single representative rate per tax may not accurately
reflect the actual situation within each tax structure. Furthermore, it is important to note that correlations
do not imply causality, and a more comprehensive analysis controlling for endogeneity issues would be
necessary to establish causality. Nevertheless, previous studies (Vegh & Vuletin, 2012 and 2015; Ilzetzki
& Vegh, 2008; Carneiro & Odawara, 2015) have demonstrated that the results hold even after addressing
these issues using instrumental variables.19
19 Thesepapers tend to use the following as instrumental variables: a weighted index for real GDP growth of trading partners,
changes in price of exports and change of real returns on US Treasury Bills.
4. Estimating the Structural (cyclically adjusted) Budget Balance for Argentina 1993-2022
Fiscal policy analysis often centers on a government's observed fiscal balance, but it is vital to consider
the impact of both permanent and temporary factors. Permanent elements are those expected under
normal circumstances, aligning with optimal economic conditions. In contrast, temporary factors, such as
increased social expenditure during a recession, lead to deviations from the economic trend. Failing to
differentiate between these components can result in inaccurate assessments, prompting unnecessary
adjustments. Notably, some fiscal deficits may not always be detrimental, and surpluses may not always
be advantageous.
The structural budget balance, or cyclically adjusted primary balance (CAPB), refines the fiscal balance by
excluding revenues and expenditures influenced by temporary factors. This adjustment provides a more
accurate assessment by eliminating the impact of economic cycles on the observed fiscal balance. For
instance, the CAPB excludes revenues boosted by transitory events or lost due to temporary recessions.
This treatment extends to taxes influenced by external shocks, like the 2000s commodities super-cycle in
Argentina and the LAC region.
4.1. Methodology to estimate CAPB
The CAPB is obtained by adjusting the original components—revenues and expenditures—based on the
economic cycle phase. In practice, observed levels are proportionally adjusted according to the potential
output to actual output ratio, considering built-in elasticities that capture the relation between the output
gap and changes in revenues and expenditures. Van den Noord (2000) defines the CAPB, ������s as:
s s
∑i Ti -G +X
������s = Ys
(1)
where ������is is the structural component for the ith category of tax, ������ s is the structural current primary
government expenditure, ������ refers to non-tax revenues, and ������s is the level of potential output. To refine
the fiscal stance, especially in Latin American countries, we deduct interest payments on public debt from
expenditures, as proposed by Alberola and Montero (2006). As mentioned, we refer to structural
components as the elements adjusted proportionally according to the ratio of potential output to actual
output and the assumed built-in elasticities. This is
Ts Ys ai Gs Ys f3
i =( ) ; = ( ) (2)
Ti Y G Y
where ������ , ������ and ������i are the observed output, current primary government expenditure, and receipt of the
ith tax category respectively. ������ refers to the elasticity of ith tax category with respect to output while ������
is the elasticity of current government expenditures with respect to output.
It is essential to note that the cyclical component of the budget balance is influenced by estimated output
gaps, weights of tax revenues per category and current primary expenditure, and the built-in elasticities.
From the observed budget balance (������ ) we can also derive the cyclical component (������e) as
������e = ������ − ������s (3)
In contrast to other regional studies this approach will focus exclusively on the Federal Government
Budget Balance. Many comparable studies center their analyses on the evolution of the General
Government balance (Alberola & Montero, 2006; Armendáriz, 2006; Basco et al, 2007; Daude et al, 2010).
However, this section opts for a more specific focus on the federal government budget balance for two
noteworthy reasons. Firstly, in Argentina´s context, the trajectory of the general government fiscal
accounts is predominantly influenced by the federal government´s accounts. With the exception of a brief
period in the late-1990s, sub-national administrations (i.e., provinces) have generally maintained
balanced or slightly deficit-ridden public accounts (Figure 18). Secondly, in accordance with Musgrave's
seminal work in 1959, the stabilizing function is typically entrusted to the central government.
Figure 18. National and Subnational fiscal balances (1961-2022)
4
2
0
-2
Percent of GDP
-4
-6
-8
-10
-12
-14
-16
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
2022
Federal Government Provinces
Source: Ministry of Economy.
Conforming to the literature, we categorize taxes into four distinct groups, each exhibiting different
elasticities to the economic cycle. Notably, considering that the federal government retains roughly 42
percent of the receipts for various tax categories in Argentina, this study simplifies its analysis by relying
on the resulting share of resources officially recorded as central government revenue, an estimation based
on data published by the Ministry of Economy (Figure 19).20 21 These categories are as follows: i) Personal
Income Tax (PIT), averaging 2.3 percent of GDP over the past decade, with nearly 1.1 percent of GDP
allocated to the Federal Government; ii) Corporate Tax (CIT), constituting an average of 2.9 percent of
GDP in the last 10 years, of which 1.3 percent of GDP remained in the Federal Government; iii) Social
Security Contributions (SSC), representing approximately 6.2 percent of GDP on average during the last
decade; iv) Indirect taxes, aligned with developed countries, emerge as the most significant source of
taxation. The portion retained by the Federal Government averaged 8.8 percent of GDP in the last decade,
with VAT serving as the primary tax source (3.5 percent of GDP).
20 More than 80 percent of taxation in Argentina is collected by the central government, as part of the fiscal federalism
arrangement. Part of the centrally collected taxes is automatically channeled to the revenue sharing scheme ( Regimen de
Coparticipacion), which shares the proceeds between the central government and the provinces. Currently almost 60% of the
eligible tax collection is transferred to provinces, while 40% remains at the central government.
