A Decomposition of International Capital Flows


We propose a method to break down capital flows into portfolio growth and portfolio reallocation components and apply it to data on US equity and bond outflows. The decomposition is part of an integrated approach that decomposes purchases of any asset into portfolio growth and reallocation components. US equity and bond outflows depend not just on portfolio growth and the reallocation between US and foreign equity and bonds, but also on reallocation decisions higher up on the decision tree. This includes reallocation between portfolio and non-portfolio assets and between equity and bonds. We also consider the decomposition of US equity and bond outflows to individual foreign countries. The data shed light on the importance of the various components in accounting for capital flows over both the short and long run.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6


  1. 1.

    Our focus on equity and bond outflows connects with the recent empirical literature on international capital flows that has increasingly focused on different types of capital flows (foreign direct investment, portfolio equity, portfolio debt, banking flows, others flows) as opposed to total capital inflows and outflows. Examples are Avdjiev et al. (2017), Broner et al. (2013), Bruno and Shin (2015), Cerutti et al. (2019), Cerutti et al. (2017a), Cerutti et al. (2017b), Fratzscher (2012), Koepke (2015), Milesi-Ferretti and Tille (2011) and Sarno et al. (2016).

  2. 2.

    While our approach is entirely empirical, it relates to the recent literature that has emphasized portfolio allocation decisions as drivers of capital flows in open economy DSGE models. Examples are Albuquerque et al. (2007, 2009), Bacchetta and van Wincoop (2017), Davis and van Wincoop (2018), Devereux and Sutherland (2007, 2010), Didier and Lowenkron (2012), Dou and Verdelhan (2017), Evans and Hnatkovska (2014), Gabaix and Maggiori (2015), Hau and Rey (2006), Hnatkovska (2010) and Tille and van Wincoop (2010a, b, 2014).

  3. 3.

    We thank Frank Warnock for helping us understand the international position and flow data.

  4. 4.

    The BT data are available at http://www.federalreserve.gov/pubs/ifdp/2007/910/ticdata.zip.

  5. 5.

    The surveys of claims before 2004 took place 3/31/1994, 12/31/1997, 12/31/2001 and 12/31/2003.

  6. 6.

    See the Appendix of BT for details.

  7. 7.

    See Warnock and Warnock (2009) for a related approach. We thank Frank Warnock for explaining to us in detail the approach used to compute restated flows in Curcuru et al. (2011).

  8. 8.

    If we take the part of the sample that starts in 2012, the standard deviation of quarterly equity outflows, scaled by the position during the previous quarter, is almost twice as big on average across countries when using the BJ data than the data computed with the BT methodology. For some countries, like Poland and Germany, the volatility is four times as large.

  9. 9.

    We have also computed all results reported in Sect. 4 when the sample ends in December, 2011, using the BT position data that are available online without extending their sample further. We find that this makes virtually no difference for the results. In addition, we have also computed the results for the decomposition of equity and bond flows to individual countries when combining the BT data through 2011 with the BJ data after that. The results are reported in Online Appendix and are very similar to those reported in Table 4 when using the BT data for the entire sample.

  10. 10.

    This is equal to line 1 of Table B.100 of the Financial Accounts (total assets), minus line 31 (liabilities).

  11. 11.

    We obtain the data from Table B.1, market value of domestic corporations.

  12. 12.

    See BIS, Debt Securities Statistics.

  13. 13.

    We do not need data on the return of non-portfolio assets. In all decompositions we treat the reallocation between portfolio and non-portfolio assets as a residual, after computing the portfolio growth component and other reallocation components.

  14. 14.

    The latter is the All Countries and IROs row in BT, code 99996. It should be noted that for equity these returns are highly correlated with the MSCI. The average correlation in our sample is 0.9968.

  15. 15.

    The latter includes Treasury securities, government agency bonds, mortgage-backed bonds, corporate bonds and a small amount of foreign bonds traded in USA. Before November, 2008, it was called the Lehman Aggregate Bond Index.

