Public versus private investment multipliers in emerging market and developing economies: Cross-country analysis with a focus on Asia

Jalles, João Tovar; Park, Donghyun; Qureshi, Irfan A. Working Paper Public versus private investment multipliers in emerging market and developing economies: Cross-country analysis with a focus on Asia ADB Economics Working Paper Series, No. 737 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: Jalles, João Tovar; Park, Donghyun; Qureshi, Irfan A. (2024) : Public versus private investment multipliers in emerging market and developing economies: Cross-country analysis with a focus on Asia, ADB Economics Working Paper Series, No. 737, Asian Development Bank (ADB), Manila, https://doi.org/10.22617/WPS240372-2 This Version is available at: https://hdl.handle.net/10419/305383 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/ ASIAN DEVELOPMENT BANK ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org ADB ECONOMICS WORKING PAPER SERIES NO. 737 August 2024 Public versus Private Investment Multipliers in Emerging Market and Developing Economies Cross-Country Analysis with a Focus on Asia This paper introduces new measures of investment shocks based on cyclically adjusted investment data to examine the effect of public and private investment shocks on growth. Results suggest that public investment shocks play a much greater role in boosting economic growth in comparison with private investment shocks. In emerging market and developing economies including in Asia, the growth response to investment shocks is positive and much stronger during recessions compared with economic expansions. Investment shocks also stimulate private investment and private consumption. About the Asian Development Bank ADB is committed to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific, while sustaining its efforts to eradicate extreme poverty. Established in 1966, it is owned by 68 members —49 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. PUBLIC VERSUS PRIVATE INVESTMENT MULTIPLIERS IN EMERGING MARKET AND DEVELOPING ECONOMIES CROSS-COUNTRY ANALYSIS WITH A FOCUS ON ASIA João Tovar Jalles, Donghyun Park, and Irfan Qureshi ASIAN DEVELOPMENT BANK The ADB Economics Working Paper Series presents research in progress to elicit comments and encourage debate on development issues in Asia and the Pacific. The views expressed are those of the authors and do not necessarily reflect the views and policies of ADB or its Board of Governors or the governments they represent. ADB Economics Working Paper Series João Tovar Jalles, Donghyun Park, and Irfan Qureshi No. 737 | August 2024 João Tovar Jalles ([email protected]) is a senior associate professor of economics at the University of Lisbon. Donghyun Park ([email protected]) is an economic advisor at the Economic Research and Development Impact Department, Asian Development Bank (ADB). Irfan Qureshi ([email protected]) is a public management specialist in the Sectors Group of ADB. Public versus Private Investment Multipliers in Emerging Market and Developing Economies: Cross-Country Analysis with a Focus on Asia Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) © 2024 Asian Development Bank 6 ADB Avenue, Mandaluyong City, 1550 Metro Manila, Philippines Tel +63 2 8632 4444; Fax +63 2 8636 2444 www.adb.org Some rights reserved. Published in 2024. ISSN 2313-6537 (print), 2313-6545 (PDF) Publication Stock No. WPS240372-2 DOI: http://dx.doi.org/10.22617/WPS240372-2 The views expressed in this publication are those of the authors and do not necessarily reflect the views and policies ofthe Asian Development Bank (ADB) or its Board of Governors or the governments they represent. ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The mention of specific companies or products of manufacturers does not imply that they are endorsed or recommended by ADB in preference to others of a similar nature that are not mentioned. By making any designation of or reference to a particular territory or geographic area inthis document, ADB does not intend to make any judgments as to the legal or other status of any territory or area. This publication is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) https://creativecommons.org/licenses/by/3.0/igo/. By using the content of this publication, you agree to be bound bytheterms of this license. For attribution, translations, adaptations, and permissions, please read the provisions andterms of use at https://www.adb.org/terms-use#openaccess. This CC license does not apply to non-ADB copyright materials in this publication. If the material is attributed toanother source, please contact the copyright owner or publisher of that source for permission to reproduce it. ADB cannot be held liable for any claims that arise as a result of your use of the material. Please contact [email protected] if you have questions or comments with respect to content, or if you wish toobtain copyright permission for your intended use that does not fall within these terms, or for permission to use theADB logo. Corrigenda to ADB publications may be found at http://www.adb.org/publications/corrigenda. ABSTRACT This paper analyzes the growth impact of public and private investment shocks based on a large sample of emerging and developing countries over the period 1980–2021 with a particular focus on the Asian region. We develop new measures of investment shocks based on cyclically adjusted investment data. Estimations using local projections suggest that public investment shocks play a much greater role in boosting economic growth in comparison with private investment shocks. In emerging market and developing economies (EMDEs) (including in Asia), the growth response to investment shocks is positive and much stronger in recessions relative to economic expansions. Finally, public investment shocks in EMDE and Asian samples crowd-in private investment and private consumption. Keywords: fiscal