Understanding the drivers of remittances to Pakistan

de Padua, David; Lanzafame, Matteo; Qureshi, Irfan A.; Taniguchi, Kiyoshi Working Paper Understanding the drivers of remittances to Pakistan ADB Economics Working Paper Series, No. 733 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: de Padua, David; Lanzafame, Matteo; Qureshi, Irfan A.; Taniguchi, Kiyoshi (2024) : Understanding the drivers of remittances to Pakistan, ADB Economics Working Paper Series, No. 733, Asian Development Bank (ADB), Manila, https://doi.org/10.22617/WPS240348-2 This Version is available at: https://hdl.handle.net/10419/301966 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 UNDERSTANDING THE DRIVERS OF REMITTANCES TO PAKISTAN David de Padua, Matteo Lanzafame, Irfan Qureshi, and Kiyoshi Taniguchi ADB ECONOMICS WORKING PAPER SERIES NO. 733 July 2024 Understanding the Drivers of Remittances to Pakistan Remittances are an important source of external financing in Pakistan, accounting for approximately 10% of gross domestic product. Combining a database of bilateral remittances between Pakistan and its main remittance-sending countries with monthly macroeconomic data over 2003–2021, we use a Bayesian vector autoregression model to understand the drivers of remittances. We find that macroeconomic variables, including economic activity, inflation, equity markets, and interest rates—both in Pakistan and migrants’ host countries—play a significant role, and their contributions vary over time. 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. 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 Understanding the Drivers of Remittances to Pakistan David de Padua, Matteo Lanzafame, Irfan Qureshi, and Kiyoshi Taniguchi No. 733 | July 2024 David de Padua ([email protected]) is an economics officer; Matteo Lanzafame ([email protected]) is a principal economist; and Kiyoshi Taniguchi ([email protected]) is a regional lead economist at the Economic Research and Development Impact Department, Asian Development Bank (ADB). Irfan Qureshi (iqur[email protected]g) is a public management specialist in the Sectors Group of ADB. 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. WPS240348-2 DOI: http://dx.doi.org/10.22617/WPS240348-2 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. Note: In this publication, “PRs” refers to Pakistan rupees and “$” refers to United States dollars. ABSTRACT Remittances are an important source of external financing in Pakistan, amounting to around 10% of gross domestic product in 2021. As such, an appropriate understanding of the key macroeconomic drivers of remittances has important policy implications. Combining a database of bilateral remittances between Pakistan and its main remittance-sending countries with monthly macroeconomic data over 2003–2021, we use a Bayesian vector autoregression model to understand the drivers of remittances to Pakistan. Specifically, we do so by estimating the impact of various structural shocks on remittance growth in Pakistan. We find that macroeconomic variables, including economic activity, inflation, equity markets, and interest rates—both in Pakistan and migrants’ host countries—play a significant role, and their contributions vary over time. Keywords: remittances, macroeconomics, Pakistan JEL codes: E7, F22, F24 1. Introduction This paper investigates the macroeconomic drivers of remittances to Pakistan. In 2022, remittance flows to low and middle-income countries were estimated at $630 billion. Understanding what drives remittances has become critical to policy in at least two ways. These flows help smooth consumption and positively impact poverty reduction (Adams and Page 2005). Policy can enhance this role of remittances, for instance, by reducing remittance transaction costs or incentivizing remittance receivers to channel spending to productive investments. Remittances have played an important role in Pakistan’s development process since the 1990s (Iqbal and Sattar 2005). These financial flows have also become crucial for macroeconomic dynamics in the country for three reasons. The first is their relative size. Following continued growth in the 2000s, remittances account for about 10% of Pakistan’s gross domestic product (GDP), considerably higher than that for the South Asia as a whole (Figure 1a). The second is their relative stability. Remittances to Pakistan proved resilient to the impact of the Global Financial Crisis (GFC) of 2007–2008 and soared during the coronavirus disease (COVID-19) pandemic (Figure 1b). In 2021, Pakistan received $31.1 billion in remittances from overseas, a 19.8% increase from the previous year. The third is that remittances have grown so large, which roughly equals the value of net imports of goods and services (Figure 2). As such, they have nontrivial effects on the balance of payments and macroeconomic stability (Bugamelli and Paterno 2005). 