An empirical study of Islamic banks and the financial crisis

After reading the IFSB Secretary General's statements about Islamic finance not being immune to crisis, I ran across a paper released last winter (Abduh, Muhamad and Raditya Sukmana. 2011. "Deposit Behaviour in Indonesia Islamic Banking: Do Crisis and Fatwa Matter?", PDF) which examined the relationship between several economic variables and the deposits with Islamic banks in Indonesia.  The connection with the IFSB comments is that the authors of the paper suggested that their empirical analysis demonstrated that Islamic banks were more resilient in the face of the latest financial crisis than conventional banks in Indonesia.  I read the paper with a keen interest to see my own preconceptions about the stability of Islamic banks in the financial crisis challenged (particularly since the data in Indonesia could tell a different story than my anecdotal story from the GCC experience). 

Their method of capturing the effects of the crisis was to include a dummy variable (one that has the value of 1 when a condition is met, and 0 otherwise) for the periods during the financial crisis, which they define as April 2007 to December 2008.  This method is likely to be effective to capture the direct impact of the financial crisis (the financial contagion which spread around the world rapidly as the value of mortgage-related assets fell).  However, it will not capture the second wave effects from a shrinking or slowing global economy and the wealth destruction that occurs as a result of the crisis. 

The authors, in their literature review, provide a similar critique of a paper which investigated the performance of Islamic banks in Malaysia before and after the Asian financial crisis of the late 1990s using data from 1994-1999, which the authors note will not reflect the post-crisis period "since the impact of the 1997/1998 global financial crisis is felt until beyond the year of 1999".  I would offer a similar critique of their study's conclusions about the resiliency of Islamic banks to financial crises: their selection of the dates for the financial crisis will only pick up first round effects on Indonesian Islamic banks and cannot be generalized to support a thesis that: "whenever crisis comes, it seems
that Islamic bank depositors will not withdraw their money substantially in both the short run and the long run periods".  What their tests showed was that in the short-run, Islamic bank depositors did not withdraw in large amounts.  The longer-run effects, as well as the resiliency of Islamic banks to crisis has not been tested. 

The authors do explore other reasons for the lack of direct impact of the financial crisis on Islamic banks:

"Other explanations on why crisis is not a significant factor to influence deposit in Indonesia Islamic bank are as the following. First is that Islamic banks do not indulge with the derivative instruments. Secondly,  during the time of crisis, institutional investors which put the huge amount of investment in US were also affected. In turn, people dump their dollar to the market which resulted in dollar being significantly decrease in value. For those banks which provide loan/financing overseas, obviously, will be greatly affected. Fortunately, Islamic bank in indonesia has spreaded the financing to domestic market which certainly adopt the local currency of rupiah. The dynamic movement of US dollar will have no influence to the performance of islamic banks."

I would agree that Indonesian Islamic banks were less exposed due to their lack of exposure to derivatives.  The second explanation, of a dollar devaluation, is puzzling since when speaking about the crisis itself, since there was a surge of money into the US during the height of the crisis (as a result of a flight to safety into US Treasuries), which led to a dramatic appreciation of the US dollar (breaking a longer term trend of depreciation). The argument which follows, however, does make sense: Islamic banks in Indonesia, with more domestic exposure, and less international exposure, would be expected to take a smaller hit than if they had more international exposure.  However, without comparing the foreign-domestic split of assets between Islamic and conventional banks in Indonesia, it is hard to say whether this explanation makes sense. 

My conclusion remains relatively unchanged after reading this paper.  Islamic banks were not exposed to the first wave of asset implosion that was led in large part by the exposure of conventional banks to mortgage-related products, including derivatives, which sparked an institutional run on the bank (most countries have deposit insurance, which limited the retail bank runs).  Islamic banks have been developing more and more complex products, but at the time of the crisis, had not developed mortgage derivatives, so were spared from the direct impact of the crisis. 

However, as the crisis led to a worldwide recession, it begun to affect Islamic banks that saw American and European banks retrench somewhat from countries with a lot of Islamic banking, which led to slower growth in these countries (along with falling oil prices, which hurt GCC countries in particular).  The slowing economy and reduced inflows of oil money and global wealth destruction led to a reversal in asset prices (notably real estate prices in Dubai), which did not spare Islamic banks, and which continues in some corners of the Islamic finance industry (e.g. Arcapita and GFH).