Fiscal consolidations in Latin America and the Caribbean: Do inequality, corruption, and informality matter?

Jalles, João Tovar; Pessino, Carola; Calderon, Ana Cristina Working Paper Fiscal consolidations in Latin America and the Caribbean: Do inequality, corruption, and informality matter? IDB Working Paper Series, No. IDB-WP-1668 Provided in Cooperation with: Inter-American Development Bank (IDB), Washington, DC Suggested Citation: Jalles, João Tovar; Pessino, Carola; Calderon, Ana Cristina (2025) : Fiscal consolidations in Latin America and the Caribbean: Do inequality, corruption, and informality matter?, IDB Working Paper Series, No. IDB-WP-1668, Inter-American Development Bank (IDB), Washington, DC, https://doi.org/10.18235/0013447 This Version is available at: https://hdl.handle.net/10419/315949 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ WORKING PAPER No IDB-WP-1668 Fiscal Consolidations in Latin America and the Caribbean Do Inequality, Corruption, and Informality Matter? João Tovar Jalles Carola Pessino Ana Cristina Calderon Inter-American Development Bank Institutions for Development Sector Fiscal Management Division March 2025 Fiscal Consolidations in Latin America and the Caribbean Do Inequality, Corruption, and Informality Matter? João Tovar Jalles Carola Pessino Ana Cristina Calderon Inter-American Development Bank Institutions for Development Sector Fiscal Management Division March 2025 Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Tovar Jalles, João. Fiscal consolidations in Latin America: do inequality, informality and corruption matter? / João Tovar Jalles, Carola Pessino, Ana Cristina Calderon. p. cm. — (IDB Working Paper Series ; 1668) Includes bibliographical references. 1. Photovoltaic power generation-Latin America. 2. Photovoltaic power generation-Caribbean Area. 3. Solar energy-Latin America. 4. Solar energy-Caribbean Area. I. Pessino, Carola. II. Calderón, Ana Cristina. III. Inter-American Development Bank. Fiscal Management Division. IV. Title. V. Series. IDB-WP-1668 JEL Codes: C23, E21, E62, H5, H62 Keywords: fiscal adjustments, filtering, panel data, binary choice models, local projection, fiscal multipliers, nonlinearities, corruption, shadow economy http://www.iadb.org Copyright © 2025 Inter-American Development Bank ("IDB"). This work is subject to a Creative Commons license CC BY 3.0 IGO (https://creativecommons.org/licenses/by/3.0/igo/legalcode). The terms and conditions indicated in the URL link must be met and the respective recognition must be granted to the IDB. Further to section 8 of the above license, any mediation relating to disputes arising under such license shall be conducted in accordance with the WIPO Mediation Rules. Any dispute related to the use of the works of the IDB that cannot be settled amicably shall be submitted to arbitration pursuant to the United Nations Commission on International Trade Law (UNCITRAL) rules. The use of the IDB's name for any purpose other than for attribution and the use of the IDB's logo shall be subject to a separate written license agreement between the IDB and the user and is not authorized as part of this license. Note that the URL link includes terms and conditions that are an integral part of this license. The opinions expressed in this work are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent. 1 Abstract*†‡* Widening income disparities, higher corruption, and increased informality in many emerging market and developing economies (EMDEs)—all with pressing and mounting fiscal problems—have rekindled interest in the empirical analysis of the key factors determining the occurrence of fiscal consolidations. Using discrete choice models, this paper examines the drivers of fiscal consolidation episodes in a sample of 148 EMDEs between 1980 and 2019, with a focus on Latin American and Caribbean countries. Inequality does not seem to drive consolidations—which are more likely during good economic times—while more informality increases the probability of their occurrence and corruption decreases it. In turn, when examining the drivers of successful consolidations, larger income inequality acts as a boost, while informality is a hinderance. In fact, while the size of the public investment multiplier in Latin America and the Caribbean is larger than in other regions, when informality is high, the multiplier effect is reduced to a much lower and insignificant magnitude. Results are robust to several sensitivity and robustness tests. * João Tovar Jalles: University of Lisbon–Lisbon School of Economics and Management (ISEG); Research in Economics and Mathematics and Research Unit on Complexity and Economics, Universidade de Lisboa–ISEG; Economics for Policy, Universidade Nova de Lisboa–Nova School of Business and Economics and IPAG Business School. Email: [email protected]. † Carola Pessino: Inter-American Development Bank. Email: cpessi[email protected]. ‡ Ana Cristina Calderon: Inter: Inter-American Development Bank. Email: [email protected]. * The authors thank Andrea Luzuriaga for helpful comments and suggestions. The opinions expressed herein are those of the authors and not necessarily those of their employers. Any remaining errors are the authors’ sole responsibility. 2 1. Introduction The COVID-19 pandemic has left a large dent in the government budgets across the world. During 2020, governments had no choice but to increase public spending to fight the pandemic at a time when shrinking economic activity depressed their falling revenues. Consequently, public debt rose by 5 percentage points (pp) on average in 2020, to 46 percent of GDP. In 2021, higher-than-projected inflation reduced public debt-to-GDP ratios in advanced economies (AEs) and emerging markets (EMs) (IMF, 2022). This reprieve is likely to be short lived. As inflation rises, government bonds become less attractive to investors, and the costs of borrowing rises. There are a considerable number of low-income developing countries (LIDCs) that are in debt distress and would need to embark on fiscal consolidation in the foreseeable future, particularly if debt relief is not forthcoming (Clements et al., 2021). In addition, in the context of EMs and LIDCs, the International Monetary Fund (IMF) recently estimated that scarring from the COVID-19 pandemic will increase the already sizable financing needed to achieve the Sustainable Development Goals (Benedek et al., 2021). Given these developments, the issue of fiscal consolidation will remain pertinent in the foreseeable future for both advanced economies and emerging markets, while being of greater urgency in EMs and LIDCs because of depleting (or a lack of) fiscal space. Fiscal retrenchment to get public finances back on the sustainability track and address fiscal solvency concerns is therefore the path to follow. International institutions such as the IMF project that a large share of countries will pursue fiscal consolidation in the coming years. Such projections are only realistic, however, to the extent that current economic and political conditions are sufficiently conducive to fiscal adjustment. There is considerable literature on fiscal consolidations in AEs—reflecting greater availability of data—focused on the type of fiscal adjustment (spending rather than tax-based) that is durable and more growth friendly. By contrast, studies on EMs and LIDCs are limited, in particular those on the Latin American and Caribbean (LAC) region. At the same time, many developing countries suffer from higher-than-average levels of inequality and informality and poor institutional quality (high corruption), all of which can compromise fiscal consolidations' ability to take off and/or succeed.4 This is 4 Some of these issues are being investigated in European countries that underwent consolidation after the global financial crisis of 2008 and failed to achieve it because of these issues (Pappa, Sajedi, and Vella, 2015). 3 particularly evident in the LAC region, where, despite nearly all countries experiencing a significant decline in income inequality since the early 2000s, structural challenges such as weak labor markets, limited social mobility, and persistent disparities in access to education and health care continue to hinder equitable economic growth.5 In fact, more generally, the region has faced significant fiscal challenges over the years, with various countries in the region grappling with high levels of public debt, persistent fiscal deficits, and macroeconomic instability (Eyzaguirre and Santos, 2018). These challenges have posed obstacles to sustainable economic growth, social development, and financial stability. Some key factors contributing to these fiscal challenges include macroeconomic volatility,6 revenue dependence,7 informal economy,8 public expenditure pressures,9 and political and governance challenges.10 The purpose of this empirical paper is fourfold. First, it identifies and characterizes in a novel way fiscal consolidations in a large sample of emerging markets and developing economies (EMDEs). Second, it empirically studies the main determinants of fiscal consolidation. Third, it adds previously neglected dimensions of inequality, corruption, and informality. Finally, it looks at whether drivers are different depending on the compositional nature of the consolidation program; their degree of success is also inspected. The paper seeks answers to the following more detailed set of questions. What are the key stylized facts characterizing fiscal consolidations in LAC countries? What are the key macroeconomic considerations that induce them to implement fiscal consolidations? Is it high debt, slowing growth, or worsening terms of trade? To what 5 See, for example, López Calva and Lustig (2010), Gasparini, Cruces, and Tornarolli (2011), Gasparini and Lustig (2011), Azevedo, Inchauste, and Sanfelice (2013), Székely and Mendoza (2015), Gasparini, Cruces, and Tornarolli (2016), and Székely and Mendoza (2016). 