A central insight in international macroeconomics over the past two decades is that the degree of international financial integration observed in the data falls significantly short of theoretical predictions (Cole and Obstfeld, 1991; Obstfeld, 2001).
Introduction
A central insight in international macroeconomics over the past two decades is that the degree of international financial integration observed in the data falls significantly short of theoretical predictions (Cole and Obstfeld, 1991; Obstfeld, 2001). Despite the dramatic expansion of trade and financial globalization since the late 1990s, international financial markets remain incomplete and segmented. In other words, there is no globally available asset that enables investors to insure against all possible states of the world, nor is there sufficient investment in assets that could provide insurance against a broad range of economic risks. Households in one country typically cannot directly access the savings of households in another without relying on financial intermediaries. These features manifest as home bias and imperfect international risk sharing—the empirical observation that investors systematically underinvest in foreign assets, whether in the form of equities, bonds, or cross-border loans. As a result, the ideal of a fully diversified global portfolio remains elusive. This limited international integration implies that the full potential benefits of globalization—in terms of higher growth, enhanced risk-sharing, and welfare improvements—have yet to be fully realized.
While advances in international macroeconomics have produced sophisticated models of incomplete and segmented financial markets,[1] empirically quantifying the degree of incompleteness and the precise nature of segmentation remains a major challenge. Fundamental questions—such as the actual extent of financial integration, its relationship to trade integration, and the roles played by different types of foreign investors in segmenting markets—remain unresolved. My research addresses these gaps by leveraging micro-level data on global investors and domestic borrowers, as well as detailed information on trade and financial transactions these agents perform, to measure the degree of international market integration and to characterize the specific forms of segmentation. Such identification is central to assessing the growth and welfare consequences of global shocks and the domestic and international transmission of macroeconomic policies. A particular focus of my work is identifying the nature of frictions, whether financial, informational, or policy-based—that shape market behavior by influencing both current transactions and future investor expectations, and hence central to driving systematic deviations from the full-integration benchmark in the data.
[1] This is an extensive literature, spanning more than two decades. While it is not practical to cite all the papers here, one can see Itskhoki and Mukhin, 2021 and Gabaix and Maggiori, 2015, for the most recent state-of the-art examples.
To date, most theoretical representations of such frictions remain ad hoc or reduced-form, due largely to the scarcity of micro-level empirical evidence. My work aims to bridge this gap.
Early empirical work in international macroeconomics primarily adopted a reduced-form approach. While theoretically motivated, this literature faced significant limitations in identification due to its reliance on macro-level time series data, which constrained its ability to establish causality and/or uncover the micro-level mechanisms underpinning macroeconomic models, given measurement problems. A seminal example is Feldstein and Horioka (1980), who documented a high correlation between national savings and investment across OECD countries—an outcome inconsistent with the prediction of zero correlation under full capital mobility. Building on this, a vast empirical literature has explored various quantity puzzles in international capital flows that parallel longstanding asset pricing and exchange rate puzzles. These include the high correlation of consumption across countries, the weak empirical link between exchange rates and relative consumption (Backus and Smith, 1993), and the surprisingly limited flow of capital from rich to poor countries (Lucas, 1990). From a finance perspective, related puzzles arise in international portfolio allocations, with evidence of persistent home bias in equity and bond holdings (French and Poterba, 1991; Coval and Moskowitz, 1999). The exchange rate disconnect puzzle encapsulates these anomalies, highlighting the weak and unstable empirical relationship between exchange rates and macroeconomic fundamentals—including capital flows (Obstfeld and Rogoff, 2000). All these anomalies relate to imperfect international integration and underscores the need for more granular, micro-founded empirical approaches to advance our understanding of international trade and financial integration.
