The Federal Reserve Board announced its intention to modify the annual bank stress tests. In the words of the Fed:

The Board’s stress test evaluates the resilience of large banks by estimating their losses, revenue, and capital levels under a hypothetical severe recession scenario. Due to the changing hypothetical nature of the test, the results also change and introduce volatility each year. The results, in part, determine the calibration of the stress capital buffer (SCB), which is one component of the amount of capital large banks must hold to absorb losses.

So, what would the Fed do? First, it would average the results of the stress tests over two consecutive years “to reduce the year-over-year changes in the capital requirements that result from the stress test.” Also, it would delay from October 1 to January 1 the date of meeting the SCB.

What should we think of these changes? At first blush, who cares? After all, the release says: “These proposed changes are not designed to materially affect overall capital requirements.” In other words, the exercise is pointless. Nevertheless, Bloomberg reports:

The Fed proposal drew sharp criticism from Governor Michael Barr, until recently the vice chair for supervision. “These changes risk turning stress testing into an ossified exercise that will provide false comfort in the resilience of the system,” he said in a separate statement.

Now, that is food for thought. The whole point of the stress test is to see if a bank can weather stress. Not had-a-bad-day-at-the-office stress; Great-Financial-Crisis-the-world-is-ending stress. One simply does not learn much, if anything, about financial crises from each year’s experience. So, a case can be made that the stress test should not change at all – not until there is new information about severe crisis. Meddling with the stress test every year sounds like cover for regulatory micromanagement and unwise attempts at fine-tuning.

Put Eakinomics firmly in the camp of the ossifiers; every two years seems too short. The stress test should be rigorous and stable.

The Fed also announced:

Later this year, the Board intends to propose additional changes to improve the transparency of the stress test. Those changes include disclosing and seeking public comment on the models that determine the hypothetical losses and revenue of banks under stress, and ensuring that the public can comment on the hypothetical scenarios used for the annual stress test before the scenarios are finalized.

This also drew the ire of Mr. Barr, who said banks are “likely to game the capital requirements once they know the details of the stress test.” This simply does not add up. Just because you allow banks to comment on the proposed test doesn’t mean you hand over the keys to the testing. The Fed does not have to agree with the comments or alter the test.

The stress tests clearly matter and that is good. If they did not concern banks, they would not be doing their job. But they are far from perfect and further debate and refinement is healthy.