multipliers, public investment, private investment JEL codes: C33, E22, H30, H50 1. Introduction Public investment is considered one of the key policy levers to foster economic growth, particularly in emerging market and developing economies (EMDEs), as it plays a vital role in developing infrastructure, such as roads and utilities, which is critical for reducing transaction costs and improving efficiency (World Bank 2017). It also addresses the market failure of under-investment in areas where the private sector may hesitate to invest due to high risks or low private returns. Moreover, government investment in health, education, and social services is pivotal for building human capital, a key ingredient for a productive workforce and sustainable economic growth. On the other hand, private investment is driven by profit motive, with efficiency and innovation leading to enhanced productivity and competitiveness. Private investment in new ventures and expansion projects creates jobs and increases incomes, which, in turn, stimulate further consumption and investment. Additionally, private investment plays a significant role in the development of financial markets, improving access to finance and fostering more investments. De Gregorio (1992) has shown its positive impact on economic growth through mechanisms like capital accumulation and efficiency gains. Against this backdrop, it is not surprising that a large economic literature has emerged to evaluate the macroeconomic effects of investment in general and public investment in particular. First, public investment in building and maintaining key infrastructure such as roads, bridges, ports, airports, and energy facilities is essential. This infrastructure reduces transportation costs, improves connectivity, and enhances the overall business environment. For example, the extensive high-speed rail network 2 expansion of the People’s Republic of China (PRC) has not only improved connectivity but also spurred economic activity across the country (Liu, Niu, and Han 2019). Second, public investment in education and healthcare contributes greatly to human capital development. Such investments enhance the quality and productivity of the workforce, contributing to economic growth (Hanushek and Woessmann 2012). Public investments in education played a major role in fostering a highly skilled labor force and innovation in highly successful Asian countries such as the Republic of Korea. Third, public investment benefits technology and innovation. Governments across Asia have recognized the importance of supporting technology-intensive industries through research grants and subsidies (Asian Development Bank [ADB] 2018). This approach helped countries such as Singapore become global technology hubs. Fourth, public investment can strategically address regional disparities. By promoting the development of less-developed areas, it can alleviate poverty, stimulate economic activity, and reduce rural–to–urban migration (World Bank 2016). For instance, India's investment in rural infrastructure and employment generation programs aims to address regional imbalances. Fifth, in an era of climate change and disasters, public investment in disaster preparedness, environmental sustainability, and clean energy infrastructure is essential for long-term sustainable development. Asian economies increasingly acknowledge the importance of investing in renewable energy sources and climate resilience measures (ADB 2017). Sixth, public investment acts as a catalyst for private sector development. Governments can invest in vital infrastructure or provide incentives for private firms, 3 encouraging them to invest, expand, and create jobs (Woo and Kumar 2015). This synergy between public and private investment can be a powerful driver of economic development. Seventh, public investment in social safety nets, healthcare, and education directly impacts poverty reduction by improving living standards and creating opportunities for disadvantaged groups (Kose et al. 2017). Programs like conditional cash transfers in countries like Indonesia and the Philippines have demonstrated the poverty-reduction potential of public investment. Finally, as Asian economies heavily rely on international trade, public investment in trade-related infrastructure, such as ports and logistics facilities, enhances a nation's ability to engage in global trade and attract foreign investment (ADB 2018). Examples include the development of export-processing zones in Bangladesh and the PRC. The empirical estimates of public investment multipliers reported to date, however, differ significantly, depending on the sample of countries and timeline investigated, estimation methodology, incorporation of non-linear effects, and fiscal shock identification strategies (Section 2). The main problem in performing such research rests in the difficulty of identifying exogenous changes in public spending that are uncorrelated with contemporaneous macroeconomic shocks. Considering these caveats, we aim to contribute to this literature in several ways. First, we use updated data on public (and private) investment covering a large sample of 98 EMDEs, including 20 Asian economies, over a long-time span, from 1980 to 2021. Previous generalizations of research that only looked at samples of emerging market economies to “developing countries” were misleading. Our results are thus based on a 4 broader, more heterogeneous sample of countries and over a longer time span compared to previous studies, which either focused on specific geographical regions, single countries, or over shorter and less recent time horizons. Inter alia, our broader sample allows us to explore heterogeneity across EMDEs. Second, we introduce a new approach to measuring spending shocks for public (and private) investment. Under this approach, we construct a measure of cyclically adjusted real public investment (CAPubI) and