  16. 16.

    Although they use a somewhat different approach, Guo and Jin (2009) also find that the “composition” effect (related to various types of reallocation) is much more important than the “growth effect” at high frequencies. They do not consider the role of portfolio growth in the long run.

  17. 17.

    The share that US investors invest in Foreign equity cannot continue to rise indefinitely. Specifically, once barriers to foreign investment stop falling further, the trend increase of the Foreign equity share will end. At that point capital outflows due to Home–Foreign reallocation will no longer be positive on average. But the effect on portfolio growth will last as investors continue to invest a larger share of their savings abroad.

  18. 18.

    We remove one extreme outlier for equity and 4 outliers for bonds. For equity the outlier is Russia. For bonds the outliers are Serbia and Montenegro, Poland, Panama and Ghana.


  1. Ahmed, Shaghil, Stephanie E. Curcuru, Francis E. Warnock and Andrei Zlate. 2017. Decomposing international portfolio flows, working paper.

  2. Albuquerque, Rui, Gregory H. Bauer, and Martin Schneider. 2007. International equity flows and returns: A quantitative equilibrium approach. Review of Economic Studies 74(1): 1–30.

    Article  Google Scholar 

  3. Albuquerque, Rui, Gregory H. Bauer, and Martin Schneider. 2009. Global private information in international equity markets. Journal of Financial Economics 94(1): 18–46.

    Article  Google Scholar 

  4. Avdjiev, Stefan, Bryan Hardy, Sebnem Kalemli-Ozcan and Luis Serven. 2017. Gross capital flows by banks, Corporates and Sovereigns, working paper.

  5. Bacchetta, Philippe and Eric van Wincoop. 2017. Gradual portfolio adjustment: Implications for global equity portfolios and returns. working paper, University of Virginia.

  6. Bertaut, Carol C. and Ruth Judson. 2014. Estimating U.S. cross-border securities positions: New data and new methods. Board of Governors of the Federal Reserve System (U.S.), International Finance Discussion Paper 1113.

  7. Bertaut, Carol C. and Ralph W. Tryon 2007. Monthly estimates of U.S. cross-border securities positions. Board of Governors of the Federal Reserve System (U.S.), International Finance Discussion Paper 910.

  8. Broner, Fernando, Tatiana Didier, Aitor Erce, and Sergio L. Schmukler. 2013. Gross capital flows: Dynamics and crises. Journal of Monetary Economics 60: 113–133.

    Article  Google Scholar 

  9. Bruno, Valentina, and Hyun Song Shin. 2015. Cross-border banking and global liquidity. Review of Economic Studies 82: 535–564.

    Article  Google Scholar 

  10. Burger, John D., Francis E. Warnock, and Veronica C. Warnock. 2018. Benchmarking portfolio flows. IMF Economic Review 66 (3): 527–563.

    Article  Google Scholar 

  11. Cerutti, Eugenio, Stijn Claessens, and Lev Ratnovski. 2017. Global liquidity and cross-border bank flows. Economic Policy 32 (89): 81–125.

    Article  Google Scholar 

  12. Cerutti, Eugenio, Stijn Claessens and Andrew K. Rose 2017. How important is the global financial cycle? evidence from capital flows. NBER Working Paper 23699.

  13. Cerutti, Eugenio, Stijn Claessens, and Damien Puy. 2019. Push factors and capital flows to emerging markets: Why knowing your lender matters more than variables. Journal of International Economics. (forthcoming)

  14. Curcuru, Stephanie E., Tomas Dvorak, and Francis E. Warnock. 2008. Cross-border return differentials. Quarterly Journal of Economics 123(4): 1495–1530.

    Article  Google Scholar 

  15. Curcuru, Stephanie E., Charles P. Thomas, Francis E. Warnock, and Jon Wongswan. 2011. US international equity investment and past and prospective returns. American Economic Review 101: 3440–3455.