6 LAC economies have experienced high levels of macroeconomic volatility, including fluctuations in commodity prices, exchange rates, and interest rates. These volatile conditions can disrupt fiscal planning and make it challenging to maintain fiscal stability (Céspedes, Chang, and Velasco, 2017). 7 Many countries in the region rely heavily on a few key sources of revenue, such as commodity exports or specific industries. This dependence exposes governments to significant revenue fluctuations, making it difficult to sustain stable fiscal policies and mitigate fiscal risks (Talvi and Végh, 2005). 8 A sizable informal economy exists in many LAC countries, leading to tax evasion, low revenue collection, and limited fiscal space. Informality poses challenges for effective fiscal management because it undermines tax compliance, hampers revenue mobilization efforts, and limits the reach of social protection programs (Melguizo and Scartascini, 2015). 9 The region faces various pressures on public expenditure, including demands for social programs, education, health care, and infrastructure development. Meeting these demands while maintaining fiscal sustainability requires careful expenditure prioritization, efficiency improvements, and effective public administration (Cetrángolo and Lema, 2017). 10 Political instability, corruption, and weak governance have affected fiscal management in several LAC countries. Lack of political consensus, short-term policy focus driven by electoral cycles, and inadequate institutions can hinder effective fiscal consolidation efforts (Levy-Yeyati and Panizza, 2011). 4 extent do inequality, corruption, and informality matter? Are there meaningful differences in terms of the composition of fiscal adjustment (i.e., in terms of whether it is expenditure based or tax based)? What makes a consolidation succeed? To answer these questions, we rely on a new dataset constructed using the first difference of new estimates of the cyclically adjusted primary balance (CAPB) obtained by means of the Hamilton (2018) filter. This is done for a large sample of 148 EMDEs between 1980 and 2019. We find that the duration of a fiscal episode is higher in AEs than in LAC and that the initial fiscal conditions prevailing just before a given episode seem to have had an impact on the size of subsequent fiscal efforts. In addition, the typical fiscal consolidation episode is of short duration (i.e., 2–3 years) and involved relatively modest gains. Then, relying on binary choice models, we find that a consolidation is more likely to take place in “good times”: when growth is high, countries experience positive terms of trade shocks, and inflation is low. High debt remains a significant determinant of consolidation because LAC countries have limited access to financial markets compared with AEs. Inequality in the region does not seem to drive consolidations, while more informality and less corruption increase the probability of their occurrence. More-corrupt countries seem less inclined to carry out expenditure-based fiscal adjustments. The opposite is true for those with larger informal sectors. In fact, while the size of LAC’s public investment multiplier is larger than in other country groups, when informality is high the multiplier effect gets reduced to a much lower and insignificant magnitude. Results are robust to several sensitivity and robustness tests, including, inter alia, alternative consolidation measures that use forecast errors, narrative approaches, and the use of other estimators. The remainder of the paper is organized as follows. Section 2 reviews the relevant literature. Section 3 discusses the empirical methodology and Section 4 presents the data together with key stylized facts. Section 5 discusses the empirical results and Section 6 concludes and elaborates on policy implications. 2. Literature Review This paper focuses on the conditions that may lead countries to embark on fiscal consolidation, the success of fiscal consolidation, and to what extent inequality, corruption, and informality have an impact on such an adjustment process. Hence, the paper relates to four strands of literature: (i) the determinants of fiscal consolidations, (ii) the relationship between inequality and fiscal consolidations, (iii) 5 the relationship between corruption and fiscal consolidations, and (iv) the relationship between informality and fiscal consolidations. Most of the existing literature is sufficiently general and, consequently, does not specifically factor in or address idiosyncrasies (or problems) in the LAC region that might hinder the success of a fiscal consolidation program. Fiscal consolidations—which involve reducing budget deficits and stabilizing public debt—are crucial for sustainable economic growth and financial stability. Understanding the determinants of fiscal consolidations can provide insights for policymakers to design effective consolidation strategies. The literature examining the determinants of fiscal consolidations focuses on the factors that influence the success or failure of such consolidation efforts, including macroeconomic factors, political factors, and structural factors. First is the set of macroeconomic factors. First and foremost, economic growth plays a significant role in determining the success of fiscal consolidations (Alesina and Ardagna, 2010; Fatás and Mihov, 2013). Higher economic growth can lead to increased tax revenues and reduced spending on unemployment benefits, thereby easing the fiscal adjustment process. Studies have shown that fiscal consolidations implemented during periods of economic expansion tend to be more successful than those implemented during recessions. However, the relationship between growth and consolidation is complex because consolidation measures themselves can impact economic growth. Therefore, policymakers need to strike a balance between fiscal discipline and supporting economic activity. Second, interest rates have implications for the cost of public debt, which is a crucial consideration in fiscal consolidations (Giavazzi and Pagano, 1990; Perotti, 1996). Higher interest rates can increase debt servicing costs, making fiscal adjustments more challenging. Conversely, lower interest rates can create more favorable conditions for consolidations by reducing the burden of interest payments. However, the relationship between interest rates and fiscal consolidations is influenced by factors such as monetary policy, inflation expectations, and market perceptions of a country's creditworthiness. Third, inflation can impact fiscal consolidations through its effects on public debt dynamics and economic growth (Alesina and Perotti, 1996; Alesina, Ardagna, and Trebbi, 2006). Higher inflation erodes the real value of public debt (denominated in local currency), making it easier to achieve debt reduction 12 1.5 times the sample standard deviation (or equal to one sample standard deviation, on average, over two years). As there is no single, agreed-upon definition in the literature, and being aware of best practices reviewed above, we adopt a middleground approach in defining CAPB-to-GDP change thresholds for the determination of fiscal consolidation episodes. We opt for the Alesina and Perotti (1997) approach, under which a fiscal consolidation episode is defined as a minimum annual improvement in the CAPB-to-GDP ratio of 0.5 pp over two consecutive years.13 Another relevant issue is the CAPB measure of choice. CAPB data can be obtained either via a publicly available source (e.g., the IMF's World Economic Outlook [WEO] database) or computed using a filtering approach (by decomposing GDP and government revenues into their cyclical and trend components). In relation to this, there is no clear consensus in the literature regarding the “optimal” way to estimate potential output. According to Borio, Disyatat, and Juselius (2017), past studies have applied (i) univariate statistical approaches, usually consisting of filtering out the trend component from the cyclical one, or (ii) structural approaches, deriving the estimates directly from a theoretical model. Aware of the shortcomings of using either of the two approaches14 and the disadvantage of not maximizing the total number of observations in our panel database when using the WEO CAPB,15 we apply a filtering technique. Once the potential output (and, consequently, the output gap) is obtained, we use it to compute a new measure of the CAPB. Reflecting the fact that the elasticity of government revenues (REV) to output growth is close to one while primary expenditure (PEXP) is largely inelastic to growth (i.e., we assume the same as Girouard and André, 2005), we multiply government revenues by the factor [1/(1+OG/100)] to calculate REV!"# (adjusted government revenues), with OG being the output gap obtained via the Hodrick-Prescott(HP) or Hamilton filters.16 Mathematically, we have: 13 The start year of a fiscal consolidation episode is, therefore, the year in which there is a minimum annual improvement in the CAPB-to-GDP ratio of 0.5 pp, if there is also a minimum annual improvement in the CAPB-to-GDP ratio of 0.5 pp in the following year. Accordingly, the end year of a fiscal consolidation episode is the last year (in a sequence of years) with a minimum annual improvement in the CAPB-to-GDP ratio of 0.5 pp, after which the annual CAPB-to-GDP either improves by less than 0.5 pp or worsens (i.e., decreases). 