2 Figure 1: Remittances to Pakistan and Comparators (a) Pakistan and South Asia (b) Remittances to Pakistan and the World GDP = gross domestic product, lhs = left-hand side, rhs = right-hand side. Source: World Development Indicators. Figure 2. Remittances in the Balance of Payments of Pakistan Source: CEIC Data company. -15,000 -10,000 -5,000 0 5,000 10,000 Primary Income Secondary Income of which: Remittances Net Goods and Services Current Account $ million 0 5 10 15 20 25 30 35 0 100 200 300 400 500 600 700 800 World - lhs Pakistan - rhs $ billions $ billions 0 1 2 3 4 5 6 7 8 9 10 Pakistan South Asia % of GDP 3 Remittances are a well-documented phenomenon, owing to a large literature of micro and macro studies. Microeconomic studies on remittances focus on microeconomic data and surveys, typically looking at individual characteristics of remitters, such as education, sex, origin, income levels, and household characteristics. These studies tend to have a development focus and investigate behavioral aspects of remittances, relating to migrants’ motives to remit or how remittance-receiving households spend the money received (Adams 2009, Lucas and Stark 1985). Macroeconomic studies, on the other hand, focus largely on the macroeconomic conditions that surround remittance decisions, looking at interest rate differentials, business cycles, or the impact of remittances on GDP growth (El-Sakka and McNabb 1999, Frankel 2009, Sayan 2006). The literature on the macroeconomic drivers of remittances is lacking as the evidence so far produced is mostly inconclusive. There are two main advantages to undertaking a study on remittances with a macro perspective. Many studies explore the development impact of remittances or the microeconomic determinants of remittances, but few endeavor to unpack the influence of domestic and foreign macroeconomic variables on remittances. Second, it is well established that remittance flows behave differently from other types of capital flows. The motivations for remitting money are different from that of foreign direct investment or portfolio flows: they are sent to compensate for income shortfalls or to be invested on behalf of the remitter. Remittances are a boon for development and have increasingly become a pillar for macroeconomic stability. Understanding how migrant characteristics or demographics may influence long-term development outcomes is important. However, this does not provide insight into how changes in macroeconomic conditions can influence remittances, which, in turn, could potentially impact the macroeconomic stability of the remittance-receiving economy. This paper aims to bring a macroeconomic perspective to the analysis of remittances to Pakistan. We use the theoretical framework Chami et al. (2008) proposed as a departure point to investigate the drivers of remittances and interpret our findings. Our analysis focuses largely on 4 the role played by receiving and sending-country macroeconomic and financial conditions in shaping remittance flows to Pakistan rather than attempting to disentangle the intertwined motivations to remit. As such, our approach does not dwell deeply on behavioral motivations but uses them to underline the economic intuition behind the dynamics of remittances and selected macroeconomic variables. Using Bayesian vector autoregression (VAR) techniques and applying historical decomposition methods, we characterize the behavior of remittances to Pakistan in response to key domestic and foreign variables. To unpack country-specific dynamics, we run the VAR analysis for specific key remittance-sending partners: Saudi Arabia, the United Kingdom (UK), and the United States (US). We dig further into the drivers of these changes, using a historical decomposition approach to understand how the contributions of the drivers have changed over time and see how these changes are related to key economic events. Finally, we zoom in to specific crisis periods. Our key results can be summed up as follows. We find macroeconomic factors to have a significant impact on remittance growth in Pakistan, notably economic activity in both home and host economies, domestic inflation, and to some extent, oil prices. The influence of these factors vary from country to country, reflecting considerable economic differences in host countries. We find that the importance of these factors evolved over time and different factors helped bolster remittance growth through crises. The remainder of the paper is organized as follows; in section 2, we provide a review of the literature; section 3 discusses the VAR approach and describes the data; section 4 presents the results, first discussing the baseline model and then country-specific case studies; and section 5 concludes the paper. 