cyclically adjusted real private investment (CAPriI). Spending shocks can then be measured as large deviations from the variation in these variables. After identification of the investment shocks, we use the local projections approach of Jordà (2005) to trace out the shortand medium-run responses of real gross domestic product (GDP). Analyses based on local projections indicate that public investment shocks enhance economic growth significantly more than private investment shocks. In EMDEs, particularly in Asia, the impact of investment shocks on growth is more pronounced and positive during downturns than upturns. Moreover, in EMDEs overall and especially Asian EMDEs, public investment shocks tend to stimulate both private investment and consumption. The rest of the paper is structured as follows. Section 2 reviews the empirical literature, with an emphasis on cross-country studies discussing the association between public and private investment shocks and macroeconomic conditions. Section 3 presents the framework for the identification of public and private investment shocks, discusses the data and estimation methodology, and compares our approach to that of previous scholarly work. Section 4 reports the stylized facts while Section 5 presents the empirical results. Section 6 concludes. 11 are discussed in more detail in the next section, along with the new method developed in this paper. 3. Methodology and Data 3.1. Constructing Government Spending Shocks A review of existing public spending shock identification frameworks. Several strategies have been devised in the literature to identify government spending shocks. Among the most widely applied methods, an approach based on (structural) vector autoregression frameworks relies on recursive identification schemes and other parameter restrictions with the assumption that government spending does not respond to macroeconomic shocks in the same period (Blanchard and Perotti 2002, Ilzetzki et al. 2013). Besides the timing assumptions behind the underlying Cholesky identification scheme, which are less compelling for annual-frequency data, this approach is agonistic about anticipation effects of fiscal spending. A second approach, by Kraay (2012, 2014), is to instrument government shocks with predicted disbursements from approved loans from official creditors—such as the World Bank and other multilateral and bilateral aid agencies—as exogenous sources of fluctuations in government spending. This approach is difficult to replicate, as it requires a calculation of “predicted” disbursements for each loan, which can then be considered as exogenous spending shocks. A third approach is the “natural experiment” approach proposed by Barro (1981) and further developed by Ramey in subsequent papers (Ramey and Shapiro 1998, Ramey and Zubairy 2018). This framework uses fluctuations in military expenditures to identify government spending shocks. This approach, however, would not work well for 12 EMDEs, as the military spending variable does not change much from one year to the next. Finally, in more recent empirical research, Auerbach and Gorodnichenko (2012a, 2012b) use forecast errors to examine how the fiscal multiplier varies with the business cycle in Organisation for Economic Co-operation and Development (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. The authors 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 utilized also in Honda, Miyamoto, and Taniguchi (2020); Miyamoto et al. (2020); and Furceri et al. (2022). This methodology has the advantage of overcoming the problem of “fiscal foresight” (Forni and Gambetti 2010; Leeper, Richter, and Walker 2012, Leeper, Walker, and Yang 2013; and Zeev and Pappa 2017). Public (and private) spending shocks based on cyclically adjusted public variables. In this paper we build on the work of Alesina and Ardagna (2010) and others that assess the effects of changes in cyclically adjusted (fiscal) variables on growth. What is new in our approach is the way in which we identify spending shocks for public (and private) investment. The shock identification framework involves several steps. First, we estimate output elasticities of both public and private investment for each country by regressing the log of real public and private investment on the log of real GDP. 13 Estimates are done by ordinary least squares (OLS) for all countries with at least 20 continuous observations of both variables.3 Summary statistics for the elasticities are shown in Appendix Table A1. Second, we obtain measures of potential output 𝐺𝐷𝑃 via a Hodrick-Prescott (HP) filter as the baseline estimate.4 Third, we compute cyclically adjusted real public investment (CAPI): 𝐶𝐴𝑃𝑢𝑏𝐼 = 𝑃𝑢𝑏𝐼󰇡   󰇢 (1a) 𝐶𝐴𝑃𝑟𝑖𝐼 = 𝑃𝑟𝑖𝐼󰇡   󰇢 (1b) where 𝑃𝑢𝑏𝐼 is real public investment, 𝑃𝑟𝑖𝐼 is real private investment, 𝜀 denotes the output elasticities of public investment and 𝜀 denote the output elasticities of private investment. Fourth, we determine whether or not a change in public and private investment stripped of business cycle influences is large enough to be included in the sample used for the econometric analysis, and to this end, define public and private investment shocks (PubIS and PriIs, respectively) as a measure taking the value of one when a country´s first difference of 𝐶𝐴𝑃𝑢𝑏𝐼 or 𝐶𝐴𝑃𝑟𝑖𝐼 exceeds its country-specific mean by one standard deviation for each country i: 𝑃𝑢𝑏𝐼𝑆 = 1 𝑖𝑓 𝛥𝐶𝐴𝑃𝑢𝑏𝐼 > 𝛥𝐶𝐴𝑃𝑢𝑏𝐼             +𝑆𝐷𝐶𝐴𝑃𝑢𝑏𝐼 , 0 otherwise (2a) 𝑃𝑟𝑖𝐼𝑆 = 1 𝑖𝑓 𝛥𝐶𝐴𝑃𝑟𝑖𝐼 > 𝛥𝐶𝐴𝑃𝑟𝚤𝐼            +𝑆𝐷𝐶𝐴𝑃𝑟𝑖𝐼 , 0 otherwise (2b) 3 Countries with less than 20 continuous observations were not included in the sample going forward. 4 We also use Baxter-King, Christiano-Fitzgerald Random Walk, and the Hamilton (2018) and other statistical filters as a robustness check. 