Casting doubt on the predictions of standard open-economy macroeconomic models, this empirical literature led to an extensive wave of theoretical research. However, lacking explanations from the reduced form evidence on why capital flows and exchange rates deviate so markedly from the predictions of standard models, theory work proceeded in the dark, without proper measures of market incompleteness and segmentation. For example, the recent influential framework by Itskhoki and Mukhin (2021)—which is capable of addressing a broad range of exchange rate puzzles theoretically—relies on the presence of significant home bias and financial shocks that affect the Uncovered Interest Parity (UIP) wedge, a central condition of international arbitrage. Yet, the magnitude of home bias required by their model exceeds empirical estimates derived from earlier macro-level reduced-form studies. In their setting, financial market equilibrium determines the expected path of exchange rates, while the response of exchange rates to shocks is governed by goods market equilibrium. Critically, without shocks to the UIP wedge, home bias and market incompleteness alone are insufficient to explain observed deviations in the behavior of capital flows and exchange rates. Thus, we not only need improved micro-data based estimates of home bias but also a clearer identification of the sources of financial shocks, their relation to economic fundamentals, and their role in driving international arbitrage deviations observed in the data.
My research agenda seeks to provide systematic empirical evidence from micro data on these issues. I do so by linking the quantity of foreign investment to its pricing at the investor–borrower–transaction level, incorporating both investor expectations and the frictions faced by investors and borrowers. The use of granular data allows me to observe foreign bank lending to domestic firms at the bank–firm level, bond issuance by domestic firms in foreign markets at the issuance level, and cross-border portfolio investments—such as sovereign bond purchases or equity holdings—at the investor–transaction level. State-of-the-art micro-level datasets further provide detailed information on the currency denomination, maturity structure, and identity of foreign investors involved in these transactions. This approach enables the identification of both borrower- and investor-side frictions and allows me to analyze how borrower- and investor-specific constraints both shape investor expectations and asset pricing across international financial markets.
Why is such granular measurement of international financial integration essential? While improved econometric identification—such as holding macroeconomic fundamentals and aggregate shocks constant—is a valuable benefit, it is not the primary motivation. The core reason lies in the insights gained from heterogeneity. Just as the heterogeneous agents revolution in macroeconomics (see recent Moll’s Review of Economic Dynamics newsletter article) transformed our understanding of impact of heterogeneity on aggregate dynamics, incorporating heterogeneity in international finance is critical for uncovering the micro-level drivers behind observed macroeconomic patterns in capital flows and exchange rates. If specific types of borrowers or investors disproportionately influence these aggregate outcomes, then failing to account for this heterogeneity hampers our ability to assess the macroeconomic consequences of financial integration on growth, welfare, and policy transmission. Understanding the transmission of international shocks is inseparable from identifying the underlying reasons why financial markets remain incomplete and segmented. Are these frictions primarily financial or informational in nature? Or are they driven by policy and regulatory barriers? Do international or domestic financial intermediaries play the dominant role in segmenting markets? Without precise, micro-level measurement of these frictions and barriers, theoretical models risk being mis-specified, and policy prescriptions based on them may be misguided. Given that multiple models with differing microeconomic assumptions can fit the same macro data, identifying the correct frictions at the micro level is not optional—it is imperative.
A salient example of the tension between theory and empirical reality lies in the predictions of standard models regarding the growth and welfare gains from increased international integration. Canonical frameworks (e.g., Obstfeld and Rogoff, 1995) suggest that as countries transition from autarky to trade and financial openness, they benefit through multiple channels: improved risk sharing, more stable consumption, and higher investment and output growth via more efficient allocation of global capital. Yet, the reality tells a more nuanced story. Many countries that have liberalized capital accounts have experienced frequent financial crises, heightened macroeconomic volatility, and disappointing growth outcomes. One potential explanation, as I show in my joint work with my colleagues, is that without sufficiently developed domestic financial institutions, recipient countries are unable to channel foreign capital effectively toward productive investment, limiting the expected growth benefits (Alfaro, Chanda, Kalemli-Özcan, and Sayek, 2004). Elsewhere, the puzzling positive correlation between growth and the current account balance (capital flows) has been interpreted to mean either that foreign investors tend to invest in low-growth countries or, conversely, that foreign capital inflows hinder growth (Gourinchas and Jeanne, 2013). However, the same correlation can also be driven by the composition of capital flows: specifically, by the dominance of public flows (e.g., sovereign borrowing) rather than private sector investment as shown by both Aguiar and Amador, 2011 and my work (Alfaro, Kalemli-Özcan, and Volosovych, 2014). In this view, high-growth countries simultaneously export capital via public savings while attracting private inflows, potentially benefiting from the latter and being constrained by the former. Conversely, low-growth countries may rely heavily on sovereign borrowing, which could itself be a drag on growth. These competing interpretations highlight the need for disaggregated analysis of capital flow composition to understand the complex relationship between international financial integration and growth.