11 For both the aggregate and country-specific models, historical decompositions of the contributions to remittance dynamics are constructed. To illustrate, consider a simple case where the VAR model has only one lag: 𝑌=𝐴𝑌 + 𝐶𝑥+𝜀 (2) By backward substitution: 𝑌=𝐴𝑌 + 𝐶𝑥+𝜀 =𝐴(𝐴𝑌 + 𝐶𝑥 +𝜀)+ 𝐶𝑥+𝜀 =𝐴𝐴𝑌 + 𝐶𝑥+ 𝐴𝐶𝑥 +𝜀+𝐴𝜀 (3) Going on, one may go back to the beginning of the sample and in general for a model with p lags, one can rewrite 𝑌 as: 𝑌=𝐴 ()   𝑌 + 𝐶   𝑥 +𝐵   𝜀 (4) where matrix series 𝐴 (), 𝐶 and 𝐵 are functions of 𝐴,𝐴 ,… ,𝐴. The t superscript emphasizes that the matrix 𝐴 () depends on 𝑡 =1,2,…,𝑇 while 𝐶 and 𝐵 do not. The matrices 𝐵,𝐵 ,… ,𝐵 provide the response of 𝑌 shocks and are thus the series of impulse response function matrices, which can then be represented in terms of structural shocks.4 𝑌 can thus be separated into two parts: the first being deterministic exogenous variables and initial conditions, and the second due to the contribution of structural disturbances affecting the dynamics of the model. These provide an interpretation of historical fluctuations in the modeled time series through the lens of the identified structural shocks to remittance flows. This extension of the analysis allows us to determine whether the importance of each of the variables 4 For a more detailed discussion, Dieppe et al. (2018) provide a step-by-step exposition of this process in in the technical guide for their Bayesian Estimation, Analysis and Regression Toolbox. Historical contribution of deterministic variables Historical contribution of structural shocks 12 in the system varies over time, providing valuable insights for a sample period that includes spells of rapid growth in remittances, the GFC, and the COVID-19 pandemic. Table 1: Model Summary Model Variable of Interest Economic Factor Other Factor Global Factor Aggregate Remittances to Pakistan Economic activity (monthly imports) Equities U.S. Federal Funds rate Exchange rate – PRs/$ Inflation – weighted average of sample Global oil price VIX Country-specific Remittances to Pakistan Economic activity (monthly imports) Equities Short term interest rates Bilateral exchange rates Inflation Global oil price VIX Source: Authors. Using VAR and historical decomposition techniques, we characterize the behavior of remittances to Pakistan in response to key domestic and foreign variables. Using a historical decomposition approach, we dig further into the determinants of these changes. This shows how the drivers’ contributions have changed over time, particularly in relation to key economic events. Finally, we zoom in to specific crisis periods to illustrate whether any of the aggregate patterns and factors driving remittances change during such episodes. We outline and discuss our results next. 4. Empirical Findings 4.1. Baseline: Aggregate Model This section presents and discusses the baseline estimates of the aggregate model, which characterizes the behavior of remittances to Pakistan in response to key domestic and foreign variables. Throughout the analysis, the interpretation of results relies on (orthogonalized) impulse response functions (IRFs) for a period covering up to 24 months. The IRFs document the impact of a shock to each of the drivers of remittances and trace their evolution over time. For ease of presentation, the discussion of the empirical results is organized by groups of endogenous 13 variables, starting with the impact of shocks related to economic activity, followed by policy factors, and finally, touching briefly on the global factors. The findings indicate that domestic and abroad economic activity is positively associated with remittances to Pakistan. Stronger economic activity abroad typically boosts average earnings for migrants, which, in turn can translate into higher remittances (Figure 3). The positive association between remittances and domestic economic activity indicates that migrants tend to remit more when economic conditions are improving back home. This suggests that there is an opportunistic dimension to the remittance motives of Pakistani migrants: that is, to some extent, they remit money to take advantage of investment opportunities in times of economic upswings in their home country. This contrasts with findings suggesting that remittances are countercyclical. The results also show evidence of a negative association between domestic equities and remittance growth: a decline in Pakistan’s stock index leads to an increase in remittances (Figure 3). One possible interpretation of this result suggests a wealth effect is at play. Migrants send more money home to compensate for losses from the stock market. However, in the period under analysis, the equity market in Pakistan was relatively underdeveloped and remains so even now compared to regional peers. At the end of 2021 the overall stock market valuation was only 15% of GDP—compared to, for instance, 48% in