14 The focus on large shocks (greater than a standard deviation) is in the spirit of Alesina and Ardagna (2010), who argue that, looking specifically to the fiscal context at relatively large fiscal adjustments helps identify changes in fiscal variables that are policyinduced, rather influenced by the business cycle. This approach also makes our results more robust to any imperfections in measuring the effect of the cycle on fiscal variables, as small changes in cyclically adjusted public and private investment are excluded from the econometric analysis. In addition to this binary shock variable, we use two versions of continuous shocks based on CAPubI and CAPriI as a robustness check: shocks based on the growth rate of the CAPubI and CAPriI and shocks based on the yearly changes in the CAPubI and CAPriI expressed in percent of GDP (winsorized to account for outliers). 3.2. Methodological framework The main hypothesis to be tested is whether public and private investment shocks lead to higher real GDP. To estimate the response of real GDP to major public and private investment shocks, we follow the local projections method proposed by Jordà (2005) to estimate impulse-response functions. This approach has been advocated by Auerbach and Gorodnichenko (2013) and Romer and Romer (2019) as a flexible alternative to vector autoregressions (VAR) for estimation purposes while being on similar footing with regards to identification.5 Given the panel data setting, we adopt the local projections method over VAR estimation for the following reasons. First, the investment shocks from both public and private sectors we’re employing have been adjusted to be independent of current and 5 See Plagborg-Møller and Wolf (2021) for a discussion on the trade-offs between VARs and local projections. 15 expected future macroeconomic conditions. Consequently, there’s no need for additional identification via constraints typical in VAR models. Second, our analysis involves a large panel dataset with various fixed effects, which makes the straightforward implementation of conventional VAR models complex. 6 In addition, the local projections approach eliminates the necessity to estimate equations for any dependent variables other than the variable of interest, substantially reducing the amount of parameters required. Holding other things constant, local projections usually offer more accurate estimates for impulse responses over shorter timeframes, which suits our analysis. Third, the local projections method is particularly suited to estimating nonlinearities, as its application is much more straightforward compared to non-linear structural VAR models, such as Markov-switching or threshold-VAR models.7 In fact, local projections tend to be easier to implement relative to VARs when a specified nonlinearity would make the inversion of the VAR form into the VMA form difficult (Plagborg-Møller and Wolf 2021). Lastly, the error term in these panel estimations is likely to be correlated across countries, which is easier to handle in the local projections method by either clustering standard errors or using the Driscoll-Kraay (1998) standard errors that take account of arbitrary correlations of the errors across countries and time. Against this background, the baseline specification we estimate takes the following form: 𝑦, −𝑦, = 𝛼+𝜏+𝛽𝑠ℎ𝑜𝑐𝑘𝑠, +𝜽𝑿𝒊,𝒕 +ε, (3) 6 If one wishes to introduce country fixed effects in a VAR in a panel environment that is also possible, for instance, by demeaning each model variable over time for each country before including them in VAR. 7 See Choi and Shin (2023) and Miyamoto et al. (2020) for the recent application of local projections to the estimation of nonlinearities and interaction effects of exogenous shocks using a large international panel dataset—an approach similar to ours. 16 in which y is the dependent key economic variable of interest; 𝛽 denotes the (cumulative) response of the variable of interest in each k year after the public and private investment shocks; 𝛼, 𝜏 are country and time fixed effects, included to account for cross-country heterogeneity and global factors (such as the world business cycle or oil price movements)8; 𝑠ℎ𝑜𝑐𝑘𝑠, denotes the public or private investment shocks PubIS or PriIS ; 𝑿𝒊,𝒕 is a set a of control variables including—as in Abiad, Furceri, and Topalova (2016) and Furceri et al. (2022)—two lags of the shocks and two lags of real GDP growth. In the robustness section, to address potential omitted variable bias, we add two lags of CPI inflation and trade openness (the sum of exports and imports as a share of GDP).9 Equation 3 is estimated using ordinary least squares (OLS).10 We calculate Spatial Correlation Consistent (SCC) standard errors as proposed by Driscoll and Kraay (1998). Impulse response functions (IRFs) are then obtained by plotting the estimated 𝛽 for k= 0,1,...,5 with 90% confidence bands computed using the standard deviations associated with the estimated coefficients 𝛽 based on robust standard errors clustered at the 8 The inclusion of 𝜏 controls for cross-sectional dependence that arises from common sources of variation in the panel, it may also potentially affect the coefficient interpretation by conditioning out the common factor component of the shock series and its effect on GDP growth in equation 3. With the inclusion of the 𝜏 term, the estimated impulse responses can thus be interpreted as responses to country-specific shocks relative to the global. 9 The dynamics of GDP growth differ across countries, so that the lagged coefficients differ across countries. We acknowledge that here they are being estimated as if they were homogeneous across countries. This could lead to the classic problem of latent heterogeneity in lagged dependent variables with the implication that the OLS consistency conditions are violated (Pesaran and Smith 1995). We work under the assumption that dynamics for the shock variables are indeed exogenous and in such a case, the coefficients will converge to the average value among countries and the OLS conditions for consistency of the estimators will be met. 10 The local projection approach has an additional benefit over autoregressive distributed lag models and vector autoregressions in that it does not require asymptotic approximations or Monte Carlo simulations for the estimation of confidence bands. Confidence bands, on the other hand, are often bigger at longer horizons than those estimated in vector autoregression specifications. 