21 It is worth mentioning the role informality might play in different phases of the cycle. In developing economies, where the
informal sector is more prominent and higher growth rates are more common, Heintz (2006) suggests that informality may
decline, posing an additional effect on tax receipts.
Figure 19. Tax categories as a share of the GDP (1993-2022)
Source: Ministry of Economy.
Elasticities were primarily calculated through regression analysis (as in Alberola & Montero, 2006) for each
tax receipt considered (CIT, PIT, SSC). In order to filter out the cyclical part of revenues, the elasticity of T
to output is computed from the following regression.
log ������t = ������ + ������ log ������t + ������t (4)
The output elasticity of indirect taxes with respect to the economic cycle is set to unity given that is mostly
proportional to its main tax base, an approach widely adopted by the OECD Secretariat (Van den Noord,
2000, Daude et al., 2010, Girouard and André, 2005). Regarding the adjustment of expenditure for
changes in economic activity, Current primary expenditure (������ ) of the Federal government is assumed to
fluctuate in proportion to unemployment-related expenditure. Nevertheless, Giorno et al. (1995) point
out that while it would be preferable to adjust unemployment benefit expenditure directly (and any other
cyclical components of government expenditure), even with significant increases in unemployment, these
expenditures remain a small part of total government expenditures. Due to the lack of data and given the
insignificance of unemployment-related benefits in Argentina, we assume, as others, that current
expenditures do not respond automatically to the cycle at all (Daude et al., 2010; Alberola & Montero,
2006).
To calculate the potential output, we apply the Hodrick-Prescott (1997) filter, a methodology widely used
to derive the long-term output trend. The HP filter refers to a data smoothing technique, typically used to
remove short-term fluctuations associated with the business cycle.22 Removing these short-term
22Some other CAPB exercises make use of the total factor productivity method to calculate potential output. However, for the
case of Argentina, the size of the informal economy undermines the confidence of any data availability, particularly regarding the
labor statistics that need to be considered for the approach. Capital stocks are also difficult to measure in general, and more
fluctuations helps reveal long-term trends of variables, such as GDP series. The availability of a long time
series of the Argentinian GDP (1961-2022) that far exceeds the period analyzed allows for a more precise
estimation of the cycle.23
This study explicitly addresses the direct impact of commodity export taxes on the structural budget
balance. Export taxes on food products are tied to global commodity prices. Accordingly, when these
prices soar, the governments’ tax revenues increase, immediately improving their fiscal position. It is
worth noting that second round effects from a spike in commodities prices boosting the agricultural
production and exports, are not being captured by this approach, as this exercise focuses exclusively on
direct tax revenue stemming from export duties. These sources of taxation are not commonly used but
are particularly relevant in a few commodities-reliant developing countries. Thus, several studies
recalculate the receipts from these sources of taxation by applying a long-term price adjustment related
to the commodities exported by the country under analysis (Marcel et al., 2001; Alberola & Montero,
2008; Basco et al., 2007; Daude, 2010). This study considers export duties generated by commodities
exports, thus leaving aside export duties applied to non-commodities products.
∑ Ts-Gs+X+Rs
������s = i i C
(5)
Ys
From (5) we can account for the impact of commodities prices through the receipts obtained from export
s) from the following equation
duties. We define structural revenues related to commodities (������e
y
������s = ������et (p lt) (6)
et pt
where ������e are the export duties obtained strictly from commodity products, ������l is the long-run price of
commodities and ������ is the spot price of commodities. Price elasticity ������ is considered to be unitary following
the approach of Marcel et al. (2001), Zettelmeyer & Hollar (2008), and Daude et al. (2010).
Argentina introduced export duties after the 2002 crisis to capitalize on the devaluation of the currency
among other considerations. This source of taxation was typically used in the past as an emergency source
of funding during a balance of payment crisis. The sharp devaluation of the peso in early 2002 and its
potential impact on domestic prices was an opportunity to establish an export tax on all products, with
heterogeneous tax rates: higher for the main agri-food export products and oil, and lower for processed
products such as heavy manufactures. Figure 20 shows the evolution of export duties as a share of the
GDP, discriminating between those arising from commodities and non-commodities sources. As observed,
the increase of global commodities prices during the 2000s contributed to raising tax revenues from this
source. Notably, the 2008 peak of export duties, contributing to fiscal revenue with almost 3 percent of
GDP, matches the first boom peak in the so-called commodities super-cycle.
To understand the impact of export duties on public accounts, long-term prices of commodities are
considered. We follow Marcel et al. (2001) and Zettelmeyer & Hollar (2008) assuming a unitary elasticity
price of the exports and adjusting the revenues applying the long-term price of the exports considered.
prominently in developing countries. It is also worth considering that in unstable economies like the Argentinian, the concept of
potential output that points to the use of the full capacity of production factors loses some clarity (Alberola & Montero, 2006).
23 Real GDP series for Argentina is constructed based on ONP 2004, Ferreres, and current INDEC data.
We do this by considering a 10-year moving average of the IMF commodity Terms of Trade database. 24
Figure 21 plots the index considered and the moving average that seeks to replicate a notional equilibrium
price of the country’s export basket.
Figure 20. Export duties as a share of the GDP (1993-2022)
3,5%
3,0%
2,5%
2,0%
1,5%
1,0%
0,5%
0,0%
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
Commodity-related
Source: Own estimates based on AFIP.
Figure 21. Export prices (spot and long-term trend; Index 2012=100; 1993-2022)
120
p p*
100
80
60
40
20
0
Source: Own estimates based on IMF Commodity Terms of Trade index.