Indonesia or 93% in the Philippines. In addition, according to the State Bank of Pakistan (SBP), household savings are low, with domestic savings averaging only 7% from 2015 to 2020. In contrast, the savings rate in Bangladesh was 22% and 29% in India over the same period. Moreover, less than half of national savings are channeled into the financial sector, suggesting that investing in equities is uncommon. Therefore, another possible interpretation of this finding is that the benchmark equities index functions as a barometer for political and economic conditions and prospects. Negative economic news is usually quickly reflected in stock valuations—as such, when business expectations and the economy’s outlook worsen, equities fall. In such a case, the positive response of remittances suggests a compensatory dynamic to this perceived negative shock. 14 Table 2: Summary of Impact on Remittances for Baseline and Country Cases Aggregate US Saudi Arabia UK Economic activity Domestic positive positive mixed positive Abroad positive positive positive mixed Equities - domestic negative negative negative negative Equitiesabroad mixed positive positive negative Policy factors Inflation – domestic positive positive positive positive Inflation - abroad positive negative mixed positive Interest rate - domestic positive mixed mixed negative Interest rate - abroad mixed mixed positive negative Exchange rate positive mixed mixed positive Global factors Oil price positive positive positive positive Financial volatility mixed positive mixed positive UK = United Kingdom, US = United States. Note: Boldface indicate relationships that are significant at 90% confidence intervals. Mixed refers to cases where the response of remittances is negative for some period and positive in others. Source: Authors’ estimates. Figure 3: Impact of Economic Activity Shocks on Remittances Note: Figures show orthogonalized impulse response functions for 24 months with 90% confidence intervals. Source: Authors’ estimates. 15 Results show that domestic inflation has a persistent positive impact on remittances, albeit not always statistically significant (Figure 4). That is, an increase in inflation in Pakistan translates into higher remittances. This result is also found in other studies (e.g., Rivera and Tullao 2020) and indicates that, in line with the altruistic motive to remit, migrants send more money back home when domestic inflation is accelerating and eroding households’ real incomes. The evidence gathered also indicates that a higher domestic interest rate initially results in lower remittances, but the effect turns positive and significant in the subsequent months (Figure 4). This is broadly consistent with several other studies that found that migrants remit more when interest rates back home are high. This suggests that migrants could be remitting money to take advantage of (relatively) improved investment opportunities in the form of higher interest rates back home, but they do so with some delay. In contrast, interest rates abroad do not exert a statistically significant impact through the 24-month period. Taken at face value, increasing interest rates in Pakistan compensate for typically higher risk premia, thus attracting more remittances for the opportunistic motive. The exchange rate is also broadly insignificant throughout the 24-month period considered—except briefly 6 months after the shock, when a depreciation is associated with more remittances. 16 Figure 4: Impact of Policy Factor Shocks on Remittances Note: Figures show orthogonalized impulse response functions for 24 months with 90% confidence intervals. Source: Authors’ estimates. For global factors, the results show that rising oil prices are associated with a persistent increase in remittances (Figure 5). This empirical finding is likely to be mainly driven by the fact that Saudi Arabia, a global oil market maker, is a key remittance-sending partner economy, accounting for nearly a quarter of remittances on average over our sample period.5 The transmission mechanism can work via the host-country economic-activity channel as well as via a rise in inflation due to higher oil prices, which we have seen is positively associated to remittances. The results for the VIX, an indicator of financial volatility or uncertainty, suggest an increase in remittances when financial uncertainty spikes. As for equity valuations, this is consistent with the compensatory motive to remit, as investor sentiment typically worsens with 5 The next largest remittance-sending economies are the UK and the US, which accounted for an average of 11% and 20% respectively over the sample period. 