Skip to main content
Log in

International Reserves: Precautionary Versus Mercantilist Views, Theory and Evidence

  • Research article
  • Published:
Open Economies Review Aims and scope Submit manuscript


This paper compares the importance of precautionary and mercantilist motives in the hoarding of international reserves by developing countries. Overall, empirical results support precautionary motives; in particular, a more liberal capital account regime increases international reserves. Theoretically, large precautionary demand for international reserves arises as a self-insurance to avoid costly liquidation of long-term projects when the economy is susceptible to sudden stops. The welfare gain from the optimal management of international reserves is of a first-order magnitude, reducing the welfare cost of liquidity shocks from a first-order to a second-order magnitude.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others


  1. See Calvo (1998), Calvo and Mendoza (2000) and Edwards (2004) for further discussion on sudden stops of short-term capital flows.

  2. See Kaminsky and Reinhart (1999) and Hutchison and Noy (2005) for further discussion on the output costs associated with sudden stops.

  3. This figure presents capital account liberalization indices as ratios to the 1980 values of industrial and developing countries, respectively.

  4. The lower panel may convey the impression that developing countries have more open capital accounts than industrial countries. This reflects, of course, the use of 1980 as a base—the average financial openness of developing countries remains below the financial openness of industrial countries. Rodrik (2006) presents a similar figure of patterns of reserves-to-GDP ratio for emerging and developed countries, calculated over a longer period since 1960 for a slightly different grouping of countries. He concludes that emerging markets over-invested in the costly strategy of reserve accumulation and under-invested in capital-account management policies to reduce their short-term foreign liabilities.

  5. See Bryant (1980); Diamond and Dybvig (1983) and Prisman et al. (1986) for earlier literature dealing with optimal reserves (liquidity) policy in a closed economy.

  6. The precautionary demand modeled in this paper supplements the precautionary demand stemming from fiscal considerations. For example, one may argue that the prospect of unification of North and South Korea [or a conflict in the worst-case scenario] may explain part of the hoarding of international reserves by Korea. Yet, we may qualify this argument by noting that one expects the US and the OECD countries to provide the credit needed to finance the unification (or the conflict). This argument, however, does not extend to the case of a sudden stop and capital flight. As the 1997 crisis illustrated, external finance at times of sudden stops is not forthcoming without stringent conditions and is frequently limited due to moral hazard considerations.

  7. The recent history of Argentina provided a vivid illustration of the limited ability to diversify away liquidity shocks. In the mid-1990s Argentina negotiated contingent commercial credit lines in an attempt to provide external insurance against liquidity shocks. These lines, however, dried up as Argentina approached the crisis.

  8. See Kravis (1984) for a classic reference on PPP, and Samuelson (1994) for the apt expression “Penn effect.”

  9. We regard this as a robust measure of exchange rate misalignment, rather than the best or complete measure which hardly exists. Frankel (2006) also used this measure of misalignment, interpreting the Penn effect as caused by the Balassa-Samuelson effect.

  10. This variable turns out to be closely correlated with the variation of capital flows. In an auxiliary regression of this variable on the volatility of capital flow, the two variables are strongly correlated. The volatility of capital flow was measured by the standard deviation of capital flow over a moving window of 5 years.

  11. The results of regressions for post-1975 sample are available from the authors upon request. Main results remain identical, despite limited data availability.

  12. When the estimation included the exchange rate regime index of Reinhart and Rogoff (2004)—with and without the exchange rate volatility—no statistically robust relationship was uncovered between the index and reserve holdings, at times producing opposite-signed coefficients.

  13. Unlike other variables, the terms of trade is a time-series index whose value is based on a particular base year (1995 in our data). Hence, the cross-country dispersion in it captures the cross-country dispersion in changes of the terms of trade relative to the base year.

  14. In regressions of Table 4 that includes Hong Kong and Singapore, Singapore easily dominates all others in the magnitude of its country-specific constant. It should come as no surprise for a country with reserves of more than 80 percent of GDP.

  15. Our model follows the tradition of Bryant (1980) or Diamond and Dybvig (1983) in that the source of liquidity shock lies with the lender, rather than the borrower (Holmstrom and Tirole 1998). However, our model assumes away the market equilibrium among lenders (be it the risk of runs or the difficulty of the decentralized provision of liquidity). Abstracting from the question whether market-based liquidity insurance is available, we focus on the implication of large adjustment cost—including but not restricted to the liquidation cost—on the demand for reserves as self-insurance. In a similar vein, no distinction is made between the private sector and the monetary authorities which maintain the stock of international reserves.

  16. The possibility that the outcome of investment is not large enough to meet the promised rate of return is discussed later. To preview, this possibility does not affect the main conclusion of our analysis, because of the assumption of risk neutrality.