    Article  Google Scholar 

  16. Curcuru, Stephanie E., Tomas Dvorak, and Francis E. Warnock. 2010. Decomposing the U.S. external returns differential. Journal of International Economics 80(1): 22–32.

    Article  Google Scholar 

  17. Davis, J. Scott, Giorgio Valente and Eric van Wincoop. 2019. The global capital flows cycle: Impact on gross and net capital flows. NBER Working Paper 25721.

  18. Davis, J.Scott, and Eric van Wincoop. 2018. Globalization and the increasing correlation between capital inflows and outflows. Journal of Monetary Economics 100: 83–100.

    Article  Google Scholar 

  19. Devereux, Michael B., and Alan Sutherland. 2007. Monetary policy and portfolio choice in an open economy macro model. Journal of the European Economic Association 5: 491–499.

    Article  Google Scholar 

  20. Devereux, Michael B., and Alan Sutherland. 2010. Country portfolio dynamics. Journal of Economic Dynamics and Control 34: 1325–1342.

    Article  Google Scholar 

  21. Didier, Tatiana, and Alexandre Lowenkron. 2012. The current account as a dynamic portfolio choice problem. Journal of The Japanese and International Economies 26: 518–541.

    Article  Google Scholar 

  22. Dou, Winston W. and Adrien Verdelhan. 2017. The volatility of international capital flows and foreign assets. working paper.

  23. Evans, Martin D.D., and Viktoria V. Hnatkovska. 2014. International capital flows, returns and world financial integration. Journal of International Economics 92(1): 14–33.

    Article  Google Scholar 

  24. Fratzscher, Marcel. 2012. Capital flows, push versus pull factors and the global financial crisis. Journal of International Economics 88: 341–356.

    Article  Google Scholar 

  25. Gabaix, Xavier, and Matteo Maggiori. 2015. International liquidity and exchange rate dynamics. The Quarterly Journal of Economics 130(3): 1369–1420.

    Article  Google Scholar 

  26. Guo, Kai, and Keyu Jin. 2009. Composition and growth effects of the current account: A synthesized portfolio view. Journal of International Economics 79: 31–41.

    Article  Google Scholar 

  27. Hau, Harald, and Helene Rey. 2006. Exchange rates, equity prices, and capital flows. Review of Financial Studies 19(1): 273–317.

    Article  Google Scholar 

  28. Hnatkovska, Viktoria V. 2010. Home bias and high turnover: Dynamic portfolio choice with incomplete markets. Journal of International Economics 80(1): 113–128.

    Article  Google Scholar 

  29. Koepke, Robin. 2015. What drives capital flows to emerging markets: A survey of the empirical literature. IIF Working Paper.

  30. Milesi-Ferretti, and Cedric Tille. 2011. The great retrenchment: International capital flows during the global financial crisis. Economic Policy 26(66): 289–346.

    Article  Google Scholar 

  31. Sarno, Lucio, Ilias Tsiakas, and Barbara Ulloa. 2016. What drives international portfolio flows? Journal of International Money and Finance 60: 53–72.

    Article  Google Scholar 

  32. Tille, Cédric, and Eric van Wincoop. 2010a. International capital flows. Journal of International Economics 80(2): 157–175.

    Article  Google Scholar 

  33. Tille, Cédric, and Eric van Wincoop. 2010b. A new perspective on the “new rule” for the current account. Journal of International Economics 80(1): 89–99.

    Article  Google Scholar 

  34. Tille, Cédric, and Eric van Wincoop. 2014. International capital flows under dispersed private information. Journal of International Economics 93(1): 31–49.

