Emerging Markets and the Systemic Sudden Stop

A year ago, I argued here about Latin America (October 14) “that, as long as there would be no meltdown of the global economy, the predominance of sustainable stock-flow interactions at both fiscal accounts and balance-of-payments in most countries of the region would preclude new ‘sudden stop’ episodes, differently from previous experiences”. Well, regardless of whether what has happened in the global economy in the aftermath of the bankruptcy of Lehman Brothers can be called a “meltdown” and qualify as my caveat, the fact is that no emerging market – in Latin America and the Caribbean (LAC) or elsewhere – has escaped from the truly “systemic sudden stop” that followed. Nevertheless, let me use the Brazilian case as an illustration of the key role played by idiosyncratic, country-specific conditions regarding fiscal and external sustainability.

A Systemic Sudden Stop hits emerging markets

What has taken place in emerging markets since mid-September fits clearly in what Calvo, Izquierdo & Mejia defined as a “Systemic Sudden Stop”, namely episodes of capital account reversals and sharp increases in aggregate spreads with a systemic – and, hence, largely exogenous – origin, i.e. “crises that are highly likely to be associated with an external trigger that is systemic in nature” (p.10).

As the global financial crisis threatened to become a global collapse, it spilled over dramatically to emerging markets. Global deleveraging and redemption pressures led to massive fire-sales of emerging-market and other risk assets. The intensity of the sudden outflow of portfolio capital, accompanied by a sudden freeze in all credit lines (including trade credit), can be gauged indirectly through the generalized spike in risk spreads (Chart 1 – upper left), the speed of unwinding of emerging market debt and equity mutual funds (Chart 1 – bottom), as well as the recent drought of international issuance (Chart 1 – upper right). Such a hasty run to the exit was triggered by external reasons, such as the cover of losses and margin calls elsewhere, foreign banks preserving liquidity, or simply as a response to the systemically heightened risk aversion.

Chart 1Emerging Markets Undergo a Systemic Sudden Stop


Source: IDB Research Department (Northwest); IFF (Northeast); IMF (bottom)

In the case of LAC, portfolio equity outflows and carry trade unwinding accelerated the pace at which they were taking place in previous months. As a result, not only domestic markets in the region dived, but also most local currencies underwent a sharp depreciation (Chart 2), while EMBI spreads went to the roof (Chart 3). There has been some relief since then, as the global financial system seems to have overcome the acute panic stage.

Chart 2 Exchange Rates in LAC


Source: IDB Research Department

Chart 3EMBI Spreads in LAC


Source: IDB Research Department

Brazil and the Systemic Sudden Stop

In Brazil, a halt in the domestic interbank market and of credit in general followed rapidly the systemic sudden stop. Domestic credit conditions deteriorated as a peculiar form of the credit freeze happening at the core-advanced economies. Notwithstanding the small proportion of foreign sources in the total of both banking and non-banking funding, a spurt of uncertainty regarding local corporate health both at banking and non-banking sectors was sparked after the sudden drought of foreign finance and local-currency devaluation. News of an unexpected vulnerability to exchange-rate depreciation by corporations and smaller banks due to exposure through derivatives immediately led to a local version of doubts about hidden “toxic assets” and financially fragile balance sheets.

The Brazilian public sector had availed itself of the current-account surpluses and foreign-capital bonanza of the last few years to reduce its foreign debt and retire dollar-denominated domestic debt, up to the point of acquiring a negative dollar-exposure in its accounts. Indeed, the recent exchange-rate depreciation has even contributed to a shrinking public-debt-to-GDP ratio. Conversely, at the private sector, confidence on a strong local currency had become so entrenched as to lead for instance some corporations to accept providing dollar put options to banks in exchange for lower funding costs. The fact is that the reversal of the theretofore-downward dollar trend was followed by a surprising revelation of – realized and unrealized – corporate losses and a domestic generalized credit squeeze.

The response by monetary authorities has been twofold, on both foreign exchange and domestic credit fronts. As of November 6, the Central Bank has sold US$ 5.2 bn (2.6% of international reserves) in the spot market; combined with derivative sales of US$25.8 bn through currency swaps. Additionally, temporary dollar liquidity has been provided through repo agreements both at the spot market ($4.8 bn) and abroad ($3.3 bn). The war chest for interventions received the confidence boost given by the inclusion of the Brazilian central bank in the U.S. Federal Reserve’s network of currency swap lines. Trade credit lines have returned to a level equivalent to half of the one prior to mid-September, whereas the exchange rate receded from the peak and has hovered around non-dramatic levels. The latter has mitigated fears of a deeper corporate financial stress, as well as of rising credit risks for those counterparty banks that were at the other side of corporations in structures of currency derivatives.