17 increasing uncertainty. But the VIX effect on remittances is short-lived and becomes insignificant in subsequent periods. Putting these results into the context of opportunistic or compensatory motives, it’s useful to consider that these motives can coexist and both likely play a role for all variables in the model. As such, the sign and size of the impact of the estimates reflect the net effect of opportunistic and compensatory motives which work in opposite directions. Domestic activity is positively associated with remittances, suggesting opportunistic motives, while domestic equities are negatively associated with remittances, suggesting compensatory motives. This apparent contradiction can be explained If the two effects are always in play, this apparent contradiction is easier to explain: depending on the variable considered, we find that one effect or the other dominates, so the sign can be either positive or negative. Figure 5: Impact of Global Factor Shocks on Remittances Note: Figures show orthogonalized impulse response functions for 24 months with 90% confidence intervals. Source: Authors’ estimates. 4.2. Decomposing Contributions to Remittance Growth Having discussed the macroeconomic drivers of remittance flows, in this section, we decompose the growth of remittances into the contributions of the various factors we identified in the previous section. This exercise works by decomposing, for every period in the sample, the value of each 18 variable into different components corresponding to the other variables in the VAR. In this way, we can identify how much of remittance growth can be attributed to each macroeconomic variable—such as oil prices—based on the empirical model. To our knowledge, ours is the first paper to do so in the context of remittance flows to Pakistan. The decompositions are available for all the variables in the system, but for the purposes of this analysis, we focus only on the decomposition of remittances. Figure 6 illustrates the results from the historical decomposition. The results show that the contributions of economic activity and other macroeconomic factors to remittance dynamics are sizable, amounting to roughly 18 percentage points of remittance growth on average over 2003– 2021 and up to as much as 40 percentage points in some subperiods. Interestingly, the net effect of these factors changes over time. For instance, their overall contribution boosted growth in remittances from 2010 to 2012, but economic activity then dragged it down from 2013 through 2018—even leading to a decline in remittances in early 2017. This highlights the fact that macroeconomic stability, as well as fluctuations, matter as drivers of remittances. This is an especially important and policy-relevant finding. One specific example of this is the estimated impact of oil prices on remittances. Recall that the IRFs for these variables showed a positive relationship—that is, they move in the same direction—that persists for up to 2 years. Oil prices collapsed in 2020 as the COVID-19 pandemic weakened global demand: based on the historical decompositions, this subtracted as much as 9 percentage points from remittance growth and an average of 4 percentage points in 2021. More generally, in an economy such as Pakistan’s, where remittances consistently buttress the current account, understanding how macroeconomic developments and macro-policy actions can affect these flows becomes crucial. In 2023, Pakistan faced severe balance-of-payments challenges, as it struggled to meet external debt payments as foreign exchange reserves plummeted. Remittances played a considerable role in this crisis, declining by 15% in January to October compared to the same period the previous year. 19 Figure 6: Historical Decomposition of Remittance Growth —Evolution of Drivers over Time Source: Authors’ estimates. One additional interesting insight that merits discussion is that the contribution of the trend component is estimated to be relatively large. This can be taken as representing the structural factors that influence the growth of remittances outside of the variables identified in our model. This outcome is consistent with the evidence indicating that remittances are relatively stable financial flows, underpinned to a significant extent by motivations outside of fluctuations in macroeconomic variables. This trend component can be thought of as what remittance growth would be in the absence of shocks to the other variables. That remittance growth would have been positive through most of the sample period, even excluding the other factors, speaks to the persistence of remittances. One implication of this is that non-macroeconomic factors, such as the cost and ease of remitting or the number of migrants in host countries, matter to remittance growth. Additional scrutiny of this finding is beyond the scope of this brief but merits further research—particularly to investigate the extent to which these structural drivers can also be affected by policy, perhaps of a more micro-nature (e.g., taxation