    Article  Google Scholar 

  35. Warnock, Francis E., and Veronica Cacdac Warnock. 2009. International capital flows and U.S. interest rates. Journal of International Money and Finance 28: 903–919.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Eric van Wincoop.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

We gratefully acknowledge financial support from the Bankard Fund for Political Economy. We thank Frank Warnock for helping us with various data aspects of the Project.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 124 KB)

Supplementary material 2 (zip 2612 KB)


Appendix A: Equity Outflow Decomposition

In this Appendix, we derive the decompositions (15) and (18) of equity outflows. First consider the decomposition in Sect. 2.2. Differentiating (14), we have

$$\begin{aligned} Q_{t}^{e,F} \Delta X_{t+1}^{e,F}+X_{t}^{e,F} \Delta Q_{t+1}^{e,F}= k_{t}^{e,F} \Delta A_{t+1}+\left( \frac{\Delta k_{t+1}^{p}}{k_{t}^{p}}+\frac{\Delta k_{t+1}^{e|p}}{k_{t}^{e|p}}+\frac{\Delta k_{t+1}^{F|e}}{k_{t}^{F|e}}\right) k_{t}^{e,F} A_{t} \end{aligned}$$

Using (3) we have

$$\begin{aligned} \Delta A_{t+1}=(1-k_{t}^{p}) \left( \frac{\Delta Q_{t+1}^{np}}{Q_{t}^{np}}-\frac{\Delta Q_{t+1}^{p}}{Q_{t}^{p}} \right) A_{t}+ \frac{\Delta Q_{t+1}^{p}}{Q_{t}^{p}} A_{t} +S_{t+1} \end{aligned}$$

Using (5)–(7), this can be written as

$$\begin{aligned} \Delta A_{t+1}= & {} (1-k_{t}^{p}) \left( \frac{\Delta Q_{t+1}^{np}}{Q_{t}^{np}}-\frac{\Delta Q_{t+1}^{p}}{Q_{t}^{p}} \right) A_{t}+ (1-k_{t}^{e|p}) \left( \frac{\Delta Q_{t+1}^{b}}{Q_{t}^{b}} -\frac{\Delta Q_{t+1}^{e}}{Q_{t}^{e}} \right) A_{t} \nonumber \\&+\, (1- k_{t}^{F|e}) \left( \frac{\Delta Q_{t+1}^{e,H}}{Q_{t}^{e,H}}- \frac{\Delta Q_{t+1}^{e,F}}{Q_{t}^{e,F}} \right) A_{t}+ \frac{\Delta Q_{t+1}^{e,F}}{Q_{t}^{e,F}}+S_{t+1} \end{aligned}$$

We can use the definition of changes in passive portfolio shares to write this as

$$\begin{aligned} \Delta A_{t+1}=-\frac{\Delta {\tilde{k}}_{t+1}^{p}}{k_{t}^{p}} A_{t}-\frac{\Delta {\tilde{k}}_{t+1}^{e|p}}{k_{t}^{e|p}} A_{t}- \frac{\Delta {\tilde{k}}_{t+1}^{F|e}}{k_{t}^{F|e}}A_{t}+ \frac{\Delta Q_{t+1}^{e,F}}{Q_{t}^{e,F}}A_{t}+S_{t+1} \end{aligned}$$

Substituting this into (A.1), using that \(X_{t}^{e,F} \Delta Q_{t+1}^{e,F}=k_{t}^{e,F} A_{t} \Delta Q_{t+1}^{e,F}/Q_{t}^{e,F}\), we obtain the decomposition of equity outflows \(Q_{t}^{e,F} \Delta X_{t+1}^{e,F}\) in (15).

We can similarly derive the decomposition of equity outflows to individual countries. Differentiating (17), we have

$$\begin{aligned}&Q_{t}^{e,F,n} \Delta X_{t+1}^{e,F,n}+X_{t}^{e,F,n} \Delta Q_{t+1}^{e,F,n}= k_{t}^{e,F,n} \Delta A_{t+1}\nonumber \\&\quad + \left( \frac{\Delta k_{t+1}^{p}}{k_{t}^{p}}+\frac{\Delta k_{t+1}^{e|p}}{k_{t}^{e|p}}+\frac{\Delta k_{t+1}^{F|e}}{k_{t}^{F|e}}+\frac{\Delta k_{t+1}^{n|e,F}}{k_{t}^{n|e,F}} \right) k_{t}^{e,F,n} A_{t} \end{aligned}$$