14 Statistical methods suffer from the end-point problem—that is, they are extremely sensitive to the addition of new data and to real-time data revisions. Structural models, on the other hand, may be difficult to implement consistently in cross-sectional environments and rely on the imposition of predetermined assumptions. 15 The IMF does not have an official method for computing potential output. While the most common IMF approach relies on a production function, assumptions vary greatly across countries. 16 For a discussion of these approaches, see, for example, Hamilton (2018). 13 CAPB =REV!"# −PEXP. (1) Composition-wise, a fiscal consolidation episode is defined as an expenditurebased episode if the ratio of the cumulative fall in the primary expenditure-to-GDP ratio (defined as the sum of all annual changes in the primary expenditure-to-GDP ratio within the episode) to the cumulative adjustment (defined as the sum of all annual changes in the CAPB-to-GDP ratio) is larger than (or equal to) 2/3 in absolute value. If the sum of all annual changes in the primary expenditure-to-GDP ratio within a fiscal consolidation episode is positive, the episode is classified as a taxbased consolidation. All remaining cases are classified as “mixed” consolidation episodes. Succinctly, a fiscal consolidation episode is defined as expenditure based when |∆&'(&)_+,&| |∆)-&.)_+,&|≥ 2/3/and ∆PEXPC_GDP < 0, with CAPBC_GDP and PEXPC_GDP denoting cumulative CAPB and primary expenditure (in percent of GDP) within a given episode. Conversely, a fiscal consolidation episode is defined as tax based when ∆PEXPC_GDP/ ≥ 0. It follows that any episodes that do not satisfy the criteria set forth above are classified as mixed fiscal consolidation episodes. Weighing the aforementioned factors, our preferred specification for the definition of fiscal consolidation episodes in this paper will employ a CAPB change threshold of 0.5 pp over two consecutive years (Alesina and Perotti, 1997), with the CAPB data obtained based on the Hamilton filter (yielding our “Hamilton-based” criterion for fiscal consolidations). This specification comes with several advantages, namely: (i) allowing us to maximize our sample size by identifying more than 1,000 (450) fiscal consolidation years (episodes) across 185 countries (37 AEs and 148 EMDEs) between 1980 and 2019 (for our purposes only EMDEs will be used in the empirics), (ii) ensuring broad consistency and comparability with the alreadyestablished literature on fiscal consolidations (most of which relies on CAPB metrics), and (iii) prioritizing relatively durable fiscal consolidations as opposed to one-off shocks to CAPB levels (given that the CAPB threshold criterion is applied to CAPBto-GDP changes over two years as opposed to a single year). 14 Table 1. Fiscal Consolidation Episodes in LAC by Criteria Country WEO FC episode HP FC episode Hamilton FC episode Argentina 2002–2004, 2018–2019 2002–2004, 2017–2019 2002–2004, 2017– 2019 Belize none 2004–2008, 2010– 2011, 2016–2018 2001–2002, 2004– 2006, 2016–2018 Bolivia none 1994–1995, 2003–2006 1989–1990, 1994– 1995, 1999–2000, 2003–2006 Brazil 1998–1999, 2016–2017 1998–1999 none Chile 2004–2006, 2010–2011 2003–2006, 2010–2011 2003–2006, 2010– 2011 Colombia 2011–2012 1990–1991, 2000–2001, 2012–2012 1990–1991, 2000– 2001, 2011–2012 Costa Rica none 2005–2007 2005–2007 Dominican Republic 2004–2005 2004–2005, 2013–2015 2004–2005, 2013– 2015 Ecuador none 1999–2000, 2010–2011, 2017–2018 1999–2000, 2010– 2011, 2017–2018 El Salvador 2003–2004, 2016–2017 2003–2004, 2016–2017 1993–1994, 2003– 2004, 2016–2017 Guyana 2009–2010, 2012–2013 2007–2008 2008–2009 Honduras none 1991–1992, 1994–1995, 2004–2005, 2010– 2011, 2014–2015 1991–1992, 1994– 1995, 2004–2005, 2010–2011, 2014– 2015 Jamaica none 1998–2000, 2012–2013 1998–2000, 2012– 2013 Mexico 1999–2001, 2016–2017 2000–2001, 2015–2017 Panama 1999–2000, 2005–2007 1995–1996, 1999–2000, 2005–2007, 2015–2016 1995–1996, 1999– 2000, 2005–2007 Paraguay none 1985–1986, 1989–1990, 1993–1994, 2003–2004, 2010–2011 1985–1986, 1989– 1990, 1997–1998, 2003–2004, 2010– 2011 Peru 2006–2007, 2010–2011, 2018–2019 2006–2007, 2010– 2011, 2018–2019 1999–2001, 2016– 2017 Suriname 2006–2007, 2016–2018 1992–1996, 2003– 2007, 2016–2017 1994–1995, 2006– 2007, 2017–2018 Trinidad and Tobago none 1989–1991, 2018–2019 1989–1991, 2007– 2008, 2018–2019 Uruguay 2001–2003 2001–2004 2002–2003 Venezuela none 1989–1990, 1995–1996, 1999–2000, 2002– 2005 1989–1990, 1995– 1996, 1999–2000, 2002–2005, 2019 Source: Authors' elaboration. Note: "FC" = fiscal consolidation. 3.2. Empirical Approach Our aim is to explore whether income inequality, corruption, and informality (ICI for short) affect the likelihood of consolidating public finances while controlling for other variables identified in the literature as affecting the implementation of fiscal consolidations. Hence, our main dependent variable of interest is the occurrence of a fiscal consolidation episode. To capture this, we rely on a fiscal consolidation (FC) 15 dummy for country i in year t that takes the value of 1 if country i is in a fiscal consolidation episode (as defined above) in year t (0 otherwise). Based on this binary characterization, our baseline empirical exercise consists of estimating logistic regressions to assess the likelihood of a given country experiencing a fiscal consolidation year, with the counterfactual being the opposite of this (i.e., not experiencing a fiscal consolidation year). We estimate the following model: Prob(FC = 1|𝑋) = 𝛷(𝐼𝐶𝐼′𝜶 + 𝑋′𝜷), (2) where α,/𝜷 are vectors of the parameters to be estimated; 𝐼𝐶𝐼 is a proxy for inequality, corruption, or informality; 𝑋 is a vector of control variables; and 𝛷(⋅) is the logistic function. Our list of control variables includes the real GDP growth rate, the rate of inflation, and the debt-to-GDP ratio. These variables are sourced from the April 2022 IMF WEO vintage. We also add trade openness (proxied by the value of imports and exports in percent of GDP), percent changes in the terms of trade, and percent changes in the real effective exchange rate from the World Bank’s World Development Indicators (WDI) database as controls. The model associated with equation (2) can be written as: 𝐹𝐶/0 = 𝜶𝐼𝐶𝐼/012 + 𝜷𝑋/012 + 𝜀/0, (3) where, again, the FC variable takes the value 1 if a fiscal consolidation episode takes place (i.e., we allow for multi-year fiscal consolidation episodes): 𝐹𝐶it = 1 if a fiscal consolidation takes place in country i during year t (0 otherwise); i = 1, …, N; t is the year; and 𝜀/0/is an independent and identically distributed (i.i.d.) error term. In this case, each set of estimates 𝜶/Cand 𝜷/ Dis interpreted as showing (the rise or fall in) the likelihood of a fiscal consolidation year being experienced by country i . 4. Data and Stylized Facts Macroeconomic data come from the IMF’s April 2022 WEO database. These include real GDP, the budget-balance-to-GDP ratio, CAPB (percent of GDP), total government revenues (percent of GDP), primary government expenditures (percent of GDP), the CPI inflation rate (percent), and government gross debt (percent of GDP). Additional information on trade openness (value of exports and imports, percent of 16 GDP), changes in the terms of trade, and changes in the real exchange rate come from the World Bank's WDI database as mentioned above. Inequality proxies are given by the Gini index, which goes from 0 to 100, with the latter denoting more unequal income distribution. Several sources are used. The first is Solt’s (2009, 2020) Standardized World Income Inequality Database (SWIID), which covers 177 countries from 1960 to the present and includes both gross (market) and net (dispensable) Gini. The SWIID dataset combines income information from the United Nations World Income Database (UNWIDER) and the Luxembourg Income Study (LIS). SWIID provides comparable standardized Gini coefficients to measure income inequality based on estimates of market (pre-taxes and transfers) and net (post-taxes and transfers) income inequality. This thus allows the comparison of income disparities before and after redistribution by taxation and transfers over time. Note that taxes determine households’ disposable income available for consumption and thus influence the income distribution. However, disposable income does not consider indirect taxes. This creates a limitation when only disposable income is considered. As a result, we look at both pre-tax-and-transfers and post-tax-and-transfers Gini indices.17 According to Poterba (2007), using the latter mitigates the reverse causality problem because post-tax-and-transfers vary “mechanically” and “economically” with the fiscal system whereas the pre-tax-andtransfers measure varies solely through the endogenous responses of labor supply or the general equilibrium effect on factor prices. We use both the market and net income Gini indices, with high coverage across countries and over time, in the estimations.18 The second is Milanovic’s All Ginis dataset, which represents a compilation and adaptation of income or consumption Gini coefficients (calculated across households or household per capita, on gross or net basis) retrieved from nine sources and ends in 2017. Out of these nine sources the following are used due to 17 The Gini indicators based on disposable income cover the total market income received by all household members (gross earnings, self-employment income, and capital income), plus the current cash transfers they receive, less income and wealth taxes, social security contributions, and current transfers that they pay to other households. 