Credit growth in CEECs – catching up or speeding?

I have just recently participated in a conference where a session was devoted to credit in developing markets - 4 papers on potential credit booms in CEECs (details and all 4 papers downloadable here) were presented. Generally the issue seems a relatively ‘hot topic’ though the scope for substantial research is limited both by the quality of data and the high degree of uncertainty of the tools used. Real credit growth of even 40-50% yoy has caused preoccupation of authorities and a discussion whether measures to curb lending should be introduced.

Before I go on to the results, it is a good idea to reflect why asking whether there is a credit boom taking place is of importance. First of all, Borio and Lowe (2002) claim that “one of the relatively few robust findings to emerge from the literature on leading indicators of banking crisis is that rapid domestic credit growth increases the likelihood of a problem” moreover a series of papers cited by Duenwald et al. (2004) estimate the probability of a lending boom ending in a bust to be as high as 20%.
Previous crises (Bulgaria, Croatia 1990s) have proven to have a high output cost, especially as CEECs are predominantly bank financed – bond and equity markets are still very small.
The macroeconomic risks are primarily due to the rapid increase in demand, and constitute the deterioration of the current account and inflationary pressures. Banking system risks are more an effect of improper accounting for risk, especially when competitive pressures push towards rapid increase of loans, and insufficient supervision. The idea being that in a boom, the pressure for results, may foster the build up of non-performing loans, overvalued collaterals and even connected lending.

On the other hand, credit markets in the CEECs practically did not exist 15 years ago. Moreover in some segments they hardly existed about a decade ago, thus we can say that not so long ago the starting points were approximately zero. I do not want to go in to details on the levels of real credit in these countries, (for now, very detailed figures can be found in the conference papers, perhaps later I will put up a post with some tables) but it is enough to say that despite substantial financial deepening, the levels of credit to GDP remain well below (average and individual) old EU levels, and in most CEECs below levels expected for countries with similar GDP per capita.
Thus, there is no doubt we can expect a strong catch up effect – financial deepening should cause rapid growth – the question is can the levels we observe be too fast?

The problems with assessing whether lending is growing too fast are substantial and not much different from the typical problems of boom/bubble literature. First it is difficult to say a priori that a rapid development is dangerous – most measures are criticized of being arbitrary (growth above a certain threshold, deviations from the trend component, additional “non-fundamental” growth, out of sample projections etc.) or of being only useful once its too late (non performing loans). In the context of the recent discussion of potential housing booms in many developed markets we see that even with longer series and relatively stable economies it is hard to agree whether there is a boom. Transition economies have an additional nuisance such as short series which make the discussions of trends, cycles and equilibrium levels even more problematic. Moreover even this short period is a transition period, abundant with changes (monetary regimes, structural reforms, EU entry etc.) which may potentially be breaks. Finally credit in the old EU is at very diverse levels, and in most countries growing, thus it is difficult to say what to expect of the series in CEECs (convergence to a level or further growth?).

The results of all four papers seem to point to Bulgaria as a country where credit growth is excessive with respect to other macroeconomic developments, putting strain on the current account, despite already prudent fiscal stance (fiscal surplus in 2004 and government is focusing on further tightening).
Moreover, there is some consensus on possibly credit boom developments in the Baltic states - thus it seems that countries with a rigid exchange rate (currency board, except for Latvia) are more likely to have rapid credit growth. Perhaps this is what one can expect, as the economies seem rather stable, and the currency board brings along lower interest rates. The authors argue they take care of developments in fundamentals such as changes in the interest rate (and rapid GDP growth), and moreover as rapid growth of credit in these countries occurred in both the domestic and foreign currency denominated loans, it may be that there is an mis-assessment of the risks associated with a currency board – though once the countries actually join the euro, this will not be a problem.
On the other hand, if this credit boom, by increasing internal demand, is feeding into inflation, it may be jeopardizing the EMU entry prospects. I have not seen any empirical studies of these phenomena, but even a small inflationary effect of the rapid credit growth in Lithuania, made have made a difference on the decision whether the country fulfilled the Maastricht inflation criterion or not.

Among other risks there is the large share of FX denominated loans, which can prove troublesome especially in household loans since households are known not to hedge (though some of them may have FX denominated income) but are attracted by lower real interest rates of FX loans. Banks do seem to enjoy this type of lending, as it provides the (partly illusive) notion of shifting the ER risk to the customer. Most of the highlighted countries have (large) CA deficits, and 2 of the papers that focus on the effect of credit growth on CA find it significant.
As for Ukraine – CA does not seem a problem, rather banking sector risks such as connected lending, high share of non performing loans, very fragmented banking market (over 150 banks! 2004). Moreover, this is data for 2004, and recent changes in terms of trade (the gas price adjustments by the Russian producer) may cause a rapid deterioration of the external balance, and thus add a macroeconomic risk. Moreover, in contrary to the other countries, Ukraine’s future seems shadowed by a higher degree of uncertainty (at the moment not much EU prospects) thus the macro risks should not be undervalued.

Back to the overall results, the argument that very high levels of foreign ownership of banks in CEECs drastically reduces the probability of a banking crises is not entirely correct. Although I fully agree foreign entry helped modernize the banking system, it does not, by any means, fully insulate from a credit bust - despite the fact that having if in trouble, instead of bailing out a foreign subsidiary, it may just decide to pack up the business (especially if the market is tiny relative to overall group revenue). This was visible to some extent in Argentina, but may be even more in the case of a relatively small market such as for instance Estonia.
Another issue that has not been given any attention is contagion – i.e. the idea that a banking crisis in one of the CEECs would trigger similar reactions in other. While previous crises in the 1990s (Bulgaria, Croatia) seemed to be rather contained in the countries, this time, with a set of countries labeled as having ‘excessive’ credit growth, a bust in one of them could initiate problems of the others.

Summarizing, the papers do not give a definite answer – as could have been expected, but contribute to creating “early warning” indicators, as well as increase awareness of the potential problem, justifying more scrutiny, better risk assessment and supervision. They are not fast to advocate measures to curb lending (which have been introduced for instance in Bulgaria) but warn on both the macroeconomic and financial risks associated with the rapid credit expansion.
Papers referred to:
1. Coricelli F., Mucci F., and Revoltella, D.,(2006) Household credit in the new Europe: lending, boom or sustainable growth? here...
2. Duenwald C., Gueorguiev N., Schaechter A., (2005) Too much of a good thing? Credit booms in transition economies: the cases of Bulgaria, Romania, and Ukraine
here...
3. Boissay F., Calvo-Gonzalez O., Kozluk T. (2005), Is Lending in Central and Eastern Europe developing too fast? here...
4. Kiss G., Nagy M., Vonnak B., (2006) Credit growth in Central and Eastern Europe:
trend, cycle or boom?here...