The ominous facts are well known – the strongest predictors of financial crises are domestic credit booms and external debts (Reinhart and Rogoff 2011). In emerging markets, credit booms are generally preceded by large capital inflows (Reinhart and Reinhart 2010). Many high-growth emerging markets have been receiving capital inflows for the last five years as the developed economies have been attending to their wounds from the Global Financial Crisis. Now the tide is reversing. Emerging markets are experiencing slowdowns in growth and widening current-account deficits.
Should we then expect a new crisis in emerging markets in the wake of the Fed’s decision to pull back from its bond-buying programme? Such ‘tapering’ has the potential to trigger large capital outflows from the emerging markets. Should we expect investors to dump a single country’s assets first – an action that might then spread contagiously to others, as happened during Asian Crisis in the late 1990s? The answer depends on the existing economic and political vulnerabilities in each country. To put it another way: “Have the emerging markets been naughty or nice during the boom years?”
The common narrative has been such that most of the emerging countries have already ‘emerged’. They have solved their problems of high external debt and inflexible exchange rates, put their macro houses in order, and proven themselves by their resilience during the Global Financial Crisis. Indeed, macroeconomic policies have been much better than in the previous decade. But have such countries really ‘emerged’ from weak rule of law, corruption, and political instability? If not, they are still vulnerable to a confidence crisis where political instability can easily trigger a financial crisis.
The case of Turkey
A case in point is Turkey, which might very well be the next ‘sudden stop’. During the last decade, under the AK Party government, Turkey enjoyed political stability and resilient growth that averaged 5% annually. Turkey earned praise from financial markets and economists alike until recently, when it was put on Morgan Stanley’s so-called ‘fragile five’ list together with Brazil, India, Indonesia, and South Africa. All of these countries have experienced a slowdown in growth and other vulnerabilities. Turkey seems to be the most fragile with the biggest current account deficit – this stood at 7.5% of GDP as of November 2013, and is mostly financed by short-term volatile capital flows.
The situation became more alarming with the major corruption scandal that is currently unfolding. Allegedly, several high-level officials, sons of ministers, top businessman, and mayors were involved in extensive graft, in which development projects which acted as a catalyst to Istanbul’s and other urban centres’ growth were shadowed by large bribes. The AK Party (AK means clean and pure in Turkish) has made transparent governance and the fight against corruption their key motto. According to the prime minister, this is the key reason for the political stability and growth that they have provided to the country during the last decade.
The last decade was a political bubble in the sense that no major structural, institutional, or legal reform had occurred (Kalemli-Ozcan 2013). It would not be surprising if there has also been a financial bubble. The prime minister’s initial reaction to the corruption charges did not suggest that there had been any institutional reform in the past decade, especially in judicial and legal institutions. The government sacked 70 top police chiefs and passed a law that restricts prosecutors from conducting probes without approval. Journalists are blocked from accessing the police, and are being asked not to report anything that would compromise the investigation in any way.
Political bubbles are likely to form during periods of political stability and strong government support. Research has shown that, in the past, emerging-economy crises are preceded by a strong increase in government support (Herrera et al. 2013). On average, in emerging markets, government stability increases by more than 50% in the five years prior to major crisis events. The AK Party’s support over the last five years was around 50%. One explanation for such a pattern is that in emerging economies, people assign a larger importance to the government in driving economic performance. This increases the incentives of governments to delay reforms, since reforms will reduce popularity, and with less popularity the chances of being re-elected are lower. Hence political bubbles and financial bubbles can go together during periods of stability and growth. Domestic credit growth helps to stimulate the economy in the short run, but can also lead to a bubble in asset prices, mostly likely in the real estate sector – especially in the absence of any domestic structural reform.
Liability dollarisation in Turkey’s corporate sector
In the case of Turkey, the jury is still out on whether there is a financial bubble. Bubbles are hard to predict. The Central Bank of Turkey, in expectation of a bubble, has introduced several measures to rein in credit growth during the past few years. The key vulnerability in Turkey’s case is not only the magnitude of the credit boom, or the sector in which it is concentrated (construction), but also the nature of this debt – namely, high levels of liability dollarisation in the corporate sector. It is well known that in emerging markets, firms tend to borrow in dollars, mostly short-term, and thus leave themselves vulnerable to exchange rate fluctuations and a rollover crisis. The Central Bank’s latest financial stability report states that as of 2013, foreign-currency debt is 58% of the total. The report does make the case that most of this debt is of reasonably long maturity, but it is not clear what this means in the absence of a system of flow-of-funds accounting, where the maturity and currency structure of corporate-sector debt can be tracked easily. Another Central Bank study (Alp 2013) shows that as of 2011, the firms in the construction sector (at the centre of the corruption probe) have debt in foreign currency as high as 70% of their total debt, whereas manufacturing firms’ share of such foreign-currency debt is around 50%. By its nature, most of the construction sector’s debt is short term to finance working capital needs.
As I have shown together with my co-authors (and others in the literature), during the Asian and Latin American crises, such levels of liability dollarisation played a significant role – even under flexible exchange rates – due to balance sheet effects (Kalemli-Ozcan et al. 2012). In the event of a sudden loss in the value of the domestic currency, a debt that is in foreign currency on the liability side of the balance sheet cannot be paid or rolled over when the asset side of the same balance sheet is in domestic currency, resulting in a decline in the firm’s net worth. I believe this is the central premise behind the Central Bank’s extensive intervention to prop up the value of the Turkish lira. The lira has depreciated almost 15% since the end of June, reaching a record low this last week amid the scandal – hitting 2.2 against the dollar. In the absence of a significant amount of reserves (in the order of magnitude that Asian central banks have), a critical question is how far the Central Bank of Turkey can defend the lira, especially in the light of the recent political scandal that gets bigger every day.
Alp, B (2013), “The Effect of Liability Dollarization on Turkish Real Sector Firms”, Central Bank of Turkey.
Herrera, H, G Ordonez and C Trebesch (2013), “Political Booms and Financial Crises”, mimeo.
Kalemli-Ozcan, S (2013), “Why Turkey is Rebelling”, Project Syndicate.
Kalemli-Ozcan, S, H Kamil, and C Villegas-Sanchez (2012), “What Hinders Investment in the Aftermath of Financial Crises: Insolvent Firms or Illiquid Banks?”, NBER Working Paper 16528.
Reinhart, C and K Rogoff (2011), This Time Is Different: Eight Centuries of Financial Folly, Princeton University Press.
Reinhart, C and V Reinhart (2010), “Capital Flow Bonanzas: An Encompassing View of the Past and Present”, NBER Working Paper 14321.