Bryan Caplan  

Ben Bernanke Explains Systemic Risk

The Case for Killing Textbook ... Will Obama Consider Geoenginee...
Last week, Arnold approvingly quoted Tyler's one-sentence explanation of "systemic risk":
If your banks are less risky, often something else is more risky, and vice versa.
This morning it just occured to me that this is precisely the opposite of what Ben Bernanke taught me.  Assuming my memory doesn't fail me, Bernanke's one-sentence explanation of "systemic risk" was basically:
If your banks are less risky, often something else is less risky, and vice versa.
The claim, in other words, is that bailing out a failing sector is supposed to make failure in other sectors less likely.

As far as I can tell, Bernanke's explanation is the standard one.  It's also an internally consistent story about why bail-outs are better than they seem.  If you bought Tyler's version, though, systemic risk would be a story about why bail-outs are worse than they seem - when you help banks, you're hurting other sectors, which will in turn need bail-outs of their own.  Am I missing something?

Comments and Sharing

Twitter: Bryan Caplan @bryan_caplan

COMMENTS (7 to date)
Robert writes:

No, you're right. Bailouts aren't necessarily making things worse, but it's true that helping the banks and helping the rest of the economy are two separate, and often opposing, aims.

By giving money to the banks, we are simultaneously -not- giving money to people who would then use it to consume or pay off debt or build their savings and return to a healthier financial position.

The same is true with regards to mortgage modifications - if we modify the principal on those mortgages the banks lose (because they make less money) and homeowners win (because they are on healthier financial ground and can make smaller monthly payments).

George Soros said something to the same effect yesterday: "What we have created now is a situation where the banks will be able to earn their way out of a hole but by doing that, they are going to weigh on the economy," Soros said. "Instead of stimulating the economy, they will draw the lifeblood, so to speak, of profits away from the real economy in order to keep themselves alive."

rvturnage writes:

Doesn't stories like this one from Forbes speak against the idea of systemic risk? If a large bank fails, aren't the entrepreneurial startups (perhaps started by former employees at the failing bank), existing smaller banks and larger competitors going to come in and pick up the pieces, much like the banker profiled in that article is attempting to do?

David writes:

Presumably, systemic risk has something to do with the risk posed by the interconnectedness of the various parties within the system and the risk that the failure of one party will cause failures of numerous other parties.

Consequently, increased riskiness of one of the parties tends to increase the riskiness of the other parties in the system.

I don't get the conclusion however that that conception of systemic risk implies that bailouts are OK. Bailouts increase moral hazard. Moral hazard encourages the assumption of undue risk and reduces system stability.

Wayne writes:

I would like to expand on what Robert wrote: George Soros is painting a picture of what the government in Japan decided to do when the Real Estate Bubble popped in Japan in 1989. Instead of winding up the bad banks in Japan, the Government decided to prop up and massively consolidate the banking system. This ended up with Japan's economy staggering along for now what is 20 years. Today, their stock market is significantly lower than it was 20 years ago. Moreover, IIRC their government deficit is aproximately 200% of GDP (please correct me if I am mistaken), according to IMF estimates for 2008 (the OECD has the 2008 estimate at 173%. This is significantly more than all of the other OECD countries, with Germany in the 60% range along with the USA. #2 most levered is Italy at 113%, also considered in bad shape. Both Japan and Italy are rated AA, and Italy is on negative watch.

Troy Camplin writes:

Is risk a bad thing? Yes and no. Obviously, risk that doesn't pan out is a problem; risk that results in a positive return isn't. If you reduce risk, you reduce bad risks, but you also reduce good ones. You get a recession after you have a period of high risk, where the bills all come due for the risks that didn't pan out. This clears the way for all those risks that did pan out to rise up and reform the economy. We see high levels of corruption during a recession precisely because these are risks that by definition won't pan out -- they get uncovered during the collapse as well.

A bailout is by definition an attempt to help those whose risk(s) didn't pan out. That money has to come from somewhere -- it has to come from those whose risk(s) did pan out. Thus, you reward failure and punish success when you bail people out. The message is that you can engage in careless risks, because the government will be there to bail you out. Thus, bailouts encourage people to take unhealthy risks, while a more typical risk environment would encourage you to at least ascertain the relative risk you will be taking. So a bailout can in fact increase the likelihood of risk-taking, but it will not discourage carelessness. Not all risks are created equally.

I have formulated a theory of what caused this recession that is related to the relationship between interest rates and risk-taking that I can't post here because I sent a short piece on it to the Wall Street Journal, but I would be willing to discuss it with people via email.

Bob writes:

Systematic risk stems from real assets. With a given asset mix in a well-functioning market, you can only move it around.

Leverage is a good example. If an asset is partly debt-financed, the total systematic risk does not rise. The equity is riskier but the debt is less risky. In this sense, TC is right.

The problem is that structural choices can cause a market to stop functioning well. This can be thought of as a systematic risk, although it isn't what we typically think. In this sense, BC is right.

In the end, though, the primary systematic risk that banks have to bear in a well-functioning capitalist economy is liquidity risk. Banks fund less-liquid assets with highly liquid liabilities. Once in a while, this is going to blow up. But there is no good alternative - if you move the liquidity risk out of the banks, it has to end up somewhere. You can't eliminate highly liquid liabilities - that's the money supply in a modern economy. And you can't eliminate less-liquid assets; those are our investments in future productivity.

So, in the end, TC is right, IMHO.

SheetWise writes:

"If your banks are less risky, often something else is less risky, and vice versa."

"The claim, in other words, is that bailing out a failing sector is supposed to make failure in other sectors less likely."

It depends what you mean by "something else".

Whether risk in banking causes more risk or less risk would depend on whether or not there are interdependencies. If there are, Bernanke's statement makes sense. If there aren't, Tyler's statement makes sense.

Since an investor is always making a tradeoff between risk and return, it doesn't make much sense that there are interdependent sectors whose risk of failure is inversely related.

Comments for this entry have been closed
Return to top