Reduced-form macroeconomic evidence also cannot address the fundamental question lurking behind: why do foreign private investors allocate capital to foreign assets at all if such investments yield no excess returns? One possible explanation is that returns are low because many recipient countries are not capital scarce (Gourinchas and Jeanne, 2006). Alternatively, as I show in my work, weak institutional quality may deter investment, suppressing inflows despite potentially high returns (Alfaro, Kalemli-Özcan, and Volosovych, 2008). Regardless of the reason, the observed paucity of capital flows implies limited international investment, insufficient risk sharing, and persistent home bias. Yet the deeper puzzle still remains: if returns are low and risks are prohibitively high, why does any international investment take place? Why is home bias incomplete in the data?
My recent research sheds light on this paradox by focusing on the role of risk and risk premia enabling higher returns. I show that international investments can offer high expected returns precisely because they carry a high risk premium. At the same time, the presence of frictions—both financial and informational—combined with investor risk aversion, can suppress the volume of these investments. Disentangling the relative importance of these mechanisms is challenging, as high risk premia may themselves reflect country-specific fundamentals or erratic policy regimes.
To advance this debate, my research leverages micro-level data. First, I summarize below how micro data can help resolve causality concerns in the relationship between capital flows and growth. Next, I lay out the details on how micro data can help to identify and quantify the role of risk in shaping international investment behavior. As the following sections describe, these two dimensions—the growth effects of capital flows and the pricing of international risk—are deeply intertwined and require a unified micro-founded approach.
Capital Flows and Growth: Evidence from the European Integration Experiment
A natural experiment to explore these questions is European integration. The European Union provides the closest real-world analogue to textbook models of financial and trade integration. With the launch of the euro in 1999, countries with historically high default and currency risk—such as Italy and Greece—were suddenly able to borrow in euros at interest rates similar to those of Germany, despite significant remaining differences in country fundamentals (and hence default risk) that later culminated into the European crisis. Using firm-level data on foreign investment and balance sheets across European countries, I, joint with my colleagues, show that (Gopinath, Kalemli-Özcan, Karabarbounis, and Villegas-Sanchez (2017)) while productivity growth increased in Northern countries like Germany during the first decade of integration, it declined in Southern countries like Spain. This divergence is explained by the domestic misallocation of capital inflows from North to South in countries that suffer from domestic financial frictions.
How do we explain these joint phenomena—where capital did flow from North to South, as it should, as part of integration, but South still stagnated? First and foremost, Germany invested in Spain as part of the integration process not because Spain was growing faster in 1999, but because Spain offered higher returns on capital, owing to its lower capital stock as of 1999. Once institutional barriers were removed, as part of the integration, capital flowed, as it should. This led to the convergence of sovereign borrowing rates, however, micro-level rates for firms and households did not converge, due to domestic distortions. These domestic financial frictions led to misallocation. Capital that flowed into Spain was misallocated toward large but unproductive firms, leading to a decline in aggregate productivity growth in Spain.
This paper, thanks to the use of micro data, demonstrates that patterns observed previously in macro data can reflect reverse causality: capital flows caused lower growth due to financial frictions and misallocation, rather than capital flowing to low-growth countries. The financial friction—observable only in micro data—operates through size-dependent constraints: large firms with high net worth receive more debt financing, while small firms with lower net worth are constrained. This channel helps explain macro patterns such as declining productivity and rising current account imbalances even in the presence of policy-driven trade and financial integration, that may look like full integration at the macro level.
