Log in / Register

Home arrow Business & Finance arrow Treasury Finance and Development Banking
< Prev   CONTENTS   Next >

4.2.3 Credit

From what we have mentioned in the previous section about emerging markets being characterized by uncertainty and risk, it is only natural that we touch upon the issue of credit risk.

In Section 3.2 we have shown how the credit risk of a borrower can be hedged through credit default swaps. As a consequence, at first glance, the CDS rates are the best indicators of the riskiness of a certain borrower. Let us add immediately that they are perfect indicators in a risk-neutral framework in the sense that through them we can imply risk-neutral survival probabilities. In other frameworks, other indicators are preferred (e.g., credit ratings published by rating agencies).

In emerging markets, CDS levels are considerably higher than in developed countries, and in Table 4.1 we show the one-year and five-year CDS rates for a selection of emerging markets. For comparison we also show the CDS rates for two developed countries (the United States and Germany) and two corporates (Ford Motor Company and Sony Corporation). To offer a common comparison tool, these are CDS rates for contracts offering protection against borrowing in USD, not in local currency. We mentioned in Section 3.2.3 that CDS levels for non-USD denominated debt would be different.

TABLE 4.1 Credit spread of selected emerging markets sovereign shown against a few developed markets' sovereign and corporate for comparison, as of March 4, 2013.


One-year CDS spread (USD)

Five-year CDS spread (USD)

United States of America, Government of



Germany, Federal Republic of



Israel, State of



South Africa, Republic of



Czech Republic, Government of the



Turkey, Republic of



Philippines, Republic of



Kazakhstan, Republic of



Ukraine, Republic of



Argentina, Republic of



Ford Motor Company



Sony Corporation



The data shown in Table 4.1 is very interesting because it shows that not all emerging markets are equally risky (and it gives an idea of why it is so difficult to lump them together), but the same data should come with a caveat. From CDS rates we can imply survival probabilities, in other words default probabilities,[1] but CDS rates are driven by many considerations that go beyond pure credit. There could be liquidity issues–the less frequently a CDS is traded, the more false could be the credit information embedded in it; there could be speculation involved in the trading of that CDS. Let us not forget that a CDS is a derivative with a link to an underlying (the credit of a borrower) but, as we discussed in Section 3.1.1, sometimes this link becomes broken and the two values (the derivative's value and the underlying's value[2]) move in very different directions. The amount of debt involved (which could or could not match the principal of the CDS contracts) can have a distorting effect through leverage.

As an example of the above, let us look at the most striking piece of data from the table: Argentina's one-year CDS level at 7, 027.76 bps. One can show (see Appendix B) that the probability of default (expressed as one minus the survival probability) can be approximated, particularly for a short maturity T, by

where CDST is the rate for a CDS with maturity T and R is the assumed recovery rate. Using a recovery of 24%, for Argentina this implies a probability of default greater than 80% within a year. This means that we are considering Argentina 170 times more likely than Kazakhstan to default within a year. While for some this might seem reasonable (some argue that Argentina has never fully left the 2002 default behind and is drifting more and more toward uncertainty), these numbers are difficult to grasp. Similarly hard to believe is that the one-year CDS spread for the Philippines and the Czech Republic is lower than the U.S. Treasury. This difficulty also originates from the fact that above we have presented a risk-neutral probability of default (we shall introduce risk neutrality in Section 7.2.1), which should not be confused with an actuarial probability of default. Rating agencies publish probabilities of default based on historical data (and therefore not implied from CDS levels), which are considerably lower.

Despite the caveats and the surprises that CDS levels can lead to, they are very useful indicators, and it is helpful to compare emerging market levels to the levels of developed countries to comprehend the different levels of magnitude of the risk involved.

Another interesting aspect of the above data is the fact that Argentina presents a so-called inverted curve, that is, the short end (represented by the one-year point) of the CDS rate's term structure is higher than the medium- long end (represented by the five-year point). This is counterintuitive since we would expect to pay more for longer-dated protection. This happens in dire situations (and almost always in emerging markets), the market thinking being roughly “if country X survives the immediate difficult situation, there is a greater chance that it might survive in the longer term as well.”

Figure 4.4 shows how for Ukraine and Kazakhstan the shape of the curve is usually normal (i.e., one-year point lower than the five-year point) except during the peak of the 2007 to 2009 financial crisis when the points switched positions.

Figure 4.4 is also useful to show what kinds of magnitudes the CDS rates of emerging markets can reach during a crisis.

  • [1] Since, of course, if there is a probability ST that a borrower is solvent, there is a probability 1 – ST that it has defaulted.
  • [2] Which, in the case of credit, unfortunately is not visible unlike, say, equities where one can easily see quoted stock and option prices. It is true that one could consider the bond as the underlying, but a bond still gives an implied view of credit as opposed to a share price, which is the underlying itself.
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science