of remittances, financial innovation, and investment vehicles such as diaspora bonds). -40 -20 0 20 40 60 80 Economic activity Inflation, interest rates, exchange rates Global factors Trend component Remittance growth 20 4.3. Country Case Studies 4.3.1 Motivation In this section, we investigate whether important country-specific dynamics may be masked by the aggregate model. To do so, we run country-specific Bayesian VAR’s for three key remittancesending partner economies: Saudi Arabia, the UK, and the US. In 2020, Saudi Arabia accounted for nearly a quarter of Pakistani migrants, while the UK and US accounted for, respectively, 8% and 6% of Pakistani migrants. Combined, the three economies are the source of roughly half of the remittances to Pakistan (Figure 7). Figure 7: Remittances to Pakistan, by Country UK = United Kingdom, US = United States. Source: State Bank of Pakistan. The impact of shocks on the drivers of remittances for the three country cases is summarized in Table 2. One finding uncovered by the aggregate model estimations that remains present across all the country cases is the positive and significant impact of domestic inflation on remittances. This evidence reinforces the view that, by eroding household incomes, accelerating Saudi Arabia UK US United Arab Emirates Oman Qatar Kuwait Bahrain Germany 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 27 the index went into free fall and plunged to as low as 5,377.4 in 2009. Plummeting equities drove remittances higher, as migrants used compensatory transfers to support family members back home for financial and other losses amid a serious economic downturn. The negative impact of economic activity for 2007 and 2009 is in line with this interpretation, shaving roughly 5 percentage points from remittance growth as economic slowdowns abroad presumably dented migrant incomes. In 2009, key remittance sending economies Saudi Arabia, the UK, and the US contracted by 2%, 4.5%, and 2.6% respectively. Inflation in Pakistan was also high during this period, averaging 13.2% from 2007 to 2009. This explains the consistent positive contribution to remittance growth through the crisis years. Migrants remitted more as their family members back home faced rising prices. The COVID-19 pandemic was a fundamentally different crisis. While the GFC was, in essence, a financial crisis, the pandemic was a health crisis that had dire economic consequences. In this case, we find economic activity abroad made the largest contribution to growth through this period. COVID-19 did not have an immediate or simultaneous impact across the world, it spread gradually with specific economies shutting down more quickly than others. The results nevertheless show that the positive contribution of economic activity declined consecutively in 2020 and 2021, suggesting that remittances slowed as host economies lost momentum. 28 Figure 11: Historical Decomposition—Global Financial Crisis and COVID-19 Pandemic COVID-19 = coronavirus disease, GFC = Global Financial Crisis. Source: Authors’ estimates. 4.4.3 Crisis Episodes—Country Cases To garner further insights, we carry out the same decomposition for the country cases of the Saudi Arabia and the US. Our findings show that the drivers of remittances from the two countries varied through the different crises (Figure 12). In the US, the contributions of inflation during the COVID19 pandemic were opposite to their contributions during the GFC. That is, high and rising inflation in Pakistan during the GFC boosted remittances from the US, while weakening (but still high) inflation during the COVID-19 downturn reduced remittances. A similar pattern is evident for inflation in the US. In contrast, accelerating US inflation negatively affected remittances to Pakistan during the COVID-19 pandemic—likely because of spiking price pressures in 2021 denting migrants’ disposable incomes. GFC COVID - 19 Pandemic 29 In Saudi Arabia, we see this “flipping” pattern for the exchange rate and domestic interest rate. During the GFC, the Pakistani rupee depreciated significantly against the Saudi riyal. A stronger riyal translated into a higher local-currency value of remittances to Pakistan. While not exhaustive, these results present some evidence of how fluctuations in key macroeconomic variables in partner economies matter. These relationships seen through periods of crises, typically characterized by extremes (e.g., high inflation, large depreciations), have also illustrated the bidirectional nature of the shock impacts. It becomes evident that the underlying economic context and circumstances driving the macroeconomic shocks are crucial to understanding how they might impact remittances. Figure 12: Historical Decomposition—Country Cases COVID-19 = coronavirus disease. Source: Authors’ estimatesp. 