17 country level. We also employ, as a robustness check, a comparison with the alternative approach of identifying fiscal shocks using forecast errors.11 Other variables used in the analysis come from the World Economic Outlook (WEO) of the International Monetary Fund (IMF) and the World Bank’s World Development Indicators (WDI) databases. 4. Stylized Facts In this section, we review major trends in public investment. We observe that for EMDEs and Asian economies, public investment–to–GDP ratio has been relatively stable over time, with a median around 5% of GDP Figure 1. There are, nonetheless, quite a degree of heterogeneity as indicated by the wide top and bottom quartiles of the distribution. This is clearer in Figure 2 where we zoom into Asia and the Pacific and observe large public investment ratios in the PRC or Maldives and low ratios in Brunei Darussalam or Pakistan. While there are variations across countries and regions, some common trends can be observed:  Post-1980s decline and subsequent recovery. Many EMDEs experienced a decline in public investment as a percentage of GDP during the 1980s and 1990s. This trend was partly due to structural adjustment programs and fiscal consolidation efforts encouraged by international financial institutions, which often led to reduced public spending on infrastructure. However, since the early 2000s, there has been a noticeable recovery in public investment levels, driven by improved fiscal positions, 11 While the presence of a lagged dependent variable and country fixed effects may, in principle, bias the estimates in small samples (Nickell 1981), the length of the time dimension mitigates this concern. The finite sample bias is in the order of 1/T, where T in our sample is 40. For robustness, we also examine a specification without the lagged dependent variables, which yields similar results. 18 increased access to international capital markets, and a renewed focus on infrastructure development as a means of stimulating economic growth.  Impact of global financial conditions. The global financial environment has had a significant impact on public investment trends in EMDEs. Periods of low global interest rates and high liquidity have often facilitated increased public investment through easier access to international borrowing. Conversely, financial crises and tightening conditions have led to reduced investment due to capital outflows and increased borrowing costs. Figure 1: Public Investment Level (percent of GDP) and Growth (percent), 1980–2019 Public investment level (percent of GDP) A: EMDEs B: Asia Real Public investment growth (percent) A: EMDEs B: Asia 2 4 6 8 10 12 1980 1990 2000 2010 2020 year 0 5 10 15 1980 1990 2000 2010 2020 year -10 -5 0 5 10 15 1980 1990 2000 2010 2020 year -5 0 5 10 15 1980 1990 2000 2010 2020 year Continued on the next page 19 EMDEs = emerging market and developing economies, GDP = gross domestic product. Note: Solid continuous lines denote the median value. Dashed lines indicate the interquartile ranges. Sources: International Monetary Fund Investment and Capital Stock Dataset; World Bank’s World Development Indicators. Figure 2: Public Investment in Asia and the Pacific, by country, 2019 (percent of GDP) AFG = Afghanistan, BAN = Bangladesh, BHU = Bhutan, BRU = Brunei Darussalam, CAM = Cambodia, FIJ = Fiji, GDP = gross domestic product, IND = India, INO = Indonesia, MAL = Malaysia, MLD = Maldives, MON = Mongolia, MYA = Myanmar, NEP = Nepal, PAK = Pakistan, PHI = Philippines, PRC = People’s Republic of China, SRI = Sri Lanka, THA = Thailand, VIE = Viet Nam. Note: The Asian Development Bank’s regular operations in Myanmar and Afghanistan have been on hold since February and August 2021, respectively. Source: International Monetary Fund Investment and Capital Stock Dataset; World Bank’s World Development Indicators. 5. Empirical Results 5.1. Baseline Estimation Results The starting point of the analysis is the estimation of the baseline equation 3 for the EMDE sample spanning the period 1980–2021 that will underpin the computation of the impulse response functions. Sub-samples of this group as well as regional breakdowns are also examined. The estimated coefficients for the binary shocks to public and private investment identified in line with the methodology outlined in Section 3 are then scaled by the average ratio of public (private) investment to GDP of the shock sample to provide a scale in the IRFs and a more direct interpretation. The impulse responses can then be BRU PAK INO FIJ CAM PHI VIE SRI NEP THA IND MAL AFG BAN MON MYA BHU PRC MLD 20 interpreted as the effect of a one-percent-of-GDP unanticipated increase in public (private) investment in year 0 on real GDP growth over the horizon of 5 years. In interpreting these results, it is important to note that as far as public investment shocks are concerned, they cover shocks occurring during both recessions and expansions, and thus cannot be used to judge whether countercyclical fiscal policies are effective or not. Figure 3 indicates that for the EMDE sample, public investment shocks, on average, lead to higher real output growth than private investment shocks. The results also suggest that both types of investment shocks lead to growth responses that are highly statistically significant throughout the IRF horizon. A public investment shock (equivalent to 1% of GDP) increases real GDP by about 1.2% over the horizon of 5 years. This is expected as public investment raises output through supply-side effects, thereby boosting potential output, in addition to its short-run demand stimulus. In turn, a private investment shock yields one-third of that growth increase, reaching 0.4% over the same horizon. A range of robustness checks with alternative shock definitions, sensitivity checks for model parametrization and sample composition, and possible bias associated with the mutual impacts of public and private investment, are discussed in Section 5.4, and confirm the baseline results. 