18 The imputation methodology to standardize observations collected from various sources makes these series subject to measurement uncertainty (Jenkins, 2015). Indeed, there are some concerns about the reliability of SWIID's imputed estimates particularly in data-poor regions (Jenkins, 2015). That said, Ferreira, Lustig, and Teles (2015) compared eight inequality datasets and conclude that “although there is much agreement across these databases, there is also a non-trivial share of country/year cells for which substantial discrepancies exist” and that “the methodological differences […] often appear to be driven by a fundamental trade-off between a wish for broader coverage on the one hand, and for greater comparability on the other.” 17 sample limitations for the region under scrutiny (LAC): POVCAL, SEDLAC, World Income Distribution (WID), and WIDER. For corruption two sources are used. The first is the World Bank's Country Policy and Institutional Assessment (CPIA) transparency, accountability, and corruption variable in the public sector (rating 1–6, with 6 denoting a better score). The second is the Corruption Perceptions Index (CPI) from Transparency International, which ranks 180 countries around the world by their perceived levels of public sector corruption (results are given on a scale of 0 [highly corrupt] to 100 [very clean]). For both indices their respective mirrors are computed so that the reading is as follows: the larger the value, the more corrupt the country in a given year. For informality two sources are used. The first is the Elgin et al. (2021) dataset on informality, which includes several proxies: Dynamic Stochastic General Equilibrium (DSGE) model–based estimates of informal output (percent GDP) and multiple indicators multiple causes (MIMIC) model–based estimates of informal output (percent GDP), self-employment (percent total employment), and informal employment (percent total employment).19 The second source for informality is from the International Labor Organization (ILO) and includes the total informality rate (percent). To maximize coverage (particularly within LAC), the preferred sources for our three dimensions are the SWIID for inequality, the CPI for corruption, and Elgin et al. (2021) DSGE-based estimates of informal output (percent GDP) for informality. That said, sensitivity is done using alternative proxies of the different concepts. Table 2 provides comparative summary statistics between AEs and Latin America. We observe that the number of fiscal episodes is significantly lower when we consider the WEO-based consolidation criterion compared with the HP-based or Hamilton-based criteria. This is particularly salient in the case of Latin America, where there were only 61 WEO-based consolidations versus 192 HP-based and 183 Hamiltonbased. Also, the average adjustment in percent of GDP is lower when the former criteria is used (less than 2 percent of GDP when using WEO-based and more than 2 percent of GDP with HPor Hamilton-based). So, using the WEO underestimates the average size of consolidations due to data limitations. In addition, the duration of a 19 MIMIC is essentially a structural model where the shadow economy is estimated from a system of equations composed of economic and institutional variables. The DSGE variable comes from a deterministic DGE model initially proposed by Elgin and Oztunali (2012). 18 fiscal episode is higher for AEs than the duration observed for Latin America. In fact, while the reported duration is, on average, three years for AEs, the duration of fiscal episodes for Latin America is slightly lower at 2.5 years. The full set of episodes by Latin American country (with corresponding years) is provided in Table A1 in the Appendix. For instance, two methods—WEO and HP—that determine fiscal consolidation episodes on the basis of the change in the CAPB essentially coincide in identifying, for instance, the fiscal contractions of Argentina in 2002–2004 or Brazil in 1998–1999. Table 2. Summary Statistics of Fiscal Consolidations by CAPB Measure and per Criterion of Economic Development Advanced Economies Total # years of FC episodes Avg. # FC episodes Avg. size of adjustment in FC episode (% GDP) Avg. duration of FC episode (years) WEObased 191 1.65 1.72 3.21 HP-based 267 2.25 1.94 3.14 Hamilton -based 276 2.35 1.87 3.05 Latin America Total # years of FC episodes Avg. # FC episodes Avg. size of adjustment in FC episode (% GDP) Avg. duration of FC episode (years) WEObased 61 1.59 1.99 2.34 HP-based 192 2.06 2.53 2.66 Hamilton -based 183 2.33 2.17 2.64 Source: Authors' calculations. Notes: "FC" = fiscal consolidation. Average size of adjustment in FC episode is the cumulative adjustment (defined as the sum of all annual changes in the CAPB-to-GDP ratio within the episode) within a given episode, divided by the duration (total number of years) of the episode. Average duration of FC episode (years) is the sum of all years during which a country has consolidated within a given episode. Figure 1 reports the distribution of changes in CAPB (percent of GDP) in LAC. In the rest of the text we will refrain from presenting WEO-based evidence and focus instead on HPand Hamilton-based consolidations. 19 Figure 1. Changes in CAPB (percent of GDP) during Fiscal Consolidations in Latin America and the Caribbean HP-based consolidations Hamilton-based consolidations Source: Authors' calculations. Note: The figure plots both the histogram and Kernel densities for each criterion (HP and Hamilton) for LAC. As far as the characteristics of fiscal consolidation episodes, initial fiscal conditions prevailing just before a given consolidation episode seem to have had an impact on the size of subsequent fiscal efforts (Figure 2). The lower the CAPB, the larger the ensuing fiscal consolidation. This may reflect that large budget deficits made it more necessary to consolidate and, at the same time, raised public awareness of the extent of the fiscal imbalance problem, making it easier to act. Figure 2. Initial Fiscal Imbalance and Subsequent Adjustments to Fiscal Consolidations in Latin America and the Caribbean HP-based consolidations Hamilton-based consolidations Source: Authors' calculations. 0.1 .2 .3 Density -10 -5 0 5 10 dcapbhp 0.1 .2 .3 Density -10 -5 0 5 10 dcapbham -15 -10 -5 0 5 Budget position in the year before consolidation 0 5 10 15 Improvement in budget position 95% CI Fitted values CAPB_hp_potpy with linear fit -LAC Initial Fiscal positions and subsequent adjustment -15 -10 -5 0 5 Budget position in the year before consolidation 0 5 10 15 Improvement in budget position 95% CI Fitted values CAPB_ham_potpy with linear fit -LAC Initial Fiscal positions and subsequent adjustment 20 Most of the fiscal consolidation episodes in LAC were of short duration (see the top two panels in Figure 3, but with some exceptions—see Table A1 in the Appendix) and involved relatively modest gains (see bottom panels of Figure 3). However, there were a number of large efforts, amounting to improvements of more than 10 percent of GDP. It is also possible to observe that, in general, sizable fiscal consolidations lasted for longer periods and smaller consolidations had a shorter duration (Figure 4). Figure 3. Duration and Size of Consolidation Episodes in Latin America and the Caribbean HP-based consolidations Hamilton-based consolidations Source: Authors' calculations. Notes: Top panels: budget position measured by the CAPB (percent of potential GDP). Bottom panels: improvement measured during the consolidation years of the identified episode. 050 100 Number of episodes 2 3 4 5 LAC The distribution of episodes by duration, in years 050 100 150 Number of episodes 2 3 4 6 LAC The distribution of episodes by duration, in years 010 20 30 Number of episodes 010 20 30 40 Improvement in budget position during the episode LAC The Distribution of episodes by the size of consolidation 010 20 30 Number of episodes 010 20 30 Improvement in budget position during the episode LAC The Distribution of episodes by the size of consolidation 21 Figure 4. Relationship between Duration and Size of Consolidation Episodes in Latin America and the Caribbean HP-based consolidations Hamilton-based consolidations Source: Authors' calculations. Notes: Budget position measured by the CAPB (percent of potential GDP). Improvement measured during the consolidation years of the identified episode. 5. Empirical Results This section presents the main findings on the role of inequality, corruption, and informality as determinants of fiscal consolidations, based on the specifications discussed in Section 3. Based on the discussion in Sections 3 and 4 and for the regression specifications mentioned below, we rely on an unbalanced panel database at an annual frequency. Summary statistics for the variables included in the regressions are shown in Table A1 in the Appendix. 5.1. Determinants of Fiscal Consolidations We begin with the estimation of logistic regression (3) to explore the main determinants of fiscal consolidations in EMDEs and Latin America (both excluding the Caribbean countries) in Tables 3 and 4 respectively.20 Both show the results when we: (i) initially exclude the ICI variables (column 1) and (ii) individually add each ICI variable one at a time. For the EMDE sample, we observe that the worse the level of informality, the more likely it is for a country to consolidate, but the result is surrounded by great uncertainty (compare columns 8–10 in Table 3 with column 11). Corruption comes out 20 Results are based on the Hamilton-based CAPB criterion for identifying fiscal consolidation episodes. The WEO-based results are available upon request. The HP-based results are partially shown in the Appendix and discussed in Section 5.3 on robustness and sensitivity. 