30 5. Conclusion This study provides an assessment of the impact of macroeconomic variables on remittances to Pakistan. We characterize how changes in these variables, both domestically and abroad, can affect remittance growth. This can help policymakers disentangle the possible impact of macroeconomic variables on remittances. Specifically, we identify economic activity, domestic interest rates, and domestic inflation as having significant effects on remittance growth. We also find that the relative importance of these macroeconomic drivers has shifted over time— particularly during the deep crises of the GFC and the COVID-19 pandemic. Additionally, we find that remittances are largely influenced by structural factors outside of the variables identified in our model and that migrants’ motivations, as identified in the microeconomic literature, likely underpin the persistence of remittance flows. Understanding and using these relationships to anticipate fluctuations in remittance growth could be useful in analyzing the Balance of Payment (BoP) framework and potentially anticipating related pressures. For instance, Pakistan’s current account deficit widened from 0.8% in 2021 to 4.6% in 2022. With global headwinds mounting alongside high food and energy prices, slowing global demand and rising interest rates abroad threaten external sector stability. Since remittances have historically served as a useful buffer to withstand BoP shocks, understanding better how these flows might evolve amid challenging global conditions is critical. In this respect, our results provide useful evidence for Pakistan’s policymakers, which can be used as a basis to think through how changing macroeconomic conditions might impact remittances to the country. Further research is necessary to unpack country-specific dynamics. While this aggregate model is useful to establish an understanding of the average impact of macroeconomic shocks, it almost certainly masks heterogeneity across remittance-sending partners where macroeconomic conditions vary greatly. For instance, growth and inflation dynamics in the US are very different from that of oil-rich Saudi Arabia, but both are major remittance-sending partners for Pakistan. Given the bilateral nature of the remittance dataset, digging into country-specific dynamics would 31 be a natural extension of the analysis presented in this paper and would provide more detailed insights to inform policy. Thinking about remittances from a more general economic development perspective, our findings provide two main insights. First, the compensatory nature of remittances during crisis episodes, such as inflationary spikes or economic crises, indicates that these financial flows remain an important shock-absorbing mechanism for households in Pakistan. Policy can enhance this role played by remittances by linking fiscal incentives for remittances to some of the macroeconomic indicators considered in this paper—so that they would be triggered by particularly large drops in economic activity or price increases. Second, since the elevated persistence of remittances to Pakistan largely reflects deterministic factors, average remittance flows can be expected to remain significant going forward. Policy can play an important role in sustaining this trend—e.g., by fostering financial development to remove structural hurdles to remittances. Perhaps even more importantly, policy measures can incentivize the allocation of remittances toward productive investment both in physical and human capital. This would boost the economy’s potential growth rate and, based on our estimates, ultimately reinforce remittance flows via the domestic economic activity channel. This virtuous circle could help underpin a more sustainable growth process in Pakistan. 32 REFERENCES Abiad, Abdul, and Irfan A. Qureshi. 2023. “The Macroeconomic Effects of Oil Price Uncertainty.” Energy Economics, 125: 106839. Acosta, Pablo, Cesar Calderon, Pablo Fajnzylber, Humberto Lopez. 2008. “What is the Impact of International Remittances on Poverty and Inequality in Latin America?” World Development, 36(1): 89–114. Adams, Richard H. 2009. “The Determinants of International Remittances in Developing Countries.” World Development, 37 (1): 93–103. Adams, Richard H. and John Page. 2005. “Do International Migration and Remittances Reduce Poverty in Developing Countries?” World Development, 33 (10): 1645–1669. Aggarwal, Reena, Asli Demirguc-Kunt and Maria Soledad Martinez Peria. 2011. “Do Remittances Promote Financial Development?” Journal of Development Economics, 96 (2): 255–264. Alleyne, Dillon. 2006.”Motivations to Remit in CARICOM: A GMM Approach.” Social and Economic Studies, 55 (3): 69–87. Arapi-Gjini, Arjola, Thomas Herzfeld and Judith Mollers. 2020. “Measuring Dynamic Effects of Remittances on Poverty and Inequality with Evidence from Kosovo.” Eastern European Economics, 58 (4): 283–308. Bettin, Giulia and Riccardo Lucchetti. 