27 Controlling for Additional Growth Short-term Drivers A potential issue with the analysis is that the results may suffer from omitted variable bias, since public and private investment programs may be implemented in response to historical economic conditions or in conjunction with other macroeconomic policy moves impacting real GDP growth. To solve this issue, we broaden the range of controls to include additional macroeconomic variables that have been shown to influence growth. In particular, we include lagged changes of (i) real exchange rates; (ii) terms of trade; (iii) short-term interest rates—to capture monetary policy actions; (iv) general government primary budget balance—to capture fiscal policy actions; (v) the Chinn-Ito index of capital controls.13 The findings of this analysis are comparable to, and not statistically different from, those found in the baseline specification, suggesting that this source of omitted variable bias is likely to be minimal in our scenario (available upon request). Controlling for GDP Growth Expectations Another possible source of endogeneity issue is that public and private investment programs may be implemented in response to the plans for future economic growth. To overcome this, we control for the expected values in t-1 of future real GDP growth rates over periods t to t+k—that is, the time horizon over which the impulse response functions are estimated. These are taken from the fall issue of the IMF World Economic Outlook for the year t-1. 13 The series (i)-(iv) are taken from the IMF World Economic Outlook database. The Chinn-Ito index is taken from http://web.pdx.edu/~ito/Chinn-Ito_website.htm. 28 6. Conclusion This paper enhances the existing body of research by incorporating a more extensive and diverse array of EMDEs than previously analyzed, and introducing a methodological approach that future researchers can readily replicate and adapt. We also take a closer look at Asia and Latin America. Our findings largely corroborate those of prior studies. Our expansion of the country sample allows for an examination of differences among various country groups and the role of initial conditions in shaping the growth outcomes of public and private investment. From a methodological perspective, the wide range of public spending multiplier estimates found in the existing literature indicates potential sensitivity to the chosen sample of countries and the time frame of analysis. Our study sheds light on this variability, specifically within the context of EMDEs, and contributes to a deeper understanding of the factors that contribute to the differing growth impacts of public and private investment. By delving into these nuances, our research contributes to policy discussions on the macroeconomic implications of both types of investment. Our evidence highlights the critical role of fiscal health and the efficiency of public spending in enhancing the growth impact of public investments. Our findings, derived from local projections across a worldwide dataset spanning from 1980 to 2021, indicate that public investment exerts a stronger influence on economic growth than private investment. Specifically, in EMDEs, a public (private) investment shock (equivalent to 1% of GDP) increases real GDP by about 1.2% (0.4%) over a horizon of 5 years. The growth impact of public investment is, on average, zero in Asia, in contrast to Latin America, which sees a positive impact. That said, within Asia, 29 we do observe a positive growth response to public investment shocks South Asia, in contrast with East Asia where there is no growth effect. One possible explanation is that the returns to public investment in infrastructure is higher in South Asia, which has an under-developed infrastructure compared to East Asia. In addition, we find that expansion of public and private investment yields greater benefits under certain circumstances. In particular, the effect of public investment in Asia in recessions is positive and significant. This finding strengthens the case for expanding fiscal space during upturns to secure policy flexibility in responding to downturns. Without sufficient fiscal space, countries might find themselves constrained in pursuing countercyclical fiscal policies during recessions. The expansion of debt during the coronavirus disease (COVID-19) pandemic (World Bank 2022) means that fiscal sustainability has become even more imperative. 30 APPENDIX List of economies in the sample Emerging market and developing economies (EMDEs) (98): Afghanistan, Albania, Algeria, Angola, Argentina, Armenia, Bangladesh, Barbados, Benin, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Burkina Faso, Burundi, Cabo Verde, Cambodia, Cameroon, Central African Republic, Chad, Chile, People’s Republic of China, Colombia, Comoros, Congo, Costa Rica, Croatia, Côte d'Ivoire, Democratic Republic of Congo, Djibouti, Dominican Republic, Egypt, Equatorial Guinea, Ethiopia, Gabon, Ghana, Guatemala, Guinea, Guinea-Bissau, Guyana, Haiti, Honduras, Hungary, India, Indonesia, Iran, Iraq, Kazakhstan, Kenya, Lao People’s Democratic Republic, Lebanon, Macedonia, Madagascar, Malawi, Malaysia, Mali, Mauritania, Mauritius, Mexico, Mongolia, Morocco, Mozambique, Myanmar, Namibia, Nepal, Nicaragua, Niger, Nigeria, North Macedonia, Pakistan, Paraguay, Peru, Philippines, Poland, Romania, Russian Federation, Rwanda, Saudi Arabia, Senegal, Serbia, Seychelles, Sierra Leone, South Africa, Sri Lanka, São