  17. This follows from the observation that \( \frac{{d\;E{\left[ \Pi \right]}_{{\left| {R = 0} \right.}} }} {{d\;z_{0} }} \cong \frac{{\partial \;E{\left[ \Pi \right]}_{{\left| {R = 0} \right.}} }} {{\partial \;D}}\frac{{d\;D}} {{d\;z_{0} }} + \frac{{\partial \;E{\left[ \Pi \right]}_{{\left| {R = 0} \right.}} }} {{\partial \;z_{0} }} = \frac{{\partial \;E{\left[ \Pi \right]}_{{\left| {R = 0} \right.}} }} {{\partial \;z_{0} }} \) (recall that the FOC determining deposits is \( \frac{{\partial \;E{\left[ \Pi \right]}_{{\left| {R = 0} \right.}} }} {{\partial \;D}} = 0 \)).

  18. With more than two states of nature, R would be preset at the ex-ante efficient level, providing full insurance for liquidity shocks below z*, and partial insurance above. While there is no way to insure complete stabilization, one expects large welfare gain from setting R at the ex-ante efficient level relative to the case of R = 0.

  19. Recalling Eq. 2, higher R reduces the stock of capital in states of nature where Z < R by ΔR, but increases the stock of capital in states of nature where Z > R by θΔR.

  20. Note that for \( {{\left( {1 + \theta z * } \right)}} \mathord{\left/ {\vphantom {{{\left( {1 + \theta z * } \right)}} {{\left( {1 + \theta } \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {1 + \theta } \right)}} = z \), output is zero, and the bank would default. Hence, a sufficient condition for the limited liability constraint to bind is \( {{\left( {1 + \theta z * } \right)}} \mathord{\left/ {\vphantom {{{\left( {1 + \theta z * } \right)}} {{\left( {1 + \theta } \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {1 + \theta } \right)}} < \lambda \). Eq. 11 implies, however, that \( \widetilde{z} < \lambda \), and the limited liability constraint may bind even if \( {{\left( {1 + \theta z * } \right)}} \mathord{\left/ {\vphantom {{{\left( {1 + \theta z * } \right)}} {{\left( {1 + \theta } \right)}}}} \right. \kern-\nulldelimiterspace} {{\left( {1 + \theta } \right)}} > \lambda \).

  21. The conventional closed-economy assumption is  = 1. The case where  < 1 can capture the presence of repatriation risk, where the banks pays foreign creditors only a fraction of output for \( z > \widetilde{z} \), or the efficiency loss associated with debt restructuring.

  22. This result holds because we assumed the absence of enforcement and monitoring costs, and that all agents are risk neutral.


  • Aizenman J, Marion NP (2003) The high demand for international reserves in the far east: what’s going on? J Jpn Int Econ 17(3):370–400

    Article  Google Scholar 

  • Aizenman J, Marion NP (2004) International reserves holdings with sovereign risk and costly tax collection. Econ J 114:569–591

    Article  Google Scholar 

  • Aizenman J, Lee Y, Rhee Y (2004) International reserves management and capital mobility in a volatile world: policy considerations and a case study of Korea. J Jpn Int Econ

  • Ben-Bassat A, Gottlieb D (1992) Optimal international reserves and sovereign risk. J Int Econ 33:345–362

    Article  Google Scholar 

  • Bryant R (1980) A model of reserves, bank runs, and deposit insurance. J Bank Financ 4:335–344

    Article  Google Scholar 

  • Calvo G (1998) Capital flows and capital-market crises: the simple economics of sudden stops. J Appl Econ 1(1):35–54

    Google Scholar 

  • Calvo G, Mendoza E (2000) Contagion, globalization, and the volatility of capital flows. In: Edwards S (ed) Capital flows and the emerging economies. University Chicago Press, Chicago

    Google Scholar 

  • Diamond D, Dybvig P (1983) Bank runs, liquidity and deposit insurance. J Polit Econ 91:401–419

    Article  Google Scholar 

  • Dooley M, Folkerts-Landau D, Garber P (2003) An essay on the revived Bretton Woods System. NBER Working paper No. 9971

  • Edison H (2003) Are foreign exchange reserves in Asia too high? IMF World Economic Outlook September

  • Edwards S (1983) The demand for international reserves and exchange rate adjustments: the case of LDCs, 1964–1972. Economica 50:269–280

    Article  Google Scholar 

  • Edwards S (2004) Thirty years of current account imbalances, current account reversals, and sudden stops. IMF Staff Papers, vol 51, Special Issue

  • Edwards S (2005) Capital controls, sudden stops and current account reversals. NBER Working Paper No. 11170

  • Flood R, Marion NP (2001) Holding international reserves in an era of high capital mobility. In: Collins SM, Rodrik D (eds) Brookings trade forum 2001. Brookings Institution Press, Washington, DC

    Google Scholar 

  • Frankel J (2006) On the Yuan: the choice between adjustment under a fixed exchange rate and adjustment under a flexible rate. In: Illing G (ed) Understanding the Chinese economy. CESifo Economic Studies, Munich

  • Frenkel J (1983) International liquidity and monetary control. In: von Furstenberg GM (ed) International Money and Credit: The Policy Roles. Washington: International Monetary Fund

