During the recent period of low global interest rates, foreign currency debt issued by firms based in emerging markets increased dramatically (see Figure 1 and Acharya et al. 2015). But to understand the risks, if any, it’s important to know the state of corporate balance sheets and what they have actually been doing. Has leverage risen? Are firms using the funds raised to refinance, to fund real investment projects or are they acting like banks and boosting financial assets? Do firms behave differently in different countries and if so, why? And how do risks vary?
Figure 1. Stock of international debt securities of private corporations
Source: Authors’ calculations based on BIS and IMF statistics.
For a sample of 17 emerging economies, outstanding bonds issued in international markets is as high as 15% of GDP for the most financially integrated countries and about 7.5% of GDP on average — with roughly half of the total issued by financial firms and the other half non-financials (Figure 1). Leverage ratios of issuing, non-financial firms are much higher than those of non-financial, non-issuers and have risen substantially, particularly in Eastern Europe and in Latin America. For example, in Latin America the average ratio of debt to equity for issuing firms increased from 58% in 2005-07 to 71% in 2013-14 (Figure 2). The regional averages mask striking differences across countries (Figure 3), the leverage ratio of the median issuing Colombian and Turkish firm rose by 28 and 70 percentage points respectively.
Figure 2. Leverage of non-financial corporations
Source: Authors’ calculations based on Thomson-Reuters and Worldscope annual balance sheet reports. Regional averages are the simple averages of the median firm in each country. The sample is the universe of listed non-financial corporations from 17 countries active as of December 4, 2015 – some 9,748 firms, of which 3,065 are issuing firms. Issuing firms have at least one bond or note reported including through any subsidiary.
Figure 3. Leverage of non-financial corporations that issue bonds in the sample period 2005-2007 vs. 2013-2014
Source: Authors’ calculations based on Thomson-Reuters and Worldscope. Each dot or triangle represents the median of issuing firms in each of 17 countries. The vertical axis shows the average 2013-2014 debt to equity ratio, while the horizontal axis shows 2005-2007 average for the same ratio.
Much borrowing from non-financial firms appears to have been to finance financial assets rather than real investment, especially where carry trade opportunities are favorable (Bruno and Shin 2015). In other words, firms have been taking advantage of low dollar interest rates to borrow but are then investing at least some of the proceeds in financial assets including bank deposits, quite possibly in local currency. Powell (2014) argued that for four countries in Latin America, non-financial firms’ external issuance funded a significant proportion of the post global financial crisis domestic credit boom.1
In a recent paper (Caballero et al. 2015) we corroborate Bruno and Shin’s finding of a strong relation between non-financial firms’ external issuance and their financial assets. In addition, we find significant differences across countries. We test three hypotheses to explain the heterogeneity:
- Financial depth varies, so firms are attempting to complete shallow credit markets;
- International banks have retreated faster from some places than others, and non-financial firms are taking their place; and
- Non-financial firms are engaging in regulatory arbitrage and the variation across countries is thus explained by capital controls.
We find no support for the first two, but we find strong evidence consistent with the regulatory arbitrage hypothesis. Specifically, we find that non-financial corporations are more likely to act as financial intermediaries when returns from carry trades are high and there are controls on capital inflows. We suggest that where capital account transactions are highly regulated, non-financial firms may have a comparative advantage. They may issue abroad and use inter-companies loans that are normally considered as FDI to transfer financial resources across countries. Specifically, we find that when domestic interest rates exceed international ones and there are capital controls (particularly on inflows), the correlation between foreign bond issuance and liquid financial asset holding is large and significant, while when we replace capital controls with variables to capture the other hypotheses we find no significant results.
How do these results relate to risks? If firms are investing the proceeds of dollar borrowing in real assets, the risks are the traditional ones of corporate, external indebtedness. The relevant questions, then, are: Are firms borrowing in dollars but investing in local currency generating (real) assets? Are they borrowing to finance investments in risky commodity sectors? Have firms hedged currency (and commodity price) risks? What do leverage and amortisation schedules look like? Do firms have the liquidity to face potential demands in the face of currency declines or other shocks? If they don’t will they demand loans from local financial systems crowding out smaller borrowers? Are the numbers large enough to be a concern? Is there a possibility of such risks becoming systemic in nature?
Many of these questions (and the implicit risks) also apply when firms borrow to invest in financial assets, but there are additional concerns as the links to the domestic financial system may be greater. In particular, when world interest rates rise, such firms may well decide to invest elsewhere and liquidity may suddenly vanish. In previous emerging economy crises larger corporate deposits have been the first to flee. Each individual firm may have a perfectly matched liquidity position but there may still be a threat to systemic liquidity.
To assess these risks in a precise way, information requirements are daunting. Indeed, given the lack of information on firm balance-sheet currency denomination, the IMF’s Global Financial Stability Report and Acharya et al. (2015) employ the historical movement of firm equity prices to measure ‘currency betas’. This is an appealing technique to sidestep a lack of data but this technique may not pick up systemic risks at all - and it’s unclear that equity investors are better informed than the IMF’s talented researchers.2
The increase in debt of emerging market non-financial firms has been large, and firms have been behaving quite differently across countries. In some cases external debt has been issued to substitute more expensive local debt, in other cases to finance real investment, and in several countries it has been used to exploit carry trade opportunities. The types and magnitude of the risks vary. But in virtually all cases, good information on corporate currency mismatches is hard to obtain. Corporate balance sheet reporting should be revised and should embrace the currency dimension. Moreover, good information on derivative positions is required. Even if firms are hedging, we don’t know who is bearing the ultimate risks. Central clearing for derivative transactions is one way to go, or at least there should be central reporting. For countries with capital controls, where non-financial firms may be arbitraging regulations and behaving like financial intermediaries, the ties to local financial systems may be particularly strong. As bond issuance falls and corporates amortise external liabilities, financial system liquidity may be particularly stressed in those cases – central banks should be ready to take action if required.
Acharya, V, S Cecchetti, J de Gregorio, S Kalemli-Ozcan, P Lane and U Panizza (2015), “Emerging economy corporate debt: The threat to financial stability”.
Bruno, V and H S Shin, (2015) “Global dollar credit and carry trades: a firm-level analysis”, BIS Working Paper, No 510, August.
Caballero, J, U Panizza and A Powell (2015), “The Second Wave of Global Liquidity: Why are Firms acting like Financial intermediaries?”, CEPR Working Paper, 10926, November.
IMF (2015), “Vulnerabilities, Legacies, and Policy Challenges Risks Rotating to Emerging Markets”, Global Financial Stability Report, IMF.
Powell, A (2014), “Global Recovery and Monetary Normalization: Escaping a Chronicle Foretold?”, Latin American and Caribbean Macroeconomic Report, Inter-American Development Bank.
1 An econometric exercise indicated that for each US$1bn of additional external issuance from non-financial firms, domestic credit increased by about US$100mn or 10%.
2 As the regressions include the market index as well as the currency, the currency beta should be interpreted as a firm idiosyncratic beta or in other words how relatively more or less sensitive an individual firm’s equity price is to the exchange rate compared to that of the market – see IMF (2015).