Tomé and Príncipe, Tajikistan, Tanzania, Thailand, Togo, Türkiye, Uganda, Ukraine, United Arab Emirates, Uruguay, Uzbekistan, Venezuela, Viet Nam, Yemen, Zambia. ADB member countries (20): Afghanistan, Armenia, Bangladesh, Cambodia, People’s Republic of China, India, Indonesia, Kazakhstan, Lao People’s Democratic Republic, Malaysia, Mongolia, Myanmar, Nepal, Pakistan, Philippines, Sri Lanka, Tajikistan, Thailand, Uzbekistan, Viet Nam.14 Table A1: Summary Statistics for Public and Private Investment Elasticities Public investment elasticities Private Investment elasticities EMDEs Obs. 5160 5200 Mean 1.204 1.362 SD 0.939 0.835 Min -1.391 -1.035 Max 4.812 5.84 Asia Obs. 800 760 Mean 0.875 1.325 SD 0.504 0.384 Min -0.038 0.622 Max 2.192 2.345 Latin America Obs. 1120 1160 Mean 1.161 1.229 SD 0.856 0.568 Min -0.071 -0.128 Max 3.422 2.281 EMDEs = emerging market and developing economies. Sources: International Monetary Fund Investment and Capital Stock Dataset; World Bank’s World Development Indicators. 14 The Asian Development Bank’s regular operations in Myanmar and Afghanistan have been on hold since February and August 2021, respectively. 31 Table A2: Summary Statistics for Public Investment Shocks Sample Number of countries Number of observations when shock=1 Mean value of respective variable in percent of GDP when shock=1 Mean value of real GDP growth in percent when shock=1 Mean value of real GDP growth percent when shock=0 Public investment shocks, based on CAP ub I EMDEs 128 557 8.108 4.475 3.793 Asia 20 106 8.634 6.002 5.474 Latin America 28 138 6.535 4.112 2.951 Private investment shocks, based on CAPriI EMDEs 128 544 17.94 4.713 3.761 Asia 19 89 17.966 6.083 5.509 Latin America 28 136 17.969 4.598 2.904 EMDEs = emerging market and developing economies, GDP = gross domestic product. Sources: International Monetary Fund Investment and Capital Stock Dataset; World Bank’s World Development Indicators. Figure A1: Growth Effects of Public and Private Investment Shocks within Asia Public Investment Private Investment Note: t=0 is the year of the government investment shock; t=1 is the first year of impact. Solid black lines denote the response to a government investment shock. Shaded areas denote 90% confidence bands, based on standard errors clustered at the country level. Source: Authors’ calculations. -1.5 -1 -.5 0 .5 0 1 2 3 4 5 year CAPI-HP1sd,East Asia Pacific -1 0 1 2 3 0 1 2 3 4 5 year CAPI-HP1sd,South Asia Region -.4 -.2 0 .2 .4 .6 0 1 2 3 4 5 year CAPriI-HP1sd,East Asia Pacific -.5 0 .5 1 0 1 2 3 4 5 year CAPriI-HP1sd,South Asia Region 32 Figure A2: Growth Effects of Public Investment Shocks, by Type of Exporter A. Agricultural exporters B. Metal Exporters C. Oil exporters D. Energy exporters Note: t=0 is the year of the government investment shock; t=1 is the first year of impact. Solid black lines denote the response to a government investment shock. Shaded areas denote 90% confidence bands, based on standard errors clustered at the country level. Source: Authors’ calculations. 0 .5 1 1.5 2 0 1 2 3 4 5 year CAPI-HP1sd,agr 0 1 2 3 4 0 1 2 3 4 5 year CAPI-HP1sd,metal -.5 0 .5 1 1.5 0 1 2 3 4 5 year CAPI-HP1sd,oil -.5 0 .5 1 1.5 0 1 2 3 4 5 year CAPI-HP1sd,energy 33 REFERENCES Abiad, Abdul, Davide Furceri, and Petia Topalova. 2016. “The Macroeconomic Effects of Public Investment: Evidence from Advanced Economies.” Journal of Macroeconomics, 50: 224–40, 2016. Agénor, Pierre-Richard. 2010. “A Theory of Infrastructure-Led Development.” Journal of Economics Dynamics and Control, 34 (5): 932–50. Alesina, Alberto, and Silvia Ardagna. 2010. “Large Changes in Fiscal Policy: Taxes versus Spending.” Tax Policy and the Economy, 24 (1). Asian Development Bank (ADB). 2017. Meeting Asia's Infrastructure Needs. ADB. ADB. 2018. Asian Development Outlook, 2018 – How Technology Affects Jobs. April. Auerbach, Alan J., and Yuriy Gorodnichenko. 2013. “Fiscal Multipliers in Recession and Expansion.” In Alberto Alesina and Francesco Giavazzi, eds. Fiscal Policy after the Financial Crisis. University of Chicago Press. Auerbach, Alan J., and Yuriy Gorodnichenko. 2012a. “Measuring the Output Responses to Fiscal Policy.” American Economic Journal: Economic Policy 4 (2): 1–27. Auerbach, Alan J., and Yuriy Gorodnichenko. 2012b. “Fiscal Multipliers in Recession and Expansion.” In A. Alesina and F. Giavazzi, eds. Fiscal Policy after the Financial Crisis, 63–98. University of Chicago Press. Barnichon, Regis, Davide Debortoli, Christian Matthes. 2022. “Understanding the Size of the Government Spending Multiplier: It’s in the Sign.” Review of Economic Studies, 89 (1): 87–117. Barro, Robert. 1990. “Government Spending in a Simple Model of Endogenous Growth.” Journal of Political Economy, 95: 103–26. Barro, Robert J. 1981. “Output Effects of Government Purchases.” Journal of Political Economy 89 (6): 1086–1121. Barro, Robert J., and Xavier Sala-I-Martin. 1992. “Public Finance in Models of Economic Growth.” Review of Economic Studies, 59 (4): 645–61. Batini Nicoletta, Luc Eyraud, Anke Weber, Luc Forni. 2014. “Fiscal Multipliers: Size, Determinants, and Use in Macroeconomic Projections.” IMF Technical Notes and Manuals 2014/004. Berg, Andrew, Jan Gottschalk, Rafael A. Portillo, and Luis-Felipe Zanna. 2010. “The Macroeconomics of Medium-Term Aid Scaling-Up Scenarios.” IMF Working Paper 10/160. 34 Berg, Andrew, Rafael A. Portillo, Susan S. Yang, and Luis-Felipe Zanna. 2013. ”Public Investment in Resource-Abundant Developing Countries.” IMF Economic Review, 61: 92–129. Berg, Andrew, Tokhir N. Mirzoev, Rafael A. Portillo, and Luis-Felipe Zanna. 2010. “The Short-Run Macroeconomics of Aid Inflows: Understanding the Interaction of Fiscal and Reserve Policy.” IMF Working Paper No.10/65. Blanchard, Olivier, and Roberto Perotti. 2002. “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output.” Quarterly Journal of Economics, 117 (4): 1329–68. Bom, Pedro R. D., and Jenny E. Ligthart. 