Tag Archives: short-term debt

Capital Flows, Credit Booms and Bank Crises

Studies of the impact of capital inflows have established that debt inflows can lead to bank crises (see here and here). Unlike equity, payments on debt are contractual and can not be cancelled if there is an economic downturn, which intensifies any shocks to the financial system. In the case of short-term debt, a foreign lender may decide not to roll over credit at the time when it is most needed. But recent papers have shown that foreign debt can also be a determinant of the credit booms that lead to the bank crises.

Philip Lane of Trinity College and Peter McQuade of the European Central Bank (working paper version here) looked at the relationship of domestic credit growth and capital flows in Europe during the period of 1993-2008. They suggest that financial flows can encourage more rapid credit growth by increasing the ability of domestic banks to extend loans, while also contributing to a rise in asset prices that encouraged financial activity. They found that debt flows contributed to domestic credit growth but equity flows did not. Moreover, the linkage of debt and domestic credit was strongest during the 2003-08 pre-crisis period.

Similarly, Julián Caballero of the Inter-American Development Bank (working paper here) investigated capital inflow booms, known as “bonanzas,” in emerging economies between 1973 and 2008. He reported that capital inflow bonanzas increased the incidence of bank crises. When he distinguished among foreign direct investment, portfolio equity and debt bonanzas, the results indicated that only the portfolio equity and debt bonanzas were associated with an increased likelihood of crises. More analysis revealed that the impact of increased debt was due in part to a lending boom. Caballero suggested that the capital inflows could also have increased asset prices, generating an asset bubble and an eventual collapse.

Deniz Iagan and Zhibo Tan of the IMF used both macroeconomic and micro-level firm data to examine the relationship of capital inflows and credit growth. They first examined the impact of capital inflows on aggregate credit to households and non-financial corporations in advanced and emerging market economies during the period of 1980-2011. They distinguished among FDI, portfolio and other inflows. They reported that portfolio and other inflows contributed to rises in household credit, and only the other inflows were significant for corporate credit.

Iagan and Tan also had data on firms in these countries, and sought to identify the determinants of leverage in these firms. They calculated an index, based on work done by Raghuram Rajan and Luigi Zingales (RZ), of a firm’s dependence on external financing. When they interacted the RZ indicator with the different types of capital inflows, the interactive term was always significant in the case of the other inflows, significant with portfolio flows in some specifications, and never significant in the case of FDI flows. The authors concluded that the results of the macro and firm level analyses were consistent: the composition of capital matters. In additional analysis, they found evidence consistent with the hypothesis that the capital inflows led to higher asset prices.

What can be done to insulate an economy from lending booms that may lead to bank crises? Nicolas E. Magud and Esteban R. Versperoni of the IMF and Carmen R. Reinhart of Harvard’s Kennedy School of Government (working paper here) examined whether the nature of the exchange rate regime was relevant. They found that less flexible exchange rate regimes are associated with increases in bank credit and a higher share of foreign currency in bank credit. On the other hand, the exchange rate regime had no impact of the size of the capital inflows. The authors of the Bank for International Settlements 85th Annual Report 2014/15, however, wrote that the insulation property of flexible exchange rates is “overstated.” An exchange rate appreciation can raise the value of firms with debt denominated in foreign currency, which increases the availability of credit.

How can regulators lower the danger of more bank crises due to debt inflows? Magud, Reinhart and Vesperoni suggest the use of macroprudential measures that affect the incentives to borrow in a foreign currency, such as currency-dependent liquidity requirements. But Caballero warns that capital controls on debt inflows may be insufficient if portfolio equity flows also contribute to lending booms that result in banking crises.

These research papers find that domestic asset prices respond to international financial flows. This makes it harder to insulate the domestic financial markets from foreign markets, and leaves these markets vulnerable to spillovers from changes in foreign conditions. The emerging markets already face downturns in their markets, and the combination of increased global volatility with a rise in the costs of servicing the dollar-denominated debt of corporations in emerging markets if the Federal Reserve raises interest rates will only add to their burdens.

Apples and Naranjas

The Economist has published its indicator of vulnerability to a “capital freeze.” (An earlier version published on September 7 was revised.) The ranking for 26 emerging markets is based on each country’s current account balance as a percent of its GDP, its short-term external debt and debt repayments relative to foreign exchange reserves and sovereign wealth fund assets, the growth rate of credit to the private sector, and the Chinn-Ito index of financial openness. The ten countries rated as most at risk are Turkey, Romania, Poland, Mexico, Colombia, Peru, Argentina, Indonesia and Chile.

Not surprisingly, there has been some pushback from officials of the countries at the top of the rankings. Mauricio Cárdenas, Colombia’s minister of finance, defends his country’s economic reputation in a letter. The minister claims that those who devised the rankings ignored the sources of vulnerability to a sudden stop, including the source of financing for the current account.  Mr. Cárdenas points out that Colombia’s current account is “fully financed by foreign direct investment instead of short-term capital flows.”

Does he have a case? We calculated Colombia’s current account/GDP and FDI/GDP ratios over the last three years (2010-12), and compared them with other Latin American economies on the index:

% Current Account/GDP FDI/GDP
Argentina -0.05 2.34
Brazil -2.24 2.92
Mexico -0.70 1.67
Venezuela 4.28 0.76
Colombia -3.05 3.54

Score one for the minister: the current account deficits of the last three years were indeed offset by inflows of FDI.  Game, set, match for Colombia?

Perhaps not. The size of the current account deficits, one of the components of the capital-freeze index, stands out. More importantly, the minister is mixing flows and stocks. The FDI inflows create FDI liabilities that are not easily reversed. But the country’s short-term external debt is the source of vulnerability to a sudden stop. Nervous lenders can simply cease renewing lines of credit or other credit facilities, and domestic borrowers will be cut off from funding. Reversals of short-term external debt were features of the Mexican, East Asian, Russian and Brazilian crises of the 1990s.

The debt data for the five South American countries in 2011 show why Brazil is rated as less risky than Colombia and the other nations. These countries are more vulnerable to a change in sentiment by foreign lenders. Venezuela does not appear in the top ten on the capital-freexe index in part because it is not as financially open as the others, and thus less exposed.

% External Debt/GNI Short-term Debt/External Debt
Argentina 26.35 14.53
Brazil 16.64 10.42
Mexico 25.20 17.88
Venezuela 21.82 24.56
Colombia 24.31 14.06

Colombia’s finance minister has justification to be proud that his country attracts sufficient FDI to finance its current account deficits. And there is no reason to expect that those flows will cease. But the country’s external debt liabilities, the result of past borrowing, are the source of potential hazard.