010 20 30 40 Improvement in budget position during the episode 2 3 4 5 Duration of the episode (years) LAC Duration and size of consolidation 010 20 30 Improvement in budget position during the episode 2 3 4 5 6 Duration of the episode (years) LAC Duration and size of consolidation 28 Table 6a. Determinants of Successful Fiscal Consolidations, EMDEs Specification (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Success definition V1 V1 V1 V1 V2 V2 V2 V2 V3 V3 V3 V3 Real GDP growth (t-1) 0.180*** 0.248*** 0.161*** 0.726*** 0.120*** 0.201** 0.148*** 0.136 0.110*** 0.193** 0.105*** 0.099 (0.042) (0.089) (0.040) (0.252) (0.043) (0.082) (0.047) (0.255) (0.035) (0.083) (0.034) (0.086) Debt ratio (t-1) 0.000 0.014 0.006+ 0.005 0.007** 0.019* 0.010** 0.122* 0.006 0.014 0.009+ -0.006 (0.005) (0.010) (0.004) (0.018) (0.004) (0.011) (0.004) (0.066) (0.005) (0.012) (0.006) (0.016) Inflation (t-1) 0.066** -0.014 0.042* 0.031 0.024 -0.187* 0.007 -0.252 0.059+ 0.033 0.048+ 0.001 (0.030) (0.040) (0.022) (0.069) (0.026) (0.096) (0.025) (0.203) (0.042) (0.046) (0.034) (0.076) Trade openness (t-1) -0.001 -0.001 0.004 0.003 -0.003 -0.007 -0.001 0.030 -0.003 0.008 0.001 0.006 (0.004) (0.004) (0.004) (0.010) (0.003) (0.006) (0.004) (0.022) (0.004) (0.006) (0.005) (0.012) Terms of trade growth (t-1) -0.006 -0.082+ -0.024 -0.159 -0.005 0.009 -0.023 -0.185+ -0.005 0.042 0.011 -0.086+ (0.025) (0.053) (0.022) (0.148) (0.024) (0.054) (0.024) (0.124) (0.022) (0.053) (0.017) (0.059) SWIID Market Gini (t-1) 0.030* 0.012 0.011 (0.015) (0.024) (0.019) Corruption Perceptions Index (CPI) (inverted) (t-1) -0.010 0.033+ -0.006 (0.018) (0.022) (0.020) MIMIC-based informality (t-1) -0.006 -0.000 0.013 (0.011) (0.011) (0.015) Informal employment (t-1) -0.068** 0.021 -0.028* (0.031) (0.033) (0.016) Observations 224 92 254 34 224 92 254 34 224 92 254 34 McFadden Pseudo-R2 0.096 0.069 0.074 0.295 0.052 0.095 0.060 0.340 0.067 0.088 0.061 0.079 Notes: The dependent variable is a dummy taking the value of 1 in a successful consolidation year (0 otherwise), defined using the Hamilton-based criterion and following one of the definitions for “success” described in the main text. Standard errors in parentheses. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively. 29 Table 6b. Determinants of Successful Fiscal Consolidations, Latin America Specification (1) (2) (3) (4) Success definition V1 V2 V3 V3 Real GDP growth (t-1) 0.476+ 0.253 0.170** 0.170 (0.329) (0.264) (0.069) (0.256) Debt ratio (t-1) 0.012 0.022** 0.051*** -0.023 (0.013) (0.011) (0.017) (0.052) Inflation (t-1) 0.269+ 0.161 0.363** 0.864+ (0.183) (0.242) (0.158) (0.566) Trade openness (t-1) 0.038 0.020 0.025 -0.084 (0.028) (0.027) (0.038) (0.075) Terms of trade growth (t-1) -0.047+ -0.069 -0.059 -0.188 (0.033) (0.085) (0.082) (0.253) SEDLAC Gini (t-1) 0.144** 0.056 -0.317 (0.062) (0.142) (0.561) MIMIC-based informality (t-1) 0.073 (0.110) Observations 32 32 32 18 McFadden Pseudo-R2 0.306 0.218 0.404 0.574 Notes: The dependent variable is a dummy taking the value of 1 in a successful consolidation year (0 otherwise), defined using the Hamilton-based criterion and following one of the definitions for “success” described in the main text. Standard errors in parentheses. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively. Box 1. Non-Linear Fiscal Multiplier Effects Depending on Inequality, Corruption, and Informality Here we estimate the conditional response of real GDP following a government spending shock, conditioned on the level of inequality, corruption, and informality. We use Jordá’s (2005) non-linear local projections to obtain impulse responses that are allowed to vary according to a continuous function 𝐹(𝑧/0),/as follows: 𝑦/,045 − 𝑦/,012 = 𝛼/+ 𝜏/+ [𝛽5 6× τ × 𝑆/,0] + [𝛽5 7× (1 − τ) × 𝑆/,0] + θ𝑋/,0 + 𝜀/,0, (B1) where y is the real GDP (in logs) and 𝑆/,0 is the government spending shock; τ/is/an/indicator/function/that/takes/the/value = 1/𝑖𝑓/𝑧80 c c c <𝑚𝑒𝑑9:;<=> such that 𝑧8 h is a country-specific average of an indicator of inequality, corruption, or informality and 𝑚𝑒𝑑9:;<=>/is the corresponding sample median value.a We are interested particularly in the 𝛽5 7 coefficient—that is, when the state is of high inequality, high corruption, or high informality. Equation B1 is estimated using Ordinary Least Squares (OLS) with Driscoll-Kraay (1998) robust standard errors (clustered at the country level). To run equation B1, the big issue relates to identifying 𝑆/,0. The observed heterogeneity in the estimates of fiscal spending multipliers reflects to a great extent the general challenges associated with the identification of exogenous shocks in public spending. To date, several approaches have been employed to address this issue.b We will rely on the approach that uses government spending forecast errors employed by Auerbach and Gorodnichenko (2012, 2013), Abiad, Furceri, and Topalova (2016), Furceri and Li (2017), and Colombo et al. (2022). These authors argue that this methodology overcomes the obstacles that often confound the causal estimation of the effect of fiscal policy on economic performance.c The measure of government spending shocks is the difference between the actual real value of public investment or consumption and the corresponding value expected by analysts as of October of the same year. More formally, and similarly to Colombo et al. (2022), the identification 30 of unexpected fiscal policy shocks using forecast errors in government spending is done in two steps. First, we compute the t-period forecast error for public spending for country i , 𝐹𝐸/,0|012 = ∆𝑙𝑛𝐺/,0 − ∆𝑙𝑛𝐺/,0|012, where ∆𝑙𝑛𝐺/,0 defines the actual government spending growth rate and ∆𝑙𝑛𝐺/,0|012 is the t-1 IMF forecast of ∆𝑙𝑛𝐺/,0 made at time t-1 for year t .d This approach also solves, by construction, the problem of “fiscal foresight,” which arises when agents react to anticipated rather than realized shocks (see, e.g., Forni and Gambetti, 2010; Ben Zeev and Pappa, 2015). Second, the corresponding public spending component is regressed on lags of the corresponding forecast error (plus country and time effects) to purge predictable components and by taking the residual of this projection (normalized by the average share in percent of GDP of the public spending variable) as the fiscal shock. Relative to Colombo et al. (2022), we have expanded the range of forecast data (and, hence, the spending shocks) from 1995–2015 as in their paper to the 1980–2019 period.e Results given by the impulse response are shown in Figure B1, which plots the unconditional responses (dashed blue line) together with the conditional ones (solid black line) with associated confidence bands. We observe a positive and insignificant unconditional output response to government consumption shocks, but a positive and significant unconditional response to government investment shocks. The size of the public investment multiplier in Latin America is larger than in other groups of countries, such as OECD economies. When informality is high the public investment multiplier effect gets reduced to a much lower and insignificant magnitude. The effect on output from a public consumption shock when informality is high remains statistically insignificant. These results are broadly in line with those obtained by Colombo et al. (2022) using a shorter time span but for a larger sample of countries. The underestimation of fiscal multipliers seems more pronounced in countries with a higher level of informality, which for Pappa, Sajedi, and Vella (2015) can be measured through tax evasion.f The same applies for high inequality. For corruption, the output impact of a public consumption shock is not statistically different from zero and is not distinguishable from the baseline. In the case of a public investment shock the effect remains positive and significant but, again, statistically not different from the unconditional result. Figure B1. Conditional Response of Real GDP Growth to Government Spending Shock: The Role of Inequality, Corruption, and Informality (in percent) -2 0 2 4 6 0 1 2 3 4 year public cons. shock in high informality -1 0 1 2 0 1 2 3 4 year public inv. shock in high informality 31 Notes : The x axis is in years; t=0 is the year of the fiscal spending shock (i.e., an unanticipated 10 percent increase); t=1 is the first year of impact. Solid black lines denote the response to a fiscal spending shock when inequality, corruption, and informality are high; the dark gray area denotes 90 percent confidence bands and the light gray area denotes 68 percent confidence bands, based on standard errors clustered at country level. The dashed blue line denotes the unconditional result together with 90 percent confidence bands depicted as dashed red lines. a Due to a lack of sufficient continuous observations for inequality, corruption, and informality proxies, the approach discussed by Auerbach and Gorodnichenko (2012, 2013) to estimate the local projection in the context of a smooth transition autoregressive (STAR) model developed by Granger and Teräsvirta (1993) is not possible. b These include identification based on the assumption that government spending does not respond to macroeconomic shocks in the same period in a Structural Vector Autoregressive (SVAR) framework (Blanchard and Perotti, 2002; Ilzetzki, Mendoza, and Végh, 2013); the natural experiment approach exploiting variation in the military spending buildups (Ramey and Shapiro, 1998; Ramey, 2011a, 2011b; Ramey and Zubairy, 2018) and official lending (Kraay, 2012, 2014) as sources of exogenous fluctuations in government spending. c According to Ramey (2016), shocks should be exogenous with respect to other current and lagged endogenous variables, they should be uncorrelated with other exogenous shocks, and they should represent unanticipated movements in exogenous variables. The forecast-error approach is arguably able to address all three. d According to An et al. (2018), IMF fiscal forecasts are accurate and preferred compared to private-sector forecasts. e Lack of a large comprehensive and cross-country comparable dataset on tax revenue forecasts prevent us from exploring the other side of the budget, namely tax hikes or shocks—that is, revenue-based consolidations. f Basile, Girardi, and Miele (2016) exploit Italian data on tax evasion and unreported income to investigate the response of the formal and informal sectors to public expenditure shocks. They find that in Italy fiscal expansions cause a reduction in the share of unreported income. 