2014. “What Have We Learnt from Three Decades of Research on the Productivity of Public Capital?” Journal of Economic Surveys, 28 (5): 889–916. Buffie, Edward F., Andrew Berg, Catherine A. Pattillo, Rafael A. Portillo, and Luis-Felipe Zanna. 2012. “Public Investment, Growth, and Debt Sustainability: Putting Together the Pieces.” IMF Working Paper No.12/144. Canning, David, and Peter Pedroni. 2008. “Infrastructure, Long-run Economic Growth and Causality Tests for Cointegrated Panels.” Manchester School, 76 (5): 504–27. Cavallo, Eduardo, and Christian Daude. 2011. “Public Investment in Developing Countries: A Blessing or a Curse?” Journal of Comparative Economics 39, 65–81. Cerra, V., S. Tekin, and S. Turnovsky. 2008. “Foreign Aid and Real Exchange Rate Adjustments in a Financially Constrained Dependent Economy.” IMF Working Paper No. 08/204. Chatterjee, Santanu, and Stephen J. Turnovsky. 2007. “Foreign Aid and Economic Growth: The Role of Flexible Labor Supply.” Journal of Development Economics, 84 (1): 507–33. Choi, Sangyup, and Junhyeok Shin. 2023. “Household Indebtedness and the Macroeconomic Effects of Tax Changes.” Journal of Economic Behavior and Organization, 209, 22–52. Choi, Jinho, and Minkyu Son. 2016. “A Note on the Effects of Government Spending on Economic Growth in Korea.” Journal of the Asia Pacific Economy, 21 (4): 651–63. Christiano, Lawrence, Martin Eichenbaum, and Sergio Rebelo. 2011. “When Is the Government Spending Multiplier Large?” Journal of Political Economy, 119 (1): 78121. 35 Clements, Benedict, Sanjeev Gupta, and João Tovar Jalles. 2022. “Fiscal Policy for Inclusive Growth in Asia.” Center for Global Development Working Paper 611. Coenen, Günter, Christopher J. Erceg, Charles Freedman, Davide Furceri, Michael Kumhof, René Lalonde, Douglas Laxton, Jesper Lindé, Annabelle Mourougane, Dirk Muir, Susanna Mursula, Carlos de Resende, John Roberts, Werner Roeger, Stephen Snudden, Mathias Trabandt, and Jan in’t Veld. 2012. “Effects of Fiscal Stimulus in Structural Models.” American Economic Journal: Macroeconomics, 4 (1): 22–68. Coenen, Günter, Roland Straub, and Mathias Trabandt. 2012. “Gauging the Effects of Fiscal Stimulus Packages in the Euro Area.” European Central Bank Working Paper Series, no. 1483. October. Colombo Emilio, Davide Furceri, Pietro Pizzuto, and Patricio Tirelli. 2022. “Fiscal Multipliers and Informality.” IMF Working Paper No. 2022/082. Dabla-Norris, Era, Jim Brumby, Annette Kyobe, Zac Mills, and Chris Papageorgiou. 2012. “Investing in Public Investment: An Index of Public Investment Efficiency.” Journal of Economic Growth, 17: 235–66. De Gregorio, José. 1992. “Economic Growth in Latin America.” Journal of Development Economics, 39 (1): 59–84. De Jong, Jasper, Marien Ferdinandusse, Josip Funda, and Igor Vetlov. 2017. “The Effect of Public Investment in Europe: A Model-Based Assessment.” ECB Working Paper 2021, February. Devarajan, Shantayanan, Vinaya Swaroop, and Heng-fu Zou. 1996. “The Composition of Public Expenditure and Economic Growth.” Journal of Monetary Economics, 37 (2): 313–44. Driscoll, John C., and Aart C. Kraay. 1998. “Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data.” Review of Economics and Statistics, 80 (4): 549–60. Eden, Maya, and Aart Kraay. 2014. “Crowding in and the Returns to Government Investment in Low-Income Countries.” World Bank Policy Research Working Paper Series No. 6781. Forni, Mario, and Luca Gambetti. 2010. “Fiscal Foresight and the Effects of Government Spending.” CEPR Discussion Paper No. 7840. Furceri, Davide, Jun Ge, Prakash Loungani, and Giovanni Melina. 2022. “The Macroeconomic (and Distributional) Effects of Government Spending Shocks in Developing Economies.” Review of Development Economics, 26 (3): 1574–99. 36 Futagami, Koichi, Yuchi Morita, and Akihisa Shibata. 1993. “Dynamic Analysis of an Endogenous Growth Model with Public Capital.” Scandinavian Journal of Economics, 95 (4): 607–25. Galí, Jordi. 2020. “The Effects of a Money-Financed Fiscal Stimulus.” Journal of Monetary Economics 115: 1–19. Ganelli, Giovanni, and Juha Tervala. 2020. “Welfare Multiplier of Public Investment.” IMF Economic Review 68: 390–420. Gaspar, Vitor, Gerd Schwartz, Marco Cangiano, Benedict Clements, Richard Hughes, Holger van Eden, Torben Hansen, Mario Pessoa, Geneviève Verdier, Richard Allen, Benoit Chevauchez, Teresa Curristine, Renaud Duplay, Israel Fainboim, Almudena Fernandez, Lesley Fisher, Suzanne Flynn, David Gentry, Jason Harris, Tim Irwin, Suhas Joshi, Javier Kapsoli, Florence Kuteesa, Duncan Last, Samah Mazraani, Peter Murphy, JeanPierre Nguenang, Eliko Pedastsaar, Maximilien Queyranne, Isabel Rial, Sandeep Saxena, Johann Seiwald, Apolinar Veloz, and Ha Vu. 2015. “Making Public Investment More Efficient.” International Monetary Fund Staff Report. Gbohoui, William. 2021. “Uncertainty and Public Investment Multipliers: The Role of Economic Confidence.” IMF Working Papers 2021/272. Gechert, Sebastian, and Ansgar Rannenberg. 2018. “Which Fiscal Multipliers Are Regime-Dependent? A Meta-Regression Analysis.” Journal of Economic Surveys, 32: 1160–82. Ghassibe, Mishel, and Francesco Zanetti. 2022. “State Dependence of Fiscal Multipliers: The Source of Fluctuations Matters.” Journal of Monetary Economics 132: 1–23. Glomm, Gerhard, and B. Ravikumar. 1994. “Public Investment in Infrastructure in a Simple Growth Model.” Journal of Economic Dynamics and Control, 18 (6): 1173– 87. Gonzalez-Garcia, Jesus, Antonio Lemus, and Mico Mrkaic. 2013. “Fiscal Multipliers in the ECCU.” IMF Working Paper, WP/13/117. Gramlich, Edward M. 1994. “Infrastructure Investment: A Review Essay.” Journal of Economic Literature, 32, 1176–96. Greiner, Alfred. 2007. “An Endogenous Growth Model with Public Capital and Sustainable Government Debt.” Japanese Economic Review, 58 (3): 345–61. Greiner, Alfred, Willi Semmler, and Gang Gong. 2005. The Forces of Economic Growth: A Time Series Perspective. Princeton University Press.