Since the introduction of export duties in 2002, commodity prices almost doubled until 2008 and then
peaked again in 2012. Since 2014 onwards, prices returned to values comparable to the earlier 1980s (not
displayed in the graph) before rising again as a consequence of the Covid-19 crisis and, more recently, the
war in Ukraine. Using the 10-year moving average prices ( p*) permits smoothing cyclical shifts. Regardless
of the discussion around the transitory or permanent feature of the 2000s so-called commodities super-
24The Commodities Terms of Trade database can be found at https://data.imf.org/. This and other indices are built from the
change in the international prices of up to 45 individual commodities and are weighted using commodity-level trade data. We
considered the yearly average of the Commodity Export Price Index, Individual commodities weighted by the Ratio of Exports to
Total Commodity Exports (June 2012 = 100).
cycle, notably Argentina maintained its export duties as a regular source of taxation even after the
economy fully recovered from 1999-2002 crisis and after the end of the commodities super cycle,
providing annual fiscal revenue for more than 1.7 percent of the GDP (on average) since its
implementation.
4.2. Results
As stated before, the CAPB considers only the structural components of tax receipts and public
expenditures. Figure 22 exhibits the Federal Government observed primary balance from 1993 to 2021 as
well as the estimated CAPB. Accordingly, the chart breaks down the difference between the non-
commodities cyclical components of the budget and the cyclical contribution of commodity-related
revenues.
Figure 22. Cyclically adjusted primary balance as share of GDP (1993-2022)
4,00%
2,00%
0,00%
-2,00%
-4,00%
-6,00%
-8,00%
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
Cyclical Commodities Structural Budget Balance Observed
Source: Own estimates based on INDEC, Min. Economy, IMF.
During the "convertibility plan"25 (1991-2001), the Federal Government successfully achieved a balanced
primary fiscal position. This accomplishment was a result of both domestic fiscal policy reforms and
favorable economic conditions. Specifically, positive cyclical resources played a significant role,
contributing over 1.4 percent of GDP at the peak of the economic cycle in 1998.
However, the outbreak and immediate aftermath of the 2001/2002 crisis had a considerable impact on
tax receipts. This can be observed from the negative cyclical component experienced during the downturn
from 2001 to 2005. To mitigate the effects of the crisis, the government introduced export duties, which
not only helped stabilize public accounts but also led to an improvement in the cyclically adjusted primary
balance (CAPB). 26
25 The Convertibility plan was the monetary reform initiated in 1991, consisting in the implementation of a hard peg between
the Argentine peso and the US dollar (AR$ 1 equals US$ 1).
26 The considerable improvement observed in the fiscal accounts after 2002 was the result of export duties introduced and a
substantial fall in the current primary government expenditure which went down in real terms by 19.6% in 2002 versus the 1999
figures.
In the late 2000s, the country experienced a sustained surplus in its fiscal accounts, primarily driven by a
robust economy and favorable commodity prices. The combination of these two factors resulted in cyclical
revenues contributing almost 3 percent of GDP in 2008, which marked the last year the country exhibited
a surplus in its fiscal accounts.
Special attention should be given to the cyclical revenue derived from commodities. During the peak 27 of
the "supercycle" in 2008, our estimates indicate that commodities-related export duties accounted for
nearly 1.1 percent of GDP in additional receipts.28 Furthermore, if a hypothetical stabilization fund based
on commodities prices had been established, it could have collected the equivalent of 2 percent of GDP
between 2006 and 2008. This period coincided with a fiscal surplus and high commodity prices, which
provided extra cyclical resources.
Starting from 2009, the public sector experienced a significant deterioration in its fiscal accounts, running
large deficits primarily due to discretionary expansionary policies until the mid-2010s. This led to a further
deepening of the structural deficit. Between 2016 and 2019, the government implemented a
consolidation path, aiming to achieve a virtual balance by 2019. This was done in the context of a program
with the IMF and a balance of payment crisis. In 2020, the response to the Covid-19 crisis resulted in an
expansion of the deficit to over 6 percent of GDP, with a cyclical impact of -3.3 percent of GDP.
Unlike the aftermath of the 2009 global financial crisis, by 2022, the country faced a challenging post-crisis
period with a worse structural fiscal position. The financial crisis taken place in 2008 shook up numerous
economies around the world, forcing severe fiscal adjustments. This was not the case in many Latin
American economies, including Argentina, for which strong fiscal positions enabled them to face the crisis
in relatively good shape (Daude et al., 2010). However, Argentina has struggled with macro-fiscal
imbalances since the end of the commodity super-cycle. Even as commodity prices began to decline from
2012 to 2015, Argentina's public recurrent spending continued to rise. Limited buffers and procyclical
policies have further exacerbated the impact of external shocks, such as drought, the pandemic, and rising
fuel prices. After a three-year recession and almost a decade of stagnation, along with a severe balance
of payments and fiscal crisis (including a debt restructuring operation in 2020), the country now faces the
challenge of resuming growth with a much weaker structural fiscal position compared to a decade ago.
4.3. Discretionary fiscal policy cyclicality
By analyzing the variation in the CAPB along the business cycle, it is possible to estimate the cyclical stance
of discretionary fiscal policy. Under this approach, a fiscal policy is considered counter-cyclical if the CAPB
increases (reducing the deficit or increasing the surplus) during a year with a positive output gap, or if the
CAPB decreases (reducing the surplus or increasing the deficit) when the output gap is negative (Daude
et al., 2010). Figure 23 depicts the changes in the CAPB and the level of the output gap from 1993 to 2021.
Our analysis confirms that the fiscal policy in Argentina exhibits a procyclical stance. Only a few years
exhibit a countercyclical mix between the output gap and variation of CAPB. Moreover, excluding 2002
and 2020 for their exceptional nature, the overall correlation between the output gap and changes in the
27 Our methodology considers the year 2008 the peak as it is the year where the spot price (p) reaches the highest point relative
to the long-term price (p*) considered, even though prices were higher in 2012.