2016. “Stead Streams and Sudden Bursts: Persistence Patterns in Remittance Decisions.” Journal of Population Economics, 29 (1): 263–292. Bouoiyour,Jamal and Amal Miftah. 2014. “The Effects of Remittances on Poverty and Inequality: Evidence from Rural Southern Morocco.” MPRA Paper, University Library of Munich, Germany. Bugamelli, Matteo, and Francesco Paterno. 2009. “Do Workers’ Remittances Reduce the Probability of Current Account Reversals?” World Development, 37 (12): 1821–38. Cazachevici, Alina, Tomas Havranek and Roman Horvath. 2020. “Remittances and Economic Growth: a Meta-analysis.” World Development 134. Chami, Ralph, Adolfo Barajas, Thomas Cosimano, Connel Fullenkamp, Michael Gapen, and Peter Montiel. 2008. “Macroeconomic Consequences of Remittances.” International Monetary Fund Occasional Paper No. 259. Washington, DC. Coulibaly, Dramane. 2009. “Macroeconomic Determinants of Migrants’ Remittances: New Evidence from a Panel VAR.” Centre d’Economie de la Sorbonne Working Paper. De Soyres, Francois, and Alexandre Gaillard. 2022. “Global Trade and GDP Comovement”. Journal of Economic Dynamics and Control, 138. Dieppe, Alistair, Bjorn Van Roye, and, Romain Legrand 2016. The Bayesian Estimation, Analysis and Regression (BEAR) Toolbox Technical Guide. 33 El-Sakka, M. I. T., and Robert McNabb. 1999. “The Macroeconomic Determinants of Emigrant Remittances.” World Development, 27 (8): 1493–502. Frankel, Jeffrey. 2009. “Are Bilateral Remittances Countercyclical?” CID Faculty Working Paper No. 185. Center for International Development (CID), Harvard University. Giuliano, Paola and Marta Ruiz-Arranz. 2009. “Remittances, Financial Development, and Growth.” Journal of Development Economics, 90 (1): 144–152 Hagen-Zanker, Jessica, and Melissa Siegel. 2007. “The Determinants of Remittances: A Review of the Literature.” Maastricht Graduate School of Governance Working Paper. Imai, Katsushi, Raghav Gaiha, Abdilahi Ali and Nidhi Kaicker. 2014. “Remittances, Growth and Poverty: New Evidence from Asian Countries.” Journal of Policy Modeling, 36 (3): 524–538. Iqbal, Zafar, and Abdus Sattar. 2005. “The Contribution of Workers’ Remittances to Economic Growth in Pakistan.” Pakistan Institute of Development Research Report No. 187. Koczan, Zsoka and Franz Loyola. 2018. “How Do Migration and Remittances Affect Inequality? A Case Study of Mexico.” International Monetary Fund Working Paper 18/136. Koechlin, Valeria and Gianmarco Leon-Ciliotta. 2007. “International Remittances and Income Inequality: An Empirical Investigation.” Journal of Policy Reform, 10 (2). Koop, Gary, and Dimitris Korobilis. 2010. “Bayesian Multivariate Time Series Methods for Empirical Macroeconomics.” Rimini Centre for Economic Analysis Working Paper. Lucas, Robert E. B., and Oded Stark. 1985. “Motivations to Remit: Evidence from Botswana.” Journal of Political Economy, 93 (5): 901–918. Meyer, Dietmar and Adela Shera. “The Impact of Remittances on Economic Growth: An Econometric Model.” Economia, 18 (2) 147–155. Ratha, Dilip. 2003. “Workers’ Remittances: An Important and Stable Source of External Development Finance.” Sayan, Serdar. 2006. “Business Cycles and Workers’ Remittances: How Do Migrant Workers Respond to Cyclical Movements of GDP at Home?” International Monetary Fund Working Paper 06/52. Shen, I-Ling, Frederic Docquier and Hillel Rapoport. 2009. “Remittances and Inequality: a Dynamic Migration Model.” Journal of Economic Inequality, 8: 197–220. Vargas-Silva, Carlos and Peng Huang. 2006. “Macroeconomic Determinants of Workers’ Remittances: Host versus Home Country’s Economic Conditions.” Journal of International Trade & Economic Development, 15 (1): 81–99. Yoshino, Naoyuki, Miyu Otsuka and Farhad Taghizadeh-Hesary. 2017. “International Remittances and Poverty Reduction: Evidence from Asian Developing Countries.” ADBI Working Papers. ASIAN DEVELOPMENT BANK ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org UNDERSTANDING THE DRIVERS OF REMITTANCES TO PAKISTAN David de Padua, Matteo Lanzafame, Irfan Qureshi, and Kiyoshi Taniguchi ADB ECONOMICS WORKING PAPER SERIES NO. 733 July 2024 Understanding the Drivers of Remittances to Pakistan Remittances are an important source of external financing in Pakistan, accounting for approximately 10% of gross domestic product. Combining a database of bilateral remittances between Pakistan and its main remittance-sending countries with monthly macroeconomic data over 2003–2021, we use a Bayesian vector autoregression model to understand the drivers of remittances. We find that macroeconomic variables, including economic activity, inflation, equity markets, and interest rates—both in Pakistan and migrants’ host countries—play a significant role, and their contributions vary over time. 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.