Failing Banks: Bankruptcy or Receivership?
In today's Wall Street Journal Stanford Professor John Taylor argues that resolving failing financial institutions via the bankruptcy process rather than a FDIC style receivership process is the only way to really limit bailouts. The heart of the argument is that the government is far more likely to inject funds if the process is controlled by political appointees and bureaucrats, rather than a judge. While this is probably the best reason to use the courts instead of a bureaucratic process, there are many other reasons to consider.
Proponents of administrative receivership often argue that the bankruptcy process is simply too slow to deal with banks and other financial institutions. "Too slow" sounds like an empirical question to me. So what does the data say? An interesting article in the Journal of Finance reports the median time in Chapter 11 to be 28 months and the median time in Chapter 7 to be 22 months. How does this compares to FDIC bank resolutions? Surprisingly, not bad. A Federal Reserve Bank of Chicago study finds the median time for FDIC resolutions to be 28 months, just as long as they typical Chapter 11, but longer than the typical Chapter 7. It could be argued that bankruptcy is actually quicker, as a Chapter 7 liquidation is more compare to an FDIC receivership than is Chapter 11.
Given that there doesn't seem much of a time advantage to receivership, is there a cost advantage? After all, bankruptcy does require all those lawyers. In regards to bankruptcy, there is good data on costs. The median cost for a Chp 7 is 2.5% of assets, and for Chp 11, 2% of assets. Interestingly, the largest expense in Chp 11 is administering the creditors committee, which would not be needed in a receivership (as creditors go unrepresented). Also of interest is that costs, as a percent of assets, decline with size. For firms above $10 million in assets, median costs are 0.8% of assets.
Unfortunately there is not good public data on the FDIC's costs. I have been told, however, that despite my initial suspicions, the $50 billion figure in the Dodd bill was calculated as the cost to resolve an entity like Lehman. At the time of its failure, Lehman's assets were around $600 billion. If we are to take $50 billion as the cost of resolution, that would imply a resolution cost in excess of 8%, considerably above what a comparable Chp 11 would cost. As the Lehman bankruptcy is resolved, we will have better data, yet at least from the various data points we have, the case for FDIC being a cheaper, or faster, alternative than the courts is far from conclusive, with some evidence suggesting the contrary.