We can write \(X_{t}^{e,F,n} \Delta Q_{t+1}^{e,F,n}=k_{t}^{e,F,n} A_{t} \Delta Q_{t+1}^{e,F,n}/Q_{t}^{e,F,n}\). From (A.4) we have

$$\begin{aligned} k_{t}^{e,F,n} \Delta A_{t+1}-k_{t}^{e,F,n} A_{t} \frac{\Delta Q_{t+1}^{e,F,n}}{Q_{t}^{e,F,n}}= & {} -\frac{\Delta {\tilde{k}}_{t+1}^{p}}{k_{t}^{p}}k_{t}^{e,F,n} A_{t}-\frac{\Delta {\tilde{k}}_{t+1}^{e|p}}{k_{t}^{e|p}} k_{t}^{e,F,n}A_{t}\nonumber \\&-\, \frac{\Delta {\tilde{k}}_{t+1}^{F|e}}{k_{t}^{F|e}}k_{t}^{e,F,n}A_{t}- \frac{\Delta {\tilde{k}}_{t+1}^{n|e,F}}{k_{t}^{n|e,F}}k_{t}^{e,F,n}A_{t}+k_{t}^{e,F,n} S_{t+1} \end{aligned}$$

Substituting this into (A.5) gives the decomposition of equity outflows \(Q_{t}^{e,F,n} \Delta X_{t+1}^{e,F,n}\) to country n shown in (18).

Appendix B: Portfolio Reallocation and Portfolio Rebalancing

As pointed out in Sect. 2.4, the relationship between portfolio rebalancing and portfolio reallocation depends on the nature of the shocks and parameters. A couple of examples will help illustrate the point. First consider an increase in the relative supply of country n equity. This will lower its relative price. There will then be a demand shift toward country n for two reasons: The lower price raises its expected return and the lower price creates a desire to rebalance the portfolio toward country n. Which of these dominate depends on how sensitive portfolios are to expected returns, which depends for example on risk aversion. If portfolios are not very sensitive to expected returns, almost all of the reallocation is associated with portfolio rebalancing. If portfolios are very sensitive to expected returns, equilibrium prices change very little and very little of the portfolio reallocation is associated with rebalancing.

A different type of shock is one where asset supplies do not change, but instead there is a portfolio demand shift toward country n by all investors, for example as a result of reduced risk of country n equity. Since asset supplies do not change, in equilibrium portfolio demand will not change either. There will be no portfolio reallocation toward or away from country n. But this is because rebalancing away from country n (due to a higher equity price) is offset by a higher equilibrium portfolio share invested in country n (supply has not changed and the price has increased).

As a final example, consider a portfolio shift toward foreign equity and away from domestic equity by investors in all countries. For each country, foreigners want to hold more of its equity and domestic investors less. If total equity demand (by domestic and foreign investors) does not change in any country, equity prices do not change either. In this case there is positive reallocation toward foreign equity, but no portfolio rebalancing as equity prices do not change.

The reported results on the relationship between portfolio reallocation and rebalancing vary by type of reallocation as well. Portfolio–non-portfolio reallocation is more closely associated with rebalancing than is the case for the other reallocation components. When two assets are not close substitutes, investors are less willing to change their allocation between the assets. A change in relative asset supplies will then lead to greater rebalancing. Portfolio rebalancing plays a less important role when agents for various reasons wish to change their allocation across the assets a lot. Since portfolio and non-portfolio assets are very different in nature (non-portfolio assets include for example real estate and bank deposits), they are not close substitutes, explaining the larger importance of rebalancing.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Meng, G., van Wincoop, E. A Decomposition of International Capital Flows. IMF Econ Rev 68, 362–389 (2020). https://doi.org/10.1057/s41308-019-00094-0

Download citation