  • García PS, Soto CG (2004) Large hoarding of international reserves: are they worth it? manuscript, Chilean Central Bank

  • Holmstrom B, Tirole J (1998) Private and public supply of liquidity. J Polit Econ 106:1–40

    Article  Google Scholar 

  • Hutchison M, Noy I (2005) How bad are twins? Output costs of currency and banking crises. J Money Credit Bank 37(4):725–752

    Article  Google Scholar 

  • Jeanne O, Ranciere R (2006) The optimal level of international reserves for emerging market economies: formulas and applications. IMF Mimeo

  • Kaminsky G, Reinhart C (1999) The twin crises. the causes of banking and balance-of-payments problems. Am Econ Rev 89:473–500

    Article  Google Scholar 

  • King M (2006) Reform of the international monetary fund. Speech:

  • Kravis IB (1984) Comparative studies of national incomes and prices. J Econ Lit 22:1–39

    Google Scholar 

  • Lee J (2004) Insurance value of international reserves. IMF Working Paper No. 04/175

  • Prisman E, Solvin M, Sushka M (1986) A general model of the banking firm under conditions of monopoly, uncertainty and recourse. J Monet Econ 17(2):293–304

    Article  Google Scholar 

  • Reinhart C, Rogoff K (2004) The modern history of exchange rate arrangements: a reinterpretation. Q J Econ 119(1):1–48

    Article  Google Scholar 

  • Rodrik D (2006) The social cost of foreign exchange reserves. NBER Working Paper No. 11952

  • Samuelson PA (1994) Facets of Balassa-Samuelson thirty years later. Rev Int Econ 2(3):201–226

    Article  Google Scholar 

  • Skidelsky R (2000) John Maynard Keynes: Fighting for freedom, 1937–1946. Penguin Putnam, New York

Download references


We thank Hali Edison for sharing the data, and Aleksandra Markovic for research assistance in the earlier phase of the project. We thank Michael Dooley, Ann Harrison, Linda Goldberg, Pierre-Olivier Gourinchas, Maury Obstfled, Brian Pinto, Ramkishen Rajan, Andy Rose, Partha Sen, George Tavlas, Tom Willett and participants in the SERC conference (Singapore 2005), Berkeley seminar, and the FRBSF conference for their useful comments. The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF or IMF policy.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Jaewoo Lee.



1.1 Data Appendix: Definitions of the regression variables

The sources of data are the International Financial Statistics and the World Economic Outlook Database (both the IMF), World Development Indicator (World Bank), and Penn World Table.

Reserves :

international reserves holdings minus gold, measured in U.S. dollars.

R to Y :

ratio of reserves to the dollar value of nominal GDP, in percent.

Population :

log of population

Openness :

log of percent import share

EX Growth :

three-year moving average of the growth rate of real exports (log change), lagged two years in the regression.

EX Volatility :

exchange rate volatility, calculated from the monthly exchange rate against the U.S. dollar.

Income :

log of per-capita real GDP, PPP based.

Relative Income :

Income of each country relative to that of the United States.

Price Level :

national price levels (measured in U.S. dollars), obtained from the Penn World Table and the World Development Indicator.

PL Deviation :

residual from the regression of Price Level on Relative Income.

K Account :

Index of capital account liberalization, constructed by Edwards (2005).

ToT :

log of the terms of trade index.


dummy variable for the period after the Mexico crisis, applied to developing and emerging market countries.


dummy variable for the period after the Asian crisis, applied to developing and emerging market countries.


dummy variable CRMEXEM, applied only to Latin America


dummy variable CRMEXEM, applied only to Asia


dummy variable CRASIAEM, applied only to Latin America


dummy variable CRASIAEM, applied only to Asia

Regressions of Tables 2, 3, and 4 all include country-specific constant terms. The primary sample for Table 2 comprises 49 countries that include advanced and emerging-market economies as well as several major developing economies. They are Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, United States, Cyprus, Israel, Korea, Argentina, Brazil, Chile, Colombia, Czech Republic, Hungary, Indonesia, Malaysia, Mexico, Pakistan, Peru, Philippines, Poland, Russia, South Africa, Thailand, Turkey, Venezuela, Algeria, China, Croatia, Egypt, India, and Morocco.

Four countries were not included in the above sample of 49 countries for varying reasons. Luxembourg and Taiwan Province of China were excluded owing to the absence of capital account liberalization indexes. They were included in Table 3, and the first column of Table 1. Hong Kong SAR and Singapore were excluded because their reserves exceeded 40 and 80% of GDP in many sample years, respectively, constituting outliers that spuriously improve the fit of the regressions. They were included in Table 4 and the second column of Table 1.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Aizenman, J., Lee, J. International Reserves: Precautionary Versus Mercantilist Views, Theory and Evidence. Open Econ Rev 18, 191–214 (2007).

Download citation

  • Published:

  • Issue Date:

  • DOI:


JEL Classification