28 These results are below other comparative studies which used other approaches (Basco et al., 2007; Alberola & Montero,
2008; Daude et al., 2010).
CAPB is negative and significant at -0.43. To deepen our comprehension of the change in the structural
balance, Figure 24 illustrates the fiscal response in different phases of the cycle.
Figure 23. Output gap and change in adjusted budget balance (1994-2022)
Source: Own estimates based on Ministry of Economy, INDEC, IMF.
Figure 24. Output gap and CAPB variation (1994-2022)
Panel A. Positive output gap Panel B. Negative output gap
Source: Own estimates based on Min Economy, INDEC, IMF.
The analysis indicates that discretionary fiscal policy in Argentina is more pro-cyclical during economic
booms, when the output gap remains positive (with a correlation of -0.41). While an uptick in economic
activity can lead to increased tax revenue and automatic stabilizers, this does not translate into a stronger
fiscal position. The higher procyclicality during booms compared to crises (with a correlation of 0.05)
suggests the need for additional mechanisms to ensure effective use of public funds during positive
phases. Unfortunately, there has been little progress over time. Figure 25 illustrates this trend by dividing
the 29-year sample into three decades. It shows that the correlation between the output gap and changes
in CAPB increased significantly in the past two decades compared to the 1990s, indicating a more pro-
cyclical fiscal stance, a finding that aligns with Daude et al. (2010).
Figure 25. Output gap and change in CAPB (1994-2021)
Panel A. The 1990s Panel B. The 2000s Panel C. The 2010s
Source: World Bank based on Ministry of Economy, INDEC, IMF.
Results show that Argentina's fiscal policy has exhibited a procyclical nature, failing to effectively stabilize
the economy. Expansionary fiscal policies during economic upswings have hindered improvements in
savings and raised sustainability concerns. This procyclicality amplifies economic fluctuations rather than
mitigating them. Adopting a countercyclical fiscal approach would benefit the country by promoting fiscal
discipline during economic booms, enhancing resilience to economic fluctuations, and ensuring a more
sustainable fiscal framework.
5. Concluding remarks
This paper contributes to the existing literature on fiscal cyclicality analysis by adopting a multi-layered
approach that encompasses both aggregate fiscal items and specific policy components. Updating
previous studies, our analysis confirms that Argentina has exhibited pronounced procyclicality in its fiscal
policies, particularly driven by public spending. Furthermore, focusing on specific fiscal items such as
pensions, the wage bill, and traditional tax sources, our findings provide up-to-date evidence that aligns
with broader literature highlighting the country's pro-cyclical tendencies.
This paper goes further than existing literature by dissecting the sources of procyclicality within these
specific fiscal items. By examining the price and quantity effects, we uncover crucial insights into the
underlying dynamics driving fiscal procyclicality in Argentina. Our analysis reveals that pension spending
in Argentina is primarily influenced by price effects, attributable to the erratic and poorly designed
indexation mechanisms. In contrast, the public wage bill displays strong procyclicality driven by both
quantity and price effects, underscoring the complexity of fiscal dynamics in the country.
Looking ahead, the methodology proposed in this paper offers a promising avenue for further research.
Expanding this approach to other fiscal items in Argentina and replicating it in other countries could
provide a comprehensive understanding of fiscal procyclicality dynamics globally. Such insights hold
significant implications for policy makers, offering valuable guidance in designing more effective policy
levers to achieve fiscal policy objectives, including macroeconomic stabilization. By leveraging these
findings, policy makers can enhance the resilience and effectiveness of fiscal policy frameworks, ultimately
contributing to more stable and sustainable economic outcomes.
References
Acemoglu D., Johnson S., Robinson J., Thaicharoen Y., 2003. Institutional Causes, Macroeco-nomic
Symptoms: Volatility, Crises and Growth. Journal of Monetary Economics 50 49–123.
Aghion P., Marinescu I., 2007. Cyclical Budgetary Policy and Economic Growth: What Do We Learn from
OECD Panel Data? NBER Macroeconomics Annual 2007, Volume 22
Alberola E., Montero J. M., Braun M., Cordella, T., 2006. Debt Sustainability and Procyclical Fiscal Policies
in Latin America. Economía, 7(1), 157–193. http://www.jstor.org/stable/20065509
Alesina A., Campante F., Tabellini G., 2008. Why is fiscal policy often procyclical? Journal of the European
Economic Association 6 (5), 1006-1036.
Álvarez-Corrales C., 2017. Parámetro de suavizamiento del filtro Hodrick-Prescott para Costa Rica.
Ardanaz M., Izquierdo A., 2017. Current expenditure upswings in good times and capital ex-penditure
downswings in bad times? New evidence from developing countries. IDB Working Paper No. 838.
Armendáriz E., 2006. La sostenibilidad de la deuda pública y postura fiscal en el ciclo económico: El
Istmo Centroamericano. México, DF : Naciones Unidas, CEPAL, Unidad de Desarrollo Económico.
Aryeetey E., 2015. The Informal Economy, Economic Growth and Poverty in Sub-Saharan Africa.
Asenjo A., Pignatti C., 2019. Unemployment insurance schemes around the world: Evidence and policy
options, International Labour Office, Research Department, Working Paper No. 49.
Avendaño R., Reisen H., Santiso J., 2008. The Macro Management of Commodity Booms: Africa and Latin
America's Response to Asian Demand, OECD Development Centre Working Papers, No. 270, OECD
Publishing, Paris, https://doi.org/10.1787/240356635678.
