After the 2008-09 global financial crisis, economists were criticized for not predicting its coming. This charge was not totally justified, as there were some who were concerned about the run-up in asset prices. Robert Schiller of Yale, for example, had warned that housing prices had escalated to unsustainable levels. But the looming debt crisis in the emerging market economies has been foreseen by many, although the particular trigger—a pandemic—was not.
Last year the World Bank released Global Waves of Debt: Causes and Consequences, written by M. Ayhan Kose, Peter Nagle, Franziska Ohnsorge and Naotaka Sugawara. The authors examined a wave of debt buildup that began in 2010. By 2018 total debt in the emerging markets and developing economies (EMDE) had risen by 54 percentage points to 168% of GDP. Much of this increase reflected a rise in corporate debt in China, but even excluding China debt reached a near-record level of 107% of GDP in the remaining countries.
The book’s authors compare the recent rise in the EMDE’s debt to other waves of debt accumulation during the last fifty years. These include the debt issued by governments in the 1970s and 1980s, particularly in Latin America; a second wave from 1990 until the early 2000s that reflected borrowing by banks and firms in East Asia and governments in Europe and Central Asia; and a third run-up in private borrowing via bank loans in Europe and Central Asia in the early 2000s. All these previous waves ended in some form of crisis that adversely affected economic growth.
While the most recent increase in debt shares some features with the previous waves such as low global interest rates, the report’s authors state that it has been “…larger, faster, and more broad-based than in the three previous waves…” The sources of credit shifted away from global banks to the capital markets and regional banks. The buildup included a rise in government debt, particularly among commodity-exporting countries, as well as private debt. China’s private debt rise accounted for about four-fifths of the increase in private EMDE debt during this period. External debt rose, particularly in the EMDEs excluding China, and much of these liabilities were denominated in foreign currency.
The World Bank’s economists report that about half of all episodes of rapid debt accumulation in the EMDEs have been associated with financial crises. They (with Wee Chian Koh) further explore this subject in a recent World Bank Policy Research Paper, “Debt and Financial Crises.” They identify 256 episodes of rapid government debt accumulation and 263 episodes of rapid private debt accumulation in 100 EMDEs over the period of 1970-2018. They test their effect upon the occurrence of bank, sovereign debt and currency crises in an econometric model, and find that such accumulations do increase the likelihood of such crises. An increase of government debt of 30 percentage points of GDP raised the probability of a debt crisis to 2% from 1.4% in the absence of such a build-up, and of a currency crisis to 6.6% from 4.1%. Similarly, a 15% of GDP rise in private debt doubled the probability of a bank crisis to 4.8% if there were no accumulation, and of a currency crisis to 7.5% from 3.9%. (For earlier analyses of the impact of external debt on the occurrence of bank crises see here and here.)
Kristin J. Forbes of MIT and Francis E. Warnock of the University of Virginia’s Darden Business School looked at episodes of extreme capital flows in the period since the global financial crisis (GFC) in a recent NBER Working Paper, “Capital Flows Waves—or Ripples? Extreme Capital Flow Movements Since the Crisis.” They update the results reported in their 2012 Journal of International Economics paper, in which they distinguished between surges, stops, flights and retrenchments. They reported that before the GFC global risk, global growth and regional contagion were associated with extreme capital flow episodes, while domestic factors were less important.
Forbes and Warnock update their data base in the new paper. They report that has been a lower incidence of extreme capital flow episodes since 2009 in their sample of 58 advanced and emerging market economies, and such episodes occur more as “ripples” than “waves.” They also find that as in the past the majority of episodes of extreme capital flows were debt-led. When they distinguish between bank versus portfolio debt, their results suggest a substantially larger role for bank flows in driving extreme capital flows.
Forbes and Warnock also repeat their earlier analysis of the determinants of extreme capital flows using data from the post-crisis period. They find less evidence of significant relationships of the global variables with the extreme capital flows. Global risk is significant only in the stop and retrenchment episodes, and contagion is significantly associated only with surges. They suggest that these results may reflect changes in the post-crisis global financial system, such as greater use of unconventional tools of monetary policy, as well as increased volatility in commodity prices.
Corporations can respond to crises by changing how and where they raise funds. Juan J. Cortina, Tatiana Didier and Sergio L. Schmukler of the World Bank analyze these responses in another World Bank Policy Research Working paper, “Global Corporate Debt During Crises: Implications of Switching Borrowing across Markets.” They point out that firms can obtain funds either via bank syndicated lending or bonds, and they can borrow in international or domestic markets. They use data on 56,826 firms in advanced and emerging market economies with 183,732 issuances during the period 1991-2014, and focus on borrowing during the GFC and domestic banking crises. They point out that the total amounts of bonds and syndicated loans issued during this period increased almost 27-fold in the emerging market economies versus more than 7 times in the advanced economies.
Cortina, Didier and Schmukler found that the issuance of bonds relative to syndicated loans increased during the GFC by 9 percentage points from a baseline of 52% in the emerging markets, and by 6 percentage points in the advanced economies from a baseline probability of 28%. There was also an increase in the use of domestic debt markets relative to international ones during the GFC, particularly by emerging economy firms. During domestic banking crises, on the other hand, firms turned to the use of bonds in the international markets. When the authors used firm-level data, they found that this switching was done by larger firms.
The authors also report that the debt instruments have different characteristics. For example, the emerging market firms obtained smaller amounts of funds with bonds as compared to bank syndicated loans. Moreover, the debt of firms in emerging markets in international markets was more likely to be denominated in foreign currency, as opposed to the use of domestic currency in domestic markets.
Cortina, Didier and Schmukler also investigated how these characteristics changed during the GFC and domestic bank crises. While the volume of bond financing increased during the GFC relative to the pre-crisis years, syndicated bank loan financing fell, and these amounts in the emerging market economies fully compensated each other. In the advanced economies, on the other hand, total debt financing fell.
The global pandemic is disrupting all financial markets and institutions. The situation of banks in the advanced economies is stronger than it was during the GFC (but this could change), and the Federal Reserve is supporting the flow of credit. But the emerging markets corporations and governments that face falling exports, currency depreciations and enormous health expenditures will find it difficult to service their debt. Kristalina Georgieva, managing director of the IMF, has announced that the Fund will come to the assistance of these economies, and next week’s meeting of the IMF will address their needs. The fact that alarm bells about debt in emerging markets had been sounding will be of little comfort to those who have to deal with the collapse in financial flows.