Risk, Risk Sharing, and Equity Investment
In fact, with all the integration happened within Europe, as observed in macro data, equity market integration remains incomplete. This suggests that the frictions hindering bond investment across countries may be less important than those affecting equity flows, or that Europe’s bank-based system introduces additional informational frictions. Since financial integration encompasses multiple channels—bank lending, equity and bond investment, and FDI—we also study equity flows across European firms. In, Kalemli-Özcan, Sorensen, and Volosovych (2014), we show that foreign equity investment is there to a certain extent but it goes with both higher firm-level and aggregate volatility. Why?
We find that foreign investors are more likely to invest in firms with higher volatility and tilt their investments toward riskier projects once involved, as conjectured by the model in Obstfeld (1994). Either mechanism implies that foreign investors are more tolerant of output volatility. This tolerance arises because foreign investors are internationally diversified. To validate this, we show that domestic investors with international portfolios also invest in relatively more volatile domestic firms. The broader implication is that international financial integration facilitates risk sharing—even without direct foreign investment into domestic firms.
Foreign Currency Debt as a Financial Friction
My research, as explained above, so far suggests that even though there is a negative correlation between foreign capital and growth in macro data, this should not imply restricting foreign investment, but rather prioritizing the resolution of domestic financial frictions since international integration does increase risk sharing and growth once properly measured and identified. A similar argument applies to the vulnerabilities caused by foreign-currency debt, especially in emerging markets. The “original sin”—where sovereigns cannot borrow abroad in their own currency (Eichengreen, Hausmann, and Panizza, 2005)—has increasingly migrated from government to corporate sectors. While governments now borrow more in local currency (Du and Schreger, 2016), many firms still borrow in foreign currencies, especially in U.S. dollars as shown by Aguiar, 2005 and my work (Kalemli-Özcan, Kamil, and Villegas-Sanchez, 2016). This currency mismatch on balance sheets between foreign-currency liabilities and local-currency revenues weakens balance sheets and creates vulnerabilities, amplifying the negative effects of financial shocks as modelled in Céspedes, Chang, and Velasco, 2004.
Implications for Domestic and International Transmission
A first-order question then is as follows: Why do emerging market borrowers (firms or governments) persistently borrow in foreign currency? Why do investors prefer to lend in dollars, even they can make higher returns from lending in local currency? To answer these questions, we must understand how foreign investors price risk and how that pricing interacts with domestic frictions. Micro data is crucial for this identification. Maggiori, Neiman, and Schreger (2020) show that even advanced-country investors prefer lending in their own currency or in dollars (global reserve currency). Using Turkish administrative data from 2003–2013—representative of the median emerging market during that time—I show that (together with my colleagues, Di Giovanni, Kalemli-Özcan, Ulu, and Baskaya, 2022) persistent domestic Uncovered Interest Parity (UIP) deviations lead firms to borrow in foreign currency as borrowing in foreign currency (dollar or euros) is cheaper. My work identifies that UIP fails domestically, by documenting domestic firms borrowing in foreign currency from local banks face lower interest rates than domestic firms borrowing in local currency from the same local banks, even when we restrict this relation to the same firm and same bank, that is within a firm-bank pair. There is a direct implication of this result for international transmission of risk shocks since this UIP wedge comoves strongly with global risk sentiment (VIX).
A key implication here is in terms of U.S. monetary policy spillovers. I study this in Kalemli-Özcan (2019), that demonstrates that U.S. monetary policy spillovers disproportionately affect emerging markets due to their effect in altering global risk sentiment and credit spreads which shows up in the UIP wedge of emerging markets. Domestic monetary policy is often ineffective in offsetting this, given weak transmission from policy rates to market interest rates. Interestingly, theoretical models (e.g., Mendoza, 2010; Bianchi, 2011) emphasize collateral constraints in such transmission, my work suggests that changes in external borrowing costs are more central in shaping macro outcomes in emerging markets. However, collateral still matters, especially when it is earnings based, even in the U.S. In, Caglio, Darst, and Kalemli-Özcan (2021), we show that small firms, which rely more on earnings-based collateral, are more responsive to monetary policy. Larger firms are less sensitive, even if risky, due to their better access to finance. This size-dependent constraint introduces strong procyclicality: small but leveraged firms struggle to recover from shocks if earnings remain subdued.