Basco E., Castagnino T., Katz S., Vargas S., 2007. Política Monetaria en Contextos de Incertidumbre,
Cambio de Régimen y Volatilidad Pronunciada.
Baumann Fonay I., Cohan L., 2018. Crecimiento Económico, PTF y PIB Potencial en Argentina. Working
Papers 0001, Secretaría de Política Económica, Ministerio de Hacienda, revised June 2018.
Baunsgaard T., Symansky S. A., 2009. Automatic Fiscal Stabilizers. IMF Staff Position Note 09/23. [141]
Barro, R. (1979). On the Determination of the Public Debt. Journal of Political Economy 87 (1979), 940–
971.
Braun M., di Gresia L., 2003. Towards effective social insurance in Latin America: the im-portance of
countercyclical fiscal policy. IDB, Departamento de Investigación.
Caballero R., Cowan K., Engel E., Micco A., 2004. Effective Labor Regulation and Microeco-nomic
Flexibility. Working Paper 10744. Cambridge, Mass.: National Bureau of Economic Re-search.
Carneiro F., Odawara R., 2016. Taming Volatility: Fiscal Policy and Financial Development for Growth in
the Eastern Caribbean. The World Bank: Washington, D.C.
Calvo G., 1998. Capital flows and capital market crises: The Simple Economics of Sudden Stops , Journal of
Applied Economics, Vol1, No1.
Cetrángolo O., Grushka C., 2020. El sistema de pensiones en la Argentina. Institucionalidad, gasto público
y sostenibilidad financiera. Serie Macroeconomía del desarrollo 208. CEPAL.
Darby J., Melitz J., 2008. Social Spending and Automatic Stabilizer in the OECD. Economic Policy 2008. pp
715-756
Daude C., Melguizo Á., Neut A., 2010. Fiscal Policy in Latin America: Countercyclical and Sustainable at
Last?, OECD Development Centre Working Papers, No. 291, OECD Publishing, Paris,
https://doi.org/10.1787/5km8zdcp7j41-en.
Duval R.A., Lougani P., 2019. “Designing Labor Market Institutions in Emerging and Developing
Economies: Evidence and Policy Options”, IMF Staff Discussion Note, SDN/19/04.
Eckardt S., Mills Z., 2014. What Goes Up Must Come Down -- Cyclicality in Public Wage Bill Spending. World
Bank Policy Research Working Paper No. 6760.
Available at SSRN: https://ssrn.com/abstract=2386354
Fatas A., Mihov I., 2001. Government size and automatic stabilizers: International and intra-national
evidence. Journal of International Economics 55 (1), 3-28.
Fatás A., 2002. The Effects of Business Cycles on Growth. Working Papers Central Bank of Chile 156,
Central Bank of Chile.
Fatas A., Mihov I., 2003. The case for restricting fiscal policy discretion. The Quarterly Journal of Economics
118 (4), 1419-1447.
Fatas A., Mihov I., 2012. Fiscal policy as a stabilization tool. The BE Journal of Macroeconomics 12 (3).
Frankel J. A., Vegh C.A., Vuletin G., 2011. On Graduation from Fiscal Procyclicality. National Bureau of
Economic Research, Working Paper 17619, Cambridge.
Freedman C., Laxton C., Kumhof M., 2008. Deflation and Countercyclical Fiscal Policy. Mimeo, December
2.
Gali J., 1994. Government size and macroeconomic stability. European Economic Review 38 (1), 117-132.
Galeano L., Izquierdo A., Puig J.P., Vegh C.A., Vuletin G., 2021. Can Automatic Government Spending be
Procyclical? National Bureau of Economic Research, Working Paper 28521, Cam-bridge.
Garcia-Cicco J., Nuñez F., 2021. El Filtro De Hodrick-Prescott Y El Covid. ¿La Puñalada Final?. Central de
Ideas. https://centraldeideas.blog/el-filtro-de-hodrick-prescott-y-el-covid-la-punalada-final/
Gavin M., Perotti R., 1997. Fiscal policy in Latin America. NBER Macroeconomics Annual 12, 11-61.
Giorno C., et al. 1995, Estimating Potential Output, Output Gaps and Structural Budget Balances, OECD
Economics Department Working Papers, No. 152, OECD Publishing, Paris,
https://doi.org/10.1787/533876774515.
Girouard N., André C., 2005. Measuring Cyclically-adjusted Budget Balances for OECD Countries, OECD
Economics Department Working Papers, No. 434, OECD Publishing, Paris,
https://doi.org/10.1787/787626008442.
Heintz J., 2006. Globalization, Economic policy and Employment: Poverty and Gender Implications,
Published Studies heintz_gender, Political Economy Research Institute, University of Massachusetts at
Amherst.
Heymann D., 2006. Buscando la tendencia: crisis macroeconómica y recuperación en la Argentina.
Hicks N., Wodon Q., 2000. Economic Shocks, Safety Nets and Fiscal Constraints: Social Pro-tection for the
Poor in Latin America. Paper presented at the XII Seminario Regional de Política Fiscal en Chile, January.
Hnatkovska V., Loayza N., 2005. Volatility and Growth. In J. Aizenmann and B. Pinto, eds., Managing
Economic Volatility and Crises. Cambridge, United Kingdom: Cambridge University Press.
Holm-Hadulla F., Kamath K., Lamo A., Perez J., Schuknecht L., 2010. Public wages in the Euro Area: Towards
Securing Stability and Competitiveness, European Central Bank, Occasional Paper Series, No 112/2010.
Ilzetzki E., Vegh C. A., 2008. Porcyclical Fiscal Policy in Developing Countries: Truth or Fiction, NBER,
Working paper 14191, DOI: 10.3386/w14191.