At the domestic credit side, besides extending its rediscount policies, the Central Bank has eased on its long-held stiff reserve requirements, in a series of moves that according to estimates may end up liberating an amount of liquidity potentially superior to 5.7% of GDP and 5.6% of total bank assets. The government has also announced the intention of resorting to public-sector majority-owned banks to fill in the blanks in the cases of credit to agriculture, automobiles and others, as well as to acquire partnership shares in Brazilian-based companies.

The phase of panic and financial absolute freeze seems to have ceased, nonetheless leading to several consequences. Preliminary figures for domestic credit in October point to a steady reversal of the long path of expansion previously in course. Most leading indicators of industrial production and demand, as well as business and household confidence surveys, are also suggesting a sharp economic deceleration in the last quarter of the year. The fusion of two large domestic private banks (Itaú and Unibanco) can become a first move of a forthcoming wave of mergers and restructuring in the Brazilian financial sector.

The macroeconomic landscape for 2009 has worsened, and GDP growth projections have been trimmed to the range of 2%-3%, a substantial slowdown after a rhythm expected to end up above 5.2% in 2008. The ongoing global deleveraging is still to continue affecting local asset markets and the balance-of-payment capital account. Together with weaker foreign demand for exports, softer commodity prices and less favorable terms of trade, that will imply a tighter external environment in the near future. Domestic absorption was running above potential GDP growth prior to the credit crunch and doubts remain on whether the current consumption and investment deceleration will be enough to counteract the inflationary effects of the now prevailing more depreciated levels of the exchange rate.

On the other hand, the Brazilian government still retains an arsenal of monetary, foreign reserves, and fiscal and quasi-fiscal instruments to be used if domestic demand decelerates too deeply. This is a major upside derived from policies of positive public primary surpluses, inflation targeting and reserve accumulation that were followed prior and along the now-ended period of favorable external environment.


Some Emerging Markets Are More Equal than Others

Calvo, Izquierdo & Mejia show that whether or not an exogenous financial trigger “develops into a full-fledged Sudden Stop depends also on country-specific variables” (p.2). In their specific empirical exercise, they focus on “Domestic Liability Dollarization, DLD, i.e., foreign-exchange denominated domestic debts towards the domestic banking system, as a share of GDP”. In addition, they use “the current account deficit as a share of absorption of tradable goods” in order to capture factors that can lead to sudden large real exchange-rate depreciations, which in turn tend to generate distress on balance sheets, output and repayment capacity (p.2-3).

They “also include as an explanatory variable a measure of financial integration with the rest of the world” (p.8). By “keeping all other variables at their sample means”, the authors detect a non-linear relationship between portfolio integration and the probability of a Sudden Stop, as depicted in Chart 4 (for which I thank my IDB colleague Eduardo Lora).

Chart 4


Indeed, the recent financial market jitters at core-advanced economies have surprisingly hit hard those countries in the region that have been singled out as cases of success in financial integration (Brazil, Mexico, Chile, Peru, Colombia and others). Not only gross inflows matter more than net ones during moments of panic and undifferentiated sell-off of assets, but also the newly-discovered perils associated with financial innovations came with that integration – as illustrated in the Brazilian case.

On the other hand, Chart 4 should not be taken too far, as it must be complemented with information on differentiated country-specific capacities of response to shocks. Levels of international reserves, as well as degrees of maneuver at both fiscal and monetary policies determine the extent to which the country is able to withstand either aggregate or key-sector specific shocks – again as illustrated in the Brazilian case. This is likely one of the factors lying behind different intensities of spread hikes and recoveries during and after the Systemic Sudden Stop.

The increasing degree of systemic relevance of large emerging markets also matters. Chart 5 exhibits the international claims by BIS reporting banks on some emerging countries, as of last June. Figures suggest a huge potential of negative feedbacks between banks in advanced countries and emerging markets, thereby self-reinforcing distress. That bidirectional possibility, as well as the fact that national rescue packages at core countries installed an additional factor of pull-out from emerging markets certainly led central banks of the former countries to consider incorporating some of the latter into their currency swap networks – once more illustrated in the Brazilian case.

Chart 5 International Claims by BIS Reporting Banks on Emerging Markets

image005.jpgBy Maturity, June 2008

There have been some traces of differentiation in currency markets between groups of countries inside and outside the “magic circle” of the Fed’s network of currency swaps (Guha & Beattie, FT, October 31). But also the easier and less heavily-conditional access to IMF resources for countries deemed to be implementing sound policies will also tend to create another layer of differentiation among emerging economies.