5.3. Robustness and Sensitivity Several sensitivity and robustness exercises were conducted, beginning with the sensitivity exercises to address omitted variable bias. Results are shown in Table A3 in the Appendix. Testing for the “original sin” (i.e., countries’ obligation to borrow in foreign currency while being able to pay in domestic currency), we obtain (based on -2 0 2 4 6 0 1 2 3 4 year public cons. shock in high inequality -1 0 1 2 0 1 2 3 4 year public inv. shock in high inequality -5 0 5 10 0 1 2 3 4 year public cons. shock in high corruption 0.5 11.5 2 0 1 2 3 4 year public inv. shock in high corruption 32 the specification under column 2 of Table 3) a statistically insignificant coefficient for (lagged) interest payments. We also tested for the “urgency to consolidate” (i.e., the distance between a country’s debt-stabilizing primary balance and CAPB in percent of GDP), obtaining a positive and statistically significant coefficient for the EMDE sample but not for Latin America. Finally, we explored the role of (lagged) international reserves minus gold retrieved from the World Bank WDI database and obtained a statistically insignificant coefficient. We found that in Latin America, the larger the stock of reserves, the less likely it is for a country to consolidate. Tables A4 and A5 in the Appendix repeat Tables 3 and 4 for the EMDE and LAC samples, respectively, by replacing the binary dependent variable with a continuous variable given by the yearly change in the Hamilton-based CAPB (percent GDP). This will capture all structurally adjusted improvements in the overall balance, from the smallest to the largest—in contrast with the dummy variable logistic approach. OLS regressions suggest that results are broadly similar with a key difference: for both the EMDE and LAC samples, the larger the level of informality, the larger the ensuing improvement in the CAPB. In terms of robustness, we start by using an alternative CAPB-based definition to identify the consolidation years—namely, the HP. Results presented in Table A6 in the Appendix, while slightly weaker in statistical terms, are generally consistent with those in Tables 3 and 4 for the EMDE and LAC samples, respectively. Next, we construct alternative measures of consolidation. The first builds on Gali and Perotti (2003), Golinelli and Momigliano (2009), Alesina and Ardagna (2010), Corsetti, Meier, and Müller (2012), and Auerbach and Gorodnichenko (2012). We estimate a fiscal policy rule of the form: ∆𝐶𝐴𝑃𝐵/0 = α/+ 𝜇0+ β∆𝐶𝐴𝑃𝐵/012 + γGAP/0 + δDEBT/012 + 𝜀/0, (4) where α/ stands for unobserved country effects, 𝜇0 captures time effects, ∆𝐶𝐴𝑃𝐵/0stands for the yearly change in the CAPB as a percent of GDP using the Hamilton-based approach, GAP/0 is the output gap obtained using the Hamiltonbased approach, and DEBT/0 stands for the debt-to-GDP ratio. To account for the contemporaneous correlation between the output gap and the dependent variable, we estimated equation (4) by means of an instrumental variable technique where the output gap is instrumented by each own lag and the first lag of real GDP growth rate. Given that for Latin America N=13<T=40, the estimates are less susceptible to the 33 so-called Nickell bias (Nickell, 1981). According to Gali and Perotti (2003), the response of the dependent variable to output gap reflects the systematic discretionary fiscal policy component, while 𝜀/0 is the random component. This reflects the nonsystematic fiscal policy response or the unanticipated fiscal policy shocks, which are independent across countries. The fiscal consolidation shock that is used as an alternative to the binary dummy variable used as baseline is defined as follows: 𝐷/0 = 1 if 𝜀/0 > 0 and 𝐷/0 = 0 if 𝜀/0 ≤ 0 (i.e., it has positive value during times of fiscal consolidation, which implies that the CAPB increases). Table 7.a shows the results. The block of fixed controls remains qualitatively similar, and as before we still get that, for Latin America, inequality does not seem to drive consolidations, while more informality increases the probability of their occurrence. 34 Table 7a. Panel Analysis: Fiscal Consolidation Based on Two-Step Regression, EMDEs versus Latin America Specification (1) (2) (3) (4) (5) (6) (7) (8) Sample EMDEs Latin America Real GDP growth (t-1) 0.141*** 0.111** 0.147*** 0.129* 0.222*** 0.122 0.342*** 0.384** (0.025) (0.054) (0.026) (0.073) (0.074) (0.238) (0.110) (0.164) Debt ratio (t-1) 0.010*** 0.011+ 0.012*** 0.033*** 0.022+ 0.075** 0.059*** 0.085*** (0.002) (0.007) (0.003) (0.013) (0.016) (0.034) (0.022) (0.026) Inflation (t-1) 0.005 0.035 0.010 -0.104** 0.030 -0.196+ 0.066 -0.070+ (0.009) (0.025) (0.009) (0.044) (0.025) (0.124) (0.061) (0.048) Trade openness (t-1) 0.001 0.011*** 0.002 -0.003 -0.007** 0.032 -0.002 -0.016+ (0.002) (0.004) (0.002) (0.007) (0.003) (0.025) (0.013) (0.010) Terms of trade growth (t-1) 0.002 0.012 0.004 0.025 0.007 0.069* -0.005 0.016 (0.006) (0.014) (0.006) (0.031) (0.014) (0.041) (0.026) (0.025) SWIID Market Gini (t-1) 0.022** (0.009) Corruption Perceptions Index (CPI) (inverted) (t-1) 0.010 -0.005 (0.015) (0.029) DSGE-based informality (t-1) 0.005 (0.007) SEDLAC Gini (t-1) -0.043 (0.035) MIMIC-based informality (t-1) 0.009 (0.008) Informal employment (t-1) -0.003 0.061*** (0.010) (0.019) Observations 1,275 514 1,451 214 130 73 115 89 McFadden Pseudo-R2 0.055 0.059 0.057 0.088 0.126 0.278 0.217 0.272 Notes: The dependent variable is a dummy taking the value of 1 in a fiscal consolidation year (0 otherwise). Standard errors in parentheses. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively. 35 The second consolidation measure relies on the alternative identification approach that uses forecast errors. Auerbach and Gorodnichenko (2013) use forecast errors to examine how the fiscal multiplier varies with the business cycle in OECD economies. This measure of government shocks is computed as the difference between the actual public spending and the public spending expected previously by professional forecasters. Using forecast error–based shocks, Abiad, Furceri, and Topalova (2016) identify the causal impact of higher public investment on output, private investment, unemployment, and public debt ratios. Abiad, Furceri, and Topalova (2016) argue that this methodology overcomes the obstacles that often confound the causal estimation of the effect of fiscal policy on economic performance. The Auerbach and Gorodnichenko (AG) approach was also utilized by Furceri and Li (2017), Honda, Miyamoto, and Taniguchi (2020), and Miyamoto et al. (2020). This methodology has the advantage of overcoming the problem of “fiscal foresight” (see Forni and Gambetti, 2010; Leeper, Richter, and Walker, 2012; Leeper, Walker, and Yang, 2013; and Ben Zeev and Pappa, 2017). More formally, we construct a measure of contractionary fiscal policy shocks as an unexpected improvement in the budget balance as a share of GDP. These “fiscal policy” forecast errors are the differences between the actual budget balances reported in the following year and the one-year-ahead forecasts in the autumn edition of the IMF’s WEO report. Mathematically, we have 𝐹𝐸/,0|012 =𝐵𝐵/,0 −𝐵𝐵?/,0|012, where 𝐵𝐵/,0 defines the actual budget balance in percent of GDP and 𝐵𝐵?/,0|012 is the IMF forecast of 𝐵𝐵/,0 made at time t-1. We use budget balance forecasts from 2003 until 2019. Positive forecast errors mean that the actual budget balance was larger than the forecast, suggesting an unanticipated fiscal retrenchment. These take the value 1 and 0 otherwise. Table 7.b shows the results. For EMDEs we get slightly conflicting results on the influence of informality on the probability of occurrence of a consolidation, while for Latin America the positive and significant result keeps surfacing as before. 36 Table 7.b. Panel Analysis: Fiscal Consolidation Based on Forecast-Error Approach, EMDEs versus Latin America Specification (1) (2) (3) (4) (5) (6) (7) (8) Sample EMDEs Latin America Real GDP growth (t-1) 0.071*** -0.008 0.053** 0.078 0.128+ -0.188* 0.123 0.094** (0.026) (0.026) (0.021) (0.056) (0.087) (0.111) (0.127) (0.037) Debt ratio (t-1) 0.007 -0.007* 0.005 -0.008 0.024 -0.007 0.010 -0.012* (0.005) (0.004) (0.004) (0.007) (0.025) (0.012) (0.023) (0.007) Inflation (t-1) 0.010 -0.013 0.016 0.051 0.154** -0.070 0.242*** -0.004 (0.011) (0.025) (0.013) (0.065) (0.072) (0.058) (0.043) (0.076) Trade openness (t-1) -0.003 0.002 0.000 0.011+ 0.005 0.001 -0.023 0.006 (0.003) (0.002) (0.003) (0.007) (0.018) (0.011) (0.017) (0.010) Terms of trade growth (t-1) 0.004 0.014 0.006 -0.044** -0.075* -0.005 -0.096*** -0.048+ (0.010) (0.020) (0.009) (0.021) (0.044) (0.051) (0.034) (0.033) SWIID Market Gini (t-1) 0.016 (0.016) Corruption Perceptions Index (CPI) (inverted) (t-1) -0.008 -0.004 (0.008) (0.021) DSGE-based informality (t-1) 0.022** (0.009) SEDLAC Gini (t-1) 0.098 (0.092) MIMIC-based informality (t-1) 0.033*** (0.008) Informal employment (t-1) -0.013* 0.002 (0.007) (0.010) Observations 1,275 514 1,451 214 130 73 115 89 McFadden Pseudo-R2 0.020 0.009 0.017 0.048 0.157 0.037 0.183 0.050 Notes: The dependent variable is a dummy taking the value of 1 in a fiscal consolidation year (0 otherwise). Standard errors in parentheses. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively. 37 The third and final consolidation measure relies on the narrative approach dataset put together by David and Leigh (2018). This dataset of fiscal consolidations for 14 LAC economies during 1989–2016 includes the size of the discretionary changes in taxes and expenditures jointly and separately. This allows us to perform a composition examination of the drivers similarly to the CAPB baseline approach we did earlier. Table 7.c shows the results; they are generally weaker, but this could be the result of both a different sample composition and a different time span under scrutiny. 44 Azevedo, J. P., G. Inchauste, and V. Sanfelice. 2013. Decomposing the Recent Inequality Decline in Latin America. World Bank Policy Research Working Paper No. 6715. Washington, DC: World Bank. Basile, R., A. Girardi, and V. Miele. 2016. Tax evasion, public expenditure and the business cycle in Italy. Economic Modelling 52: 40–49. Benedek, J., I. Kinga, I. Török, A. Temerdek, and J. Holobâcă. 2021. Indicator-based assessment of local and regional progress toward the Sustainable Development Goals (SDGs): An integrated approach from Romania . Sustainable Development 29(5): 860–875. Ben Zeev, N., and E. Pappa. 2015. Multipliers of Unexpected Increases in Defense Spending: An Empirical Investigation. Journal of Economic Dynamics and Control 57(C): 205–226. ———. 2017. Chronicle of a War Foretold: The Macroeconomic Effects of Anticipated Defense Spending Shocks. Economic Journal 127(603): 1568–1597. Berg, A., and J. D. Ostry. 2011. Inequality and Unsustainable Growth: Two Sides of the Same Coin? IMF Staff Discussion Note No. 11/08. Washington, DC: International Monetary Fund. Besley, T., and T. Persson. 2011. Pillars of Prosperity: The Political Economics of Development Clusters. Princeton, NJ: Princeton University Press. Blanchard, O., and R. Perotti. 2002. An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output. Quarterly Journal of Economics 117(4): 1329–1368. Borio, C., P. Disyatat, and M. Juselius. 2017. Rethinking potential output: Embedding information about the financial cycle. Oxford Economic Papers 69(3): 655–677. Céspedes, L. F., R. Chang, and A. Velasco. 2017. Financial Intermediation, Exchange Rates, and Unconventional Policies in an Open Economy. Journal of International Economics 108: S76-S86. Cetrángolo, O., and D. Lema. 2017. Inclusive Growth and Fiscal Policies: Challenges for Latin America. United Nations Economic Commission for Latin America and the Caribbean (ECLAC) Working Paper No. 388. Santiago de Chile: Economic Commission for Latin America and the Caribbean. Clements, B., S. Gupta, J. Jalles, and V. Mylonas. 2021. Low-Income Developing Countries Will Surely Need More Debt Relief Down the Line. Center for Global Development blog post. June 23. Washington, DC: Center for Global Development. 45 Cingano, F. 2014. Trends in Income Inequality and Its Impact on Economic Growth. OECD Social, Employment and Migration Working Paper No. 163. Paris: OECD Publishing. Colombo, E., D. Furceri, P. Pizzuto, and P. Tirelli. 2022. Fiscal Multipliers and Informality (No. 2022-2082). Washington, DC: International Monetary Fund. Corsetti, G., A. Meier, and G. J. Müller. 2012. What Determines Government Spending Multipliers? Economic Journal 122(561): 555–579. David, A., and D. Leigh. 2018. A New Action-Based Dataset of Fiscal Consolidation in Latin America and the Caribbean. IMF Working Paper 18/94. Washington, DC: International Monetary Fund. David, A., T. Komatsuzaki, and S. Penknagura. 2020. The Macroeconomic Effects of Structural Reforms in Latin America and the Caribbean. IMF Working Paper 20/195. Washington, DC: International Monetary Fund. Devries, P., J. Guajardo, D. Leigh, and A. Pescatori. 2011. A New Action-Based Dataset of Fiscal Consolidation. IMF Working Paper 11/128. Washington, DC: International Monetary Fund. Drazen, A., and M. Eslava. 2010. Electoral Manipulation Via Voter-Friendly Spending: Theory and Evidence. Journal of Development Economics 92(1): 39–52. Dreher, A., and M. Gassebner. 2013. Greasing the Wheels? The Impact of Regulations and Corruption on Firm Entry. Public Choice 155(3–4): 413–432. Driscoll, J. C., and A. C. Kraay. 1998. Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data. Review of Economics and Statistics 80(4): 549–560. Elgin, C., M. A. Kose, F. Ohnsorge, and S. Yu. 2021. Understanding Informality. CEPR Discussion Paper No. 16497. Washington, DC: Center for Economic and Policy Research. Elgin, C., and O. Oztunali. 2012. Shadow Economies around the World: Model Based Estimates. Bogazici University Department of Economics Working Papers 2012/05. Eyzaguirre, N., and P. Santos. 2018. Tackling Fiscal Challenges in Latin America: Progress and Opportunities. IMF Regional Economic Outlook: Western Hemisphere. Washington, DC: International Monetary Fund. Fatás, A., and I. Mihov. 2013. Fiscal Policy as a Stabilization Tool. Handbook of Macroeconomics 2: 2301–2358. 46 Ferreira, F., N. Lustig, and D. Teles. 2015. Appraising Cross-National Income Inequality Databases: An Introduction. Journal of Economic Inequality 13: 497–526. Forni, M., and L. Gambetti. 2010. Fiscal Foresight and the Effects of Government Spending. CEPR Discussion Paper 049. Washington, DC: Center for Economic and Policy Research. Furceri, D., and B. Li. 2017. The Macroeconomic (and Distributional) Effects of Public Investment in Developing Economies. IMF Working Paper 17/217. Washington, DC: International Monetary Fund. Gali, J., and R. Perotti. 2003. Fiscal Policy and Monetary Integration in Europe. Economic Policy 18(37): 533–572. Gao, S., and J. Shen. 2007. Asymptotic Properties of a Double Penalized Maximum Likelihood Estimator in Logistic Regression. Statistics and Probability Letters 77: 925–930. Gasparini, L., G. Cruces, and L. Tornarolli. 2011. Recent Trends in Income Inequality in Latin America. Economia 10(2): 147–201. ———. 2016. Chronicle of a Deceleration Foretold: Income Inequality in Latin America in the 2010s. Revista de Economía Mundial 43: 2546. Gasparini, L., and N. Lustig. 2011. The Rise and Fall of Income Inequality in Latin America. Working Paper No. 1110. New Orleans: Tulane University. Gates, S. 2001. Empirically Assessing the Causes of Civil War. Working Paper (unpublished). Giavazzi, F., and M. Pagano. 1990. Can Severe Fiscal Contractions Be Expansionary? Tales of Two Small European Countries. NBER Macroeconomics Annual 5: 75– 122. ———. 1996. Non-Keynesian Effects of Fiscal Policy Changes: International Evidence and the Swedish Experience. Swedish Economic Policy Review 3: 67–103. Girouard, N., and C. André. 2005. Measuring Cyclically-Adjusted Budget Balances for the OECD Countries. OECD Economics Department Working Paper No. 434. Paris: OECD Publishing. Golinelli, R., and S. Momigliano. 2009. The Cyclical Reaction of Fiscal Policies in the Euro Area: The Role of Modelling Choices and Data Vintages. Fiscal Studies 30(1): 39–72. Granger, C. W. J., and T. Terasvirta. 1993. Modeling Nonlinear Economic Relationships . New York: Oxford University Press. 47 Guajardo, J., D. Leigh, and A. Pescatori. 2014. Expansionary Austerity: New International Evidence. Journal of the European Economic Association 12(4): 949–968. Guichard, S., M. Kennedy, and E. Wurzel. 2007. What Promotes Fiscal Consolidation: OECD Country Experiences. OECD Economics Department Working Paper No. 553. Paris: OECD Publishing. Hall, R. E., and C. I. Jones. 1999. Why Do Some Countries Produce So Much More Output per Worker Than Others? Quarterly Journal of Economics 114(1): 83–116. Hallerberg, M., and P. Marier. 2004. Executive Authority, the Personal Vote, and Budget Discipline in Latin American and Caribbean Countries. American Journal of Political Science 48(3): 571–587. Hamilton, J. D. 2018. Why You Should Never Use the Hodrick-Prescott Filter. Review of Economics and Statistics 100(5): 831–843. Hilscher, J., A. Raviv, and R. Reis. 2022. Inflating Away the Public Debt? An Empirical Assessment. Review of Financial Studies 35(3): 1553–1595. Honda J., H. Miyamoto, and M. Taniguchi. 2020. Exploring the Output Effect of Fiscal Policy Shocks in Low Income Countries. IMF Working Paper 2020/012. Washington, DC: International Monetary Fund. Ilzetzki, E., E. G. Mendoza, and C. A. Végh. 2013. How Big (Small?) Are Fiscal Multipliers? Journal of Monetary Economics 60(2): 239–254. IMF (International Monetary Fund). 2017. Fiscal Monitor: Tackling Inequality. Washington, DC: International Monetary Fund. ———. 2022. IMF Fiscal Monitor: “Fiscal Policy from Pandemic to War.” Washington, DC: International Monetary Fund. Jenkins, S. P. 2015. World Income Inequality Databases: An Assessment of WIID and SWIID. Journal of Economic Inequality 13(4): 629–671. Jordà, O. 2005. Estimation and Inference of Impulse Responses by Local Projections. American Economic Review 95: 161–182. Jordà, Ò., and A. Taylor. 2016. The Time for Austerity: Estimating the Average Treatment Effect of Fiscal Policy. Economic Journal 126(590): 219–255. Kalemli-Ozcan, S., B. E. Sørensen, and O. Yosha. 2012. Fiscal Policy in a Small Open Economy with Oil Sector and Non-Oil Sector. Journal of the European Economic Association 10(7): 1577–1618. Kaufmann, D., and P. C. Vicente. 2011. Legal Corruption. Economics and Politics 23(2): 195–219. 