Izquierdo A., Pessino C., Vuletin G., 2018. Better spending for better lives: How Latin America and the
Caribbean can do more with less. DIA (Development in the Americas) Flagship, Inter-American
Development Bank, Washington, DC.
Kaminsky G., Reinhart C., Vegh C. 2004. When it rains, it pours: Procyclical capital flows and
macroeconomic policies. NBER Macroeconomics Annual 19, 11-82.
Lamo A., Pérez J., Schuknecht L., 2007. The Cyclicality of Consumption, Wages, and Employment of the
public sector in the Euro Area, European Central Bank, Working Paper Series No 757/2007.
Loayza N.V., Ranciere R., Servén L., Ventura J., 2007. Macroeconomic volatility and welfare in developing
countries: an introduction. The World Bank Economic Review 21 (3), 343-357.
Maravalle A., Rawdanowicz Ł., 2020a. How effective are automatic fiscal stabilisers in the OECD
countries?, OECD Economics Department Working Papers, No. 1635, OECD Publishing, Paris,
https://doi.org/10.1787/f1fb9d6a-en.
Maravalle A., Rawdanowicz Ł., 2020b. Automatic fiscal stabilisers: Recent evolution and policy options to
boost their effectiveness, OECD Economics Department Working Papers, No. 1636, OECD Publishing,
Paris, https://doi.org/10.1787/816b1b06-en.
Marcel M., Tokman M., Valdés R., Benavides P., 2001. Balance Estructural: la Base de la Nueva Regla de
Política Fiscal Chilena. Economía chilena, Vol. 4, Nº. 3, 2001, pags. 5-27. 4.
McKay A., Reis R., 2016. The role of automatic stabilizers in the US business cycle. Econometri-ca 84 (1),
141-194.
Muller P., Price R., 1984. Structural Budget Deficits and Fiscal Stance, OECD Economics Department
Working Papers, No. 15, OECD Publishing, Paris, https://doi.org/10.1787/050184860264.
OECD (2019), "Budgetary Governance in Practice: Chile", in Budgeting and Public Expenditures in OECD
Countries 2019, OECD Publishing, Paris, https://doi.org/10.1787/1891f943-en.
OECD (2019). Pensions at a glance 2019: OECD and G20 indicators. OECD Publishing, Paris.
Ramey, G., V. Ramey (1995). Cross-Country Evidence on the Link between Volatility and Growth. American
Economic Review 85 (1995), 1138–1151.
Perrotti D., 2021. Growth Cycles in Argentina: The Recent Behavior. Económica, 67, 023.
https://doi.org/10.24215/18521649e023
Price R., Dang T., Botev J., 2015. Adjusting fiscal balances for the business cycle: New tax and
expenditure elasticity estimates for OECD countries, OECD Economics Department Working Papers, No.
1275, OECD Publishing, Paris, https://doi.org/10.1787/5jrp1g3282d7-en.
Rapetti, M. Fondo Fiscal Contracíclico: una herramienta para el desarrollo. Publicaciones Cippec.
https://www.cippec.org/publicacion/fondo-anticiclico/
Ramey G., Ramey V., 1995. Cross Country Evidence on the Link between Volatility and Growth. American
Economic Review, 85, 1138-1151.
Rofman R., 2020. La movilidad en el sistema previsional argentino. Documento de políticas públicas #227.
Programa de protección social. CIPPEC.
Spilimbergo A., Symansky S., Blanchard O., Cottarelli C., 2008. Fiscal policy for the crisis, in R. W. Kolb (ed.),
Lessons from the financial crisis. New York: Wiley, 587-594.
Talvi E., Vegh C., 2005. Tax base variability and procyclicality of fiscal policy. Journal of Devel-opment
Economics 78 (1), 156.190.
Van den Noord P., 2000. The Size and Role of Automatic Fiscal Stabilizers in the 1990s and Beyond,
OECD Economics Department Working Papers, No. 230, OECD Publishing, Paris,
https://doi.org/10.1787/816628410134.
Vegh C., Lederman D., Bennet F.R., 2017, Leaning against the Wind: Fiscal Policy in Latin America and the
Caribbean in a Historical Perspective. LAC Semiannual Report (April), Washington, DC: World Bank. Doi:
10.1596/978-1-4648-1094-7.
Vegh C., Vuletin G., 2012. How is Tax Policy Conducted over the Business Cycle? NBER Working Paper No.
17753
Vegh C., Vuletin G., 2015. How is tax policy conducted over the business cycle?, American Economic
Journal: Economic Policy, Vol 7, pp. 327-370.
Woo J., 2009. Why do more polarized countries run more pro-cyclical Fiscal policy? Review of Economics
and Statistics 91 (4), 850-870.
World Bank (2009). Pensions in crisis: Europe and Central Asia regional policy note. World Bank,
Washington, DC.
Wibbels E., 2006. Dependency Revisited: InternationalMarkets, Business Cycles, and Social Spending in
the Developing World. International Organization 60(2): 433–68.
Zettelmeyer J., Hollar I. V., 2008. Fiscal Positions in Latin America: Have They Really Improved?, IMF
Working Papers, 2008(137), A001. Retrieved Feb 4, 2023, from
https://www.elibrary.imf.org/view/journals/001/2008/137/article-A001-en.xml
Annex A. Methodology to capture procyclicality
The purpose of the analysis is to see how the fiscal authority is conducting fiscal policy over the business
cycle. Accordingly, the initial step is to have historical data both on GDP and fiscal accounts. In
particular, the methodology consists in the following:
1. Define any fiscal instrument/variable we want to analyze (e.g. government spending)
2. Put together the historical series of both the fiscal instrument and the GDP, in real terms.
3. Get the trend series of both variables by applying HP filter or similar methods in order to
decompose the cyclical and the trend components of the series.