Takeaways
The current state of theoretical research in international macro is another form of “theory ahead of measurement» as stated by late Robert Lucas. He used this idea to critique the state of macroeconomic research in the mid-20th century, particularly large-scale econometric models that lacked strong microeconomic foundations. As theory often advances faster than the empirical tools needed to test or validate it, Lucas’s broader criticism of empirical macro models focused on how agents’ behavior might change in response to policy shifts—famously leading to the Lucas Critique. I argue that in international macro and finance, this argument is reversed: we are now in a position with several state-of-the-art models with strong micro foundations, however, none of those micro foundations are measured and tested in the data. This is a must to guide the theory further and inform policy making.
My research shows that micro-level data is critical in this endeavor. It plays a crucial role in identifying the frictions underlying incomplete international financial integration and market segmentation, and in understanding how these frictions shape the pricing of risk and deviations from international arbitrage. Risk pricing, in turn, influences equilibrium outcomes—such as firm- and country-level borrowing from domestic and foreign lenders—as well as the currency and maturity structure of debt. All this will lead to a better understanding of the relation between, or lack there-of, capital flows, exchange rates, and economic growth. My research highlights the importance of time-varying risk premium in this context as an important driver of imperfect international trade and economic integration, leading to lower growth and welfare. Several fundamental questions remain—most notably, what drives the risk premia? Financial frictions, policy uncertainty, limited commitment, monetary policy credibility, and more complex interactions are likely to be central, but further work is needed to disentangle their respective roles.
References
Aguiar, Mark, “Investment, devaluation, and foreign currency exposure: The case of Mexico,” Journal of Development Economics, 2005, 78 (1), 95–113.
_ and Manuel Amador, “Growth in the Shadow of Expropriation,” The Quarterly Journal of Economics, 2011, 126 (2), 651–697.
Alfaro, Laura, Areendam Chanda, Şebnem Kalemli-Özcan, and Selin Sayek, “FDI and economic growth: the role of local financial markets,” Journal of International Economics, 2004, 64 (1), 89–112.
_ , Şebnem Kalemli-Özcan, and Vadym Volosovych, “Why doesn’t capital flow from rich to poor countries? An empirical investigation,” The Review of Economics and Statistics, 2008, 90 (2), 347–368.
Alfaro, Şebnem Kalemli-Özcan, and Vadym Volosovych, “Sovereigns, Upstream Capital Flows, Global Imbalances,” Journal of the European Economic Association, 2014, 12 (5), 1240–1284.
Backus, David K and Gregor W Smith, “Consumption and real exchange rates in dynamic economies with non-traded goods,” Journal of International Economics, 1993, 35 (3-4), 297–316.
Bianchi, Javier, “Overborrowing and systemic externalities in the business cycle,” American Economic Review, 2011, 101 (7), 3400–3426.
Bruno, Valentina and Hyun Song Shin, “Capital flows and the risk-taking channel of monetary policy,” Journal of Monetary Economics, 2015, 71, 119–132.
Caglio, Cecilia, Matthew Darst, and Şebnem Kalemli-Özcan, “Collateral Heterogeneity and Monetary Policy Transmission: Evidence from Loans to SMEs and Large Firms,” NBER working paper, 2021, (w28685).
Céspedes, Luis Felipe, Roberto Chang, and Andres Velasco, “Balance sheets and exchange rate policy,” American Economic Review, 2004, 94 (4), 1183–1193.