48 Keen, M., and B. Lockwood. 2010. The Value-Added Tax: Its Causes and Consequences. Journal of Development Economics 92(2): 138–151. King, G., and L. Zeng. 1999a. Logistic Regression in Rare Events Data. Department of Government. Cambridge, MA: Harvard University. ———. 1999b. Estimating Absolute, Relative, and Attributable Risks in Case-Control Studies. Department of Government. Cambridge, MA: Harvard University. ———. 2001. Explaining Rare Events in International Relations. International Organization 55: 693–715. Kopits, G., and S. Symansky. 1998. Fiscal Policy Rules (No. 162). Washington, DC: International Monetary Fund. Kraay, A. 2012. How Large Is the Government Spending Multiplier? Evidence from World Bank Lending. Quarterly Journal of Economics 127(2) 829–887. ———. 2014. Government Spending Multipliers in Developing Countries: Evidence from Lending by Official Creditors. American Economic Journal: Macroeconomics 6(4) 170–208. Lambertini, L., and J. Tavares. 2005. Exchange Rates and Fiscal Adjustments: Evidence from the OECD and Implications for the EMU. Contributions in Macroeconomics 5(1): 1–28. Leeper, E. M., A. W. Richter, and T. B. Walker. 2012. Quantitative Effects of Fiscal Foresight. American Economic Journal: Economic Policy 4(2): 115–144. Leeper, E. M., T. B. Walker, and S. Yang. 2013. Fiscal Foresight and Information Flows. Econometrica 81(3): 1115–1145. Levy-Yeyati, E., and U. Panizza. 2011. The Elusive Costs of Sovereign Defaults. Journal of Development Economics 94(1): 95–105. López Calva, L., and N. Lustig (eds). 2010. Declining Inequality in Latin America: A Decade of Progress? New York/Washington, DC: United Nations Development Programme/Brookings Institution Press. Maloney, W. F., and F. Valencia. 2017. Engineers of Growth? Economic Development and the Role of the Engineering Sector. World Bank Policy Research Working Paper No. 8211. Washington, DC: World Bank. Mati, A., and J. Thornton. 2005. Developments in Domestic Fuel Prices, Taxes, and Subsidies in Selected Countries (unpublished). Fiscal Affairs Department. Washington, DC: International Monetary Fund. Mauro, P. 1998. Corruption and the Composition of Government Expenditure. Journal of Public Economics 69(2): 263–279. 49 Melguizo, A., and C. Scartascini. 2015. Informality, Productivity, and State Support to the Markets. IDB Working Paper No. 573. Washington, DC: Inter-American Development Bank. Miyamoto, H., et al. 2020. Growth Impact of Public Investment and the Role of Infrastructure Governance. In G. Schwartz et al. (eds), Well Spent: How Strong Infrastructure Governance Can End Waste in Public Investment . Washington, DC: International Monetary Fund. Nickell, S. 1981. Biases in Dynamic Models with Fixed Effects. Econometrica 49(6): 1417– 1426. Pappa, E., R. Sajedi, and E. Vella. 2015. Fiscal Consolidation with Tax Evasion and Corruption. Journal of International Economics 96: S56-S75. Perotti, R. 1996. Growth, Income Distribution, and Democracy: What the Data Say. Journal of Economic Growth 1(2): 149–187. ———. 1999. Fiscal Policy in Good Times and Bad. Quarterly Journal of Economics 114(4): 1399–1436. Perry, G., W. F. Maloney, and O. Arias. S. 2007. Informality: Exit and Exclusion . Washington, DC: World Bank. Persson, T., and G. Tabellini. 1994. Is Inequality Harmful for Growth? American Economic Review 84(3): 600–621. Potrafke, N. 2013. Political Cycles and Economic Performance in OECD Countries: Empirical Evidence from 1951–2006. Public Choice 155(3–4): 393–409. Poterba, J. M. 2007. Income Inequality and Income Taxation. Journal of Policy Modeling 29(4): 623–633. Ramey, V. A. 2011a. Can Government Purchases Stimulate the Economy? Journal of Economic Literature 3(49): 673–685. ———. 2011b. Identifying Government Spending Shocks: It’s All in the Timing. Quarterly Journal of Economics 1(126): 51–102. ———. 2016. Macroeconomic Shocks and Their Propagation. NBER Working Paper Series. Working Paper 21978. Cambridge, MA: National Bureau of Economic Research. Ramey, V. A., and M. D. Shapiro. 1998. Costly Capital Reallocation and the Effects of Government Spending. Carnegie-Rochester Conference Series on Public Policy 48: 145–194. 50 Ramey, V. A., and S. Zubairy. 2018. Government Spending Multipliers in Good Times and in Bad: Evidence from US Historical Data. Journal of Political Economy 126(2): 850–901. Romer, C. D., and D. H Romer. 2010. The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks. American Economic Review 100(3): 763–801. Schneider, F., and D. H. Enste. 2000. Shadow Economies: Size, Causes, and Consequences. Journal of Economic Literature 38(1): 77–114. Slemrod, J. 2019. Tax Compliance and Enforcement. Journal of Economic Literature 57(4): 904–54. Solt, F. 2009. Standardizing the World Income Inequality Database. Social Science Quarterly 90(2): 231–242. ———. 2020. Measuring Income Inequality across Countries and over Time: The Standardized World Income Inequality Database. Social Science Quarterly 101(3): 1183–1199. Székely, M., and P. Mendoza. 2015. Is the Decline in Inequality in Latin America Here to Stay? Journal of Human Development and Capabilities 16: 397–419. ———. 2016. Declining Inequality in Latin America: Structural Shift or Temporary Phenomena? Oxford Development Studies 44. Talvi, E., and C. Végh. 2005. Tax Base Variability and Procyclical Fiscal Policy in Developing Countries. Journal of Development Economics 78(1): 156–190. Tommasi, M., and A. Velasco. 1996. Choosing and Reneging: The Stability of Exchange Regimes. NBER Macroeconomics Annual 11: 235–283. Wei, S. J. 2001. Corruption and Economic Development. Pacific Economic Review 6(3): 317–337. 51 Appendix Table A1. Summary Statistics of Regression Variables for the Whole EMDE Sample Variable Observations Mean Standard deviation Minimum Maximum Real GDP growth 2,222 4.11 4.30 -41.88 38.2 Debt ratio 1,953 55.36 37.62 0.07 451.33 Inflation 2,213 6.94 12.86 -72.72 256.10 Change in Real Effective Exchange Rate (REER) 889 -0.40 7.90 -77.93 36.20 Trade openness 2,000 79.46 41.19 6.07 347.99 Terms of trade growth 2,024 0.24 8.93 -97.51 150.28 Gini Disposable 1,564 41.20 7.21 23.4 66.4 Gini Market 1,564 45.55 6.79 22.4 70 gini_SEDLAC 153 51.40 4.55 40.97 59.41 Gini POVCAL 703 40.44 9.11 24.03 65.77 ~cpi_inv 631 62.38 11.96 26 92 sims_informal 133 59.65 21.10 22.47 88.64 DSGE_informal 1,738 35.97 9.77 8.55 66.43 informal_emp 220 62.91 21.33 18.91 99.65 52 Table A2. Panel Analysis Including the Exchange Rate: Hamilton-Based Fiscal Consolidations, EMDEs Specification (1) (2) (3) (4) (5) (6) (7) (8) (9) regressors Real GDP growth (t-1) -0.045* -0.029 -0.022 -0.052 -0.062+ -0.135** -0.141** - 0.072*** - 0.063*** (0.028) (0.031) (0.031) (0.042) (0.043) (0.055) (0.058) (0.024) (0.023) Debt ratio (t-1) -0.006 -0.004 -0.006 -0.016* -0.015+ 0.014 -0.016 -0.002 -0.003 (0.006) (0.006) (0.007) (0.008) (0.009) (0.018) (0.016) (0.006) (0.006) Inflation (t-1) -0.021 - 0.045** -0.036+ -0.033+ -0.022 -0.070 0.067* -0.041* -0.024 (0.021) (0.023) (0.023) (0.021) (0.031) (0.077) (0.037) (0.024) (0.024) REER growth (t-1) - 0.058*** - 0.061** - 0.067*** - 0.104*** - 0.098*** -0.078* -0.075* -0.053** - 0.054*** (0.022) (0.025) (0.024) (0.025) (0.032) (0.040) (0.047) (0.022) (0.021) Trade openness (t-1) 0.008*** 0.013*** 0.012*** 0.011*** 0.012*** 0.002 -0.010 0.011*** 0.010*** (0.003) (0.003) (0.003) (0.004) (0.004) (0.008) (0.011) (0.003) (0.003) Terms of trade growth (t-1) 0.009 0.001 0.001 0.012 0.008 0.141*** 0.005 0.009 0.008 (0.010) (0.010) (0.010) (0.012) (0.011) (0.052) (0.010) (0.010) (0.009) SWIID Market Gini (t-1) -0.037* (0.020) SWIID Disposable Gini (t-1) -0.027 (0.023) POVCAL Gini (t-1) - 0.038** (0.019) World Bank Gini (t-1) -0.014 (0.020) Corruption Perception Index (CPI) (inverted) (t1) -0.001 (0.027) Country Policy and Institutional Assessment (CPIA) (inverted) (t-1) - 0.726+ (0.486) DSGE-based informality (t-1) 0.012 (0.015) MIMIC-based informality (t-1) 0.011 (0.015) Observations 652 534 534 288 211 205 170 606 621 McFadden Pseudo-R2 0.044 0.076 0.072 0.114 0.109 0.130 0.119 0.064 0.054 Notes: The dependent variable is a dummy taking the value of 1 in a fiscal consolidation year (0 otherwise), defined using the Hamilton-based criterion. Standard errors in parentheses clustered at the country level. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively. 53 Table A3. Panel Analysis Addressing Omitted Variable Bias: Hamilton-Based Fiscal Consolidations, EMDEs versus Latin America Specification regressors (1) (2) (3) (4) (5) (6) Sample EMDEs Latin America EMDEs Latin America EMDEs Latin America Real GDP growth (t-1) -0.047*** 0.007 -0.020 0.049 -0.044** 0.024 (0.017) (0.066) (0.022) (0.079) (0.019) (0.077) Debt ratio (t-1) 0.006** 0.014** 0.004+ 0.014+ 0.002 0.009 (0.003) (0.005) (0.003) (0.010) (0.003) (0.007) Inflation (t-1) -0.007 0.028 0.004 0.077* -0.011 0.022 (0.009) (0.031) (0.010) (0.045) (0.011) (0.036) Trade openness (t-1) 0.005* 0.006 0.004* 0.005 0.004+ -0.004 (0.002) (0.006) (0.002) (0.007) (0.002) (0.008) Terms of trade growth (t-1) 0.015* 0.005 0.014* 0.000 0.024*** 0.009 (0.008) (0.009) (0.008) (0.007) (0.008) (0.010) Interest payments (t-1) -0.056 -0.023 (0.050) (0.153) Urgency to consolidate (t-1) 0.044** 0.047 (0.018) (0.046) Log reserves minus gold (t-1) -0.059 -0.329* (0.045) (0.174) Observations 1,595 271 1,554 265 1,461 274 McFadden Pseudo-R2 0.019 0.048 0.022 0.044 0.020 0.071 Notes: The dependent variable is a dummy taking the value of 1 in a fiscal consolidation year (0 otherwise), defined using the Hamilton-based criterion. Standard errors in parentheses clustered at the country level. Constant term omitted. +, *, **, *** denote statistical significance at the 15, 10, 5, and 1 percent levels, respectively.