4. The difference between the original series and its trend is the cyclical component of each
variable. For instance, if real GDP (at any given point in time) is above its (HP) trend, the
economy is in “good times” and the cyclical component would be positive (see Example Figure
below).
Cyclical component i: Original Series i – HP Trend i
5. Once we have the cycle of both series (fiscal instrument and GDP), we calculate the correlation
of both cyclical components.
6. If the correlation is positive, then we define the fiscal instrument as procyclical as it increases
in good times and decreases in bad times, and vice versa.
Figure 26. Country A Real GDP vs HP Filter Trend Figure 27. Country A GDP Cycle (Real GDP-HP
Trend)
35 1.5
1.0
30
0.5
25
0.0
20
-0.5
15
-1.0
10 -1.5
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
Source: IMF Database Source: IMF Database
Annex B. The cyclicality of personnel spending in subnational governments: 24 different
realities
Sub-national governments (SNGs) bear most of the public spending on salaries in Argentina, as they are
accountable for providing the main public services. The type of federalism in Argentina, characterized by
a strong original empowerment of SNGs, has shaped most of the multiple fiscal reforms over time.
Following a long period of little decentralization and virtually no vertical fiscal imbalance (SNGs collected
and spent close to 30 percent of total public revenue and spending, respectively), Argentina decentralized
a large part of its public service provision through the last quarter of the last century. Therefore, since the
1990s, almost 45 percent of total spending has been executed by SNGs, while they collect no more than
22 percent of total revenue. Namely, provinces are now (and for the past three decades) accountable for
the provision of public education (except for higher education), public health care and security, as the
three main responsibilities. Consequently, almost ¾ of the total public wage bill in Argentina is currently
executed by SNGs.
SNGs have conducted a strongly procyclical public wage bill policy in recent decades, both by procyclical
“price” (wages) and “quantities” (staff) policies. In general terms, this is true for most of the provinces.
However, as is usually the case when referring to Argentine provinces, there are relevant heterogeneities
worth pointing out. For instance, three provinces stand out for being the less procyclical (or acyclical)
quite far from the rest, broadly speaking (Chubut, San Luis, Neuquén) (Figure 28). In aggregate terms,
both “price” (wages) and “quantities” (number of staff) effects have been strongly procyclical (correlation
coefficients above 0.60, statistically significant). However, while price effect appears more procyclical
across the board, the staffing policies (quantities effect) have been quite different between provinces. For
instance, provinces such as San Luis, Córdoba and Santa Fe, differ from the rest, by having conducted
personnel hiring policies with barely any (or even negative in the case of San Luis) link to the business
cycle.
Figure 28. Decomposition of SNG Wage Bill Figure 29. Public wage bill cyclicality over time and
cyclicality 1990-2021 (number of public service per province (1990-2021; 1990-2000; 2001-2021)
staff and average wage)
1,00
0,80
0,60
0,40
0,20
0,00
-0,20
-0,40
Santa Cruz
CABA
Tucuman
Santa Fe
Misiones
Entre Rios
Salta
Catamarca
Chubut
Neuquen
S. Del Estero
Jujuy
T. Del Fuego
Corrientes
TOTAL
San Luis
La Pampa
San Juan
La Rioja
Rio Negro
Mendoza
Formosa
Cordoba
Chaco
Buenos Aires
Neuquen
Misiones
Tucuman
Catamarca
Jujuy
Corrientes
Chubut
TOTAL
Santa Cruz
S. Del Estero
T. Del Fuego
La Rioja
Rio Negro
San Luis
San Juan
Salta
La Pampa
Santa Fe
Chaco
Buenos Aires
CABA
Entre Rios
Formosa
Mendoza
Cordoba
Wage Bill Staff Avg Wage 1990-2021 1990-2000 2001-2021
Source: Ministry of Economy. Notes: the cyclical components of GDP and public spending have been estimated using the Hodrick-Prescott
filter. Public spending is considered in real terms, deflated by CPI.
Moreover, public spending on wages became more procyclical in recent decades for most of the
provinces, with only a few exceptions. Regardless of the level of cyclicality, 19 provinces increased their
levels of cyclical correlation between public spending on wages and GDP in the past decades. As
mentioned before, the unprecedented expansion of the public wage bill during the 2000s (specifically
between 2004 and 2015) may have contributed significantly to these dynamics. There are only a few cases,
where the public wage bill policies became less procyclical than in the 1990s. Only Misiones, the City of
Buenos Aires, and La Pampa managed to mildly reduce their cyclical correlation between their public wage
bill and GDP.
Annex C. Sources of Data
Annex: Data Sources
Public Spending by Country and GDP Growth Rates:
The data on public spending by country and GDP growth rates were sourced from the
International Monetary Fund's World Economic Outlook (IMF WEO) database. These figures
provide a comprehensive overview of fiscal dynamics across various nations.
Wage Bill Public Spending per Country:
Information on wage bill public spending per country was obtained from the Government Finance
Statistics (GFS) of both the International Monetary Fund (IMF) and the Organisation for Economic
Co-operation and Development (OECD) statistics.
Pensions Public Spending per Country:
The data on pensions public spending per country was collected from two primary sources: the
Organisation for Economic Co-operation and Development (OECD) statistics and the Economic
Commission for Latin America and the Caribbean (CEPAL).