Cole, Harold L and Maurice Obstfeld, “Commodity trade and international risk sharing: How much do financial markets matter?,” Journal of Monetary Economics, 1991, 28 (1), 3–24.
Coval, Joshua D and Tobias J Moskowitz, “Home bias at home: Local equity preference in domestic portfolios,” The Journal of Finance, 1999, 54 (6), 2045–2073.
Du, Wenxin and Jesse Schreger, “Local currency sovereign risk,” The Journal of Finance, 2016, 71 (3), 1027–1070.
Eichengreen, Barry, Ricardo Hausmann, and Ugo Panizza, “The pain of original sin,” Other people’s money: Debt denomination and financial instability in emerging market economies, 2005, pp. 13–47.
Feldstein, Martin and Charles Horioka, “Domestic saving and international capital flows,” The Economic Journal, 1980, 90 (358), 314–329.
French, Kenneth R and James M Poterba, “Were Japanese stock prices too high?,” Journal of Financial Economics, 1991, 29 (2), 337–363.
Gabaix, Xavier and Matteo Maggiori, “International liquidity and exchange rate dynamics,” The Quarterly Journal of Economics, 2015, 130 (3), 1369–1420.
Giovanni, Julian Di, Şebnem Kalemli-Özcan, Mehmet Fatih Ulu, and Yusuf Soner Baskaya, “International spillovers and local credit cycles,” The Review of Economic Studies, 2022, 89 (2), 733– 773.
Gopinath, Gita, Şebnem Kalemli-Özcan, Loukas Karabarbounis, and Carolina Villegas- Sanchez, “Capital allocation and productivity in South Europe,” The Quarterly Journal of Economics, 2017, 132 (4), 1915–1967.
Gourinchas, Pierre-Olivier and Olivier Jeanne, “The elusive gains from international financial integration,” The Review of Economic Studies, 2006, 73 (3), 715–741.
_ and _ , “Capital flows to developing countries: The allocation puzzle,” Review of Economic Studies, 2013, 80 (4), 1484–1515.
Itskhoki, Oleg and Dmitry Mukhin, “Exchange rate disconnect in general equilibrium,” Journal of Political Economy, 2021, 129 (8), 2183–2232.
Kalemli-Özcan, Şebnem, “US monetary policy and international risk spillovers,” Technical Report, Proceedings – Economic Policy Symposium – Jackson Hole. 2019.
_ , Bent Sorensen, and Vadym Volosovych, “Deep Financial Integration and Macroeconomic Volatility,” Journal of the European Economic Association, 2014, 12 (6), 1558–1585.
_ , Herman Kamil, and Carolina Villegas-Sanchez, “What hinders investment in the aftermath of financial crises: Insolvent firms or illiquid banks?,” Review of Economics and Statistics, 2016, 98 (4), 756–769.
Lucas, Robert E., “Why Doesn’t Capital Flow from Rich to Poor Countries?,” The American Economic Review, 1990, 80 (2), 92–96.
Maggiori, Matteo, Brent Neiman, and Jesse Schreger, “International currencies and capital allocation,” Journal of Political Economy, 2020, 128 (6), 2019–2066.
Mendoza, Enrique G, “Sudden stops, financial crises, and leverage,” American Economic Review, 2010, 100 (5), 1941–1966.
Obstfeld, Maurice, “Risk-Taking, Global Diversification, and Growth,” American Economic Review, 1994, 84 (5), 1310–1329.
_ , “International macroeconomics: beyond the Mundell-Fleming model,” Technical Report, National Bureau of Economic Research Cambridge, Mass., USA 2001.
_ and Kenneth Rogoff, “Exchange rate dynamics redux,” Journal of Political Economy, 1995, 103 (3), 624–660.
_ and _ , “The six major puzzles in international macroeconomics: is there a common cause?,” NBER macroeconomics annual, 2000, 15, 339–390.
Rey, Hélène, “Dilemma not trilemma: the global financial cycle and monetary policy independence,” Technical Report, Proceedings – Economic Policy Symposium – Jackson Hole. 2013.