Truth is that it is too early for any safe call on where emerging economies will head to as the current global crisis unfolds, either individually or as a whole. However, the quality of domestic policies will certainly be among the differentiating factors. Paraphrasing an important flagship study of the IDB Research Department, “some of what glitters may end up being gold after all”.

9 Responses to "Emerging Markets and the Systemic Sudden Stop"

  1. bsetser   November 11, 2008 at 8:18 pm

    Otaviano — nice post. One question, from a reserves geek. How did you calculate that Brazil has done $5 billion in spot intervention? I haven’t looked at the November data, but I didn’t see a fall in Brazil’s reserves in October that would be consistent with that kind of intervention. There was a surprising jump in brazil’s reserves tho on one day in the middle of October — I don’t actually doubt you at all. Brazil certainly was intervening; I was just puzzled by the absence of a bigger fall in reported reserves in Oct.

  2. Anonymous   November 12, 2008 at 5:36 am

    What an article

  3. Jose Seligmann-Silva   November 12, 2008 at 1:32 pm

    Otaviano,Hard to disagree with your points!Just a smaller comment.I do not know how the probability of a sudden stop was calculated in chart 4. I assume that it was mostly based on a “backward looking” type of analysis, not taking into account that certain economies and policy makers, as would be the case of Brazil, have learned from the past and are now more prepared to face a financial crisis. If the current shock had taken place 9 years ago, the damage would already have been much worse that the one we’ve seen up to now. It is not clear in their analysis how much did they really took into account the whole complex set of particular preventive measures/policies/regulations that have kept Brazil in a safer ground for eventual external shocks.Having said that, one could not deny, that this is a very different type of shock, a crisis generated at the center of the international financial system, and a VERY BIG one, that continues to unwind with a process of deleveraging that could continue for a good period of time maintaining the stress of capital flow reversals from emerging economies as well as disorderly exchange rate depreciations. If there is a meltdown, nobody will escape.But looking forward, in the scenario of developed economies’ facing a long but U-shaped global activity, a point worth mentioning is that in addition to what have already been done, the way this crisis is being managed by emerging markets will affect how fast these economies will rebound. And in this aspect I believe that Brazil, again is in an upside position.

  4. Otaviano   November 12, 2008 at 2:54 pm

    BradI borrowed the $5 bi figure from some reports that have been tracking the Brazilian case. BNP Paribas, e.g., has a daily track report called “measures to tackle the crisis” that can be reached here in RGE. Trying to respond to your main question: volume of interventions is an official figure, one which is continuously checked by analysts by looking at Brazilian figures on changes of the monetary base. On the other hand, let’s keep in mind that reserves are a portfolio of not-so-homogenous and not-necessarily-low-risk assets, which have also displayed high volatility. Those assets are marked-to-market, and the Brazilian reserves include also things like gold. That’s why in the Brazilian case there is no one-to-one correspondence between variation of reserves and interventions. Tks OC

  5. Otaviano   November 12, 2008 at 3:00 pm

    JoseYou are right in your caveats regarding backward-looking exercises, particularly in situations so peculiar as the one we are going through. And financial integration cannot be taken as the sole dimension through which global shocks will affect emerging economies (even non financially-integrated economies will suffer the consequences of lower commodity prices). However, the degree of financial integration on the eve of the crisis has brought many (differentiating) implications among emerging markets.Anonymous: tksOC

  6. Vitoria Saddi   November 14, 2008 at 8:30 am

    Otaviano,It is a really good piece. Thanks a lot!Vic

  7. bsetser   November 15, 2008 at 8:40 am

    Otaviano — many thanks for your response; now i just need to find the data on volume of intervention!

  8. Paulo Manoel Lenz Cesar Protasio   December 1, 2008 at 9:51 am

    Prezado Vice-Presidente,Você poderia proferir palestra sobre essa sua visão, em SP, na Fecomercio, no próximo dia 12 de dezembro? Precisamos de uma mensagem bem construida, como a apresentada no artigo – e que sirva de degrau para subir a presente visão que tem que ter esperança de voltar a crescer no futuro, e que enquanto isso, no presente, pode ser orientada a olhar com foco para a infra-estrutura.Vamos nos falar?Grande abraço,Paulo Manoel Protasio

    • OC   December 1, 2008 at 1:24 pm

      hablamos, si, Almirante! Mas nao posso me afastar dos HQs ate’ o fim do ano!Abc OC