Tax Collection per Country:
The data on tax collection per country is derived from the Organisation for Economic Co-
operation and Development (OECD) statistics. This dataset offers a comprehensive perspective
on tax revenue across different nations, facilitating international comparisons based on general
government statistics.
International Comparison using General Government Statistics:
All international comparisons consider general government statistics whenever available. This
approach ensures consistency and comparability across countries, contributing to a robust cross-
country analysis.
Tax Rates:
Tax rate data was sourced from the database developed by Vegh et al. in 2017.
Argentina-Specific Data:
For public spending data specific to Argentina, information was gathered from the Oficina
Nacional de Presupuesto (National Budget Office) and the Cuenta de Inversion (Investment
Account). Tax revenue data for Argentina was sourced from the Secretaria de Ingresos Publicos
(Secretary of Public Revenues) and the AFIP database, providing a detailed breakdown of tax
collections in the country.
Annex D. Additional figures and tables
Source: Ministry of Economy and Orlando Ferreres.
Figure 31. Public Employees 1990-2020
10
12
14
16
0
2
4
6
8
Kazakhstan
Azerbaijan
Singapore
India
Indonesia
Mexico
Figure 30. Public Wage bill per country (as percent of GDP – Year 2021)
Armenia
Note: Data refers to 2020/2021, except for Argentina (year 2005-2017-2022).
Japan
Korea
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Thailand
Colombia
Peru
Ireland
Albania
Chile
Source: GFS IMG, Ministry of Economy of Argentina.
Mongolia
Moldova
Germany
ARG 2005
Switzerland
Netherlands
Nacion Total (presup)
Turkey
New Zealand
8,4
Argentina 2005
Paraguay
Empleados Públicos
United States
Kosovo
United Kingdom
Uruguay
Russia
Slovak Rep.
Czech Rep.
Ukraine
Provincias
Australia
Italy
Bulgaria
Lithuania
Serbia
Israel
Belarus
Municipios
Poland
Spain
Bosnia & H.
Hungary
Portugal
Latvia
Austria
ARG 2022
El Salvador
Romania
10,9
Argentina 2022
Estonia
Slovenia
Greece
Croatia
Belgium
Finland
France
Canada
ARG 2017
Sweden
Costa Rica
13,2
Argentina 2017
Brazil
Iceland
Norway
South Africa
Jordan
Denmark
As percent of total government spending
10
15
20
25
30
35
40
45
50
0
5
Iceland
Korea
Australia
Millions
Canada
Israel
Netherlands
New Zealand
Norway
United Kingdom
Ireland
Sweden
Denmark
Estonia
Latvia
Slovak Republic
Hungary
United States
OECD
Lithuania
Czech Republic
Luxembourg
Belgium
Turkey
Finland
As percent of total government spending
Germany
Slovenia
France
Japan
Poland
Austria
Source: Ministry of Economy
Spain
Portugal
Italy
Greece
Percent of GDP (RHS)
El salvador
Mexico
Peru
Colombia
Source: OECD stats and Ministry of Economy of Argentina
Costa Rica
LATAM
Chile
Uruguay
Argentina
Brasil (*)
0
2
4
6
8
10
12
14
16
18
percent of GDP
Figure 32. Public spending on pensions (as percent of GDP – 2021 or closest year)
Figure 33. Number of Benefits under the ARG National pensions system (2000-2020)
Source: World Bank data, Ministry of Economy.
rate and the GDP cycle
Figure 36. Correlation between changes in VAT
rate and the GDP cycle
Figure 34. Correlation between changes in CIT
-0.8
-0.6
-0.4
-0.2
0.2
0.4
0.6
0.8
-0.8
-0.6
-0.4
-0.2
-1
0.2
0.4
0.6
0.8
0
1
-1
0
1
Lithuania
Latvia Azerbaijan
Peru Portugal
Lithuania Latvia
Mexico Ecuador
Spain United Kingdom
Austria
Russia
United States
Venezuela
Australia
Uruguay Colombia
Japan Chile
Greece Russia
Botswana Canada
New Zealand Spain
Austria Uruguay
South Africa
France
Costa Rica
Ecuador
Germany
Portugal Italy
Honduras Brazil
Belgium Japan
Colombia Mexico
Chile Belgium
Denmark Honduras
France
South Africa
Peru
Zambia
Korea
Costa Rica Malta
Azerbaijan Botswana
Italy Philippines
Sweden New Zealand
Canada Greece
Malta Venezuela
Germany Norway
Denmark
United Kingdom
Turkey
Norway
Bolivia
Argentina Sweden
Turkey Argentina
Philippines Zambia
rate and the GDP cycle
Figure 35. Correlation between changes in PIT
Index and the GDP cycle
Figure 37. Correlation between changes in Tax
-0.8
-0.6
-0.4
-0.2
0.2
0.4
0.6
0.8
-0.8
-0.6
-0.4
-0.2
-1
0.2
0.4
0.6
0.8
0
1
-1
0
1
Peru Chile
Spain Korea
Italy Russia
Mexico Spain
Portugal
Korea Costa Rica
Latvia Italy
Chile Denmark
Honduras
South Africa
United Kingdom
Australia Azerbaijan
Austria Austria
Portugal Colombia
Japan
Denmark
Belgium
United Kingdom Australia
Honduras Turkey
Malta Peru
Greece
Belgium
France
United States United States
Greece Germany
Uruguay Malta
South Africa
Japan
Botswana
Colombia Norway
Brazil Mexico
France Brazil
Latvia
Philippines
Bolivia
Germany Philippines
Costa Rica Ecuador
Norway Venezuela
Canada
Bolivia
Uruguay
Canada Argentina
Lithuania New Zealand
Argentina Sweden
Lithuania
New Zealand
Sweden