Saturday, December 19, 2015

Economic regulation and institution design

I've been having some thoughts about how the economy is regulated and the design of the institutions doing it that I've been meaning to put down in a blog post.  So here goes.

I haven't read Ben Bernanke's memoirs myself and maybe I should but I've been reading various economic bloggers reactions to them.  Here is an example.  One thing that stands out to many people is that during the financial crisis the Fed was so preoccupied with the banking system that they essentially forgot that they were also the agency in charge of steering the macro economy at large, trading off between inflation and unemployment and so forth.

"Nothing is as important as you think it is while you're thinking about it."

It's easy to see how this could happen.  In addition to trying to steer the macro-economy the Fed is in charge of bank regulation and that was generating an enormous amount of work in 2007 to 2008.  As the subprime crisis hit bank after bank went to the Fed for help.  Some were proactive but some, like the Lehman Brothers, only went to the Fed for help a few days before they collapsed.  But they all ended up screaming for attention at some point.

By contrast news from the macro-economy trickles in slowly.  If you follow the economic blogosphere you know the excitement when a new dollop of data is released about the economy.  When that happens we get a best guess about conditions a month in the past but also exciting updates about what the conditions really were for the months before that.  Inflation is complicated so it's hard to figure out what the various measures of inflation really are.  And data on employment isn't always passed around quickly.  That means that the Fed is always working with outdated information about the larger economy and is guessing as best they can.

Which causes problems when you're dealing with a crisis.  When Lehman et al were collapsing inflation was crashing too but the Fed's most recent information showed that it was higher than normal.  That's why the Fed took the unprecedented step of paying banks not to lend money when they were shoveling out cash trying to stop the financial crisis.

That's a problem.  If there was some macro committee in charge of inflation and and a bank committee in charge of the banks then the bank committee could have happily gone into panic mode while the macro committee could have looked at all the signs that a larger economic slowdown was causing the financial crisis just as much as vice versa and taken appropriate steps.  The signs were there in the TIPS spread and other places but you can't expect a committee with a thousand banks screaming at them to go out and proactively look for economic data.

Another problem that separate committees would solve would be the question of who gets a seat at the Fed board.  The Fed has a number of brilliant economists like Bernanke but also a large number of people who are familiar with running or regulating banks but are generally ignorant of macro-economic theory.  During the crisis I was sometimes pulling my hair out at some of the things the president of the Texas Fed was saying about he thought that the economy was probably going to be ok 'cause his business buddies said they were doing ok.

There was some hope during the great recession that Obama would nominate some Doves to the Fed committee and balance it out but sadly that was not to be.  After a long delay Obama finally nominated some candidates who I'm sure were very progressive in their ideas about bank regulation but who were very conservative in their views of monetary stimulus.

It's not surprising that people who have spent their lives thinking about what is prudent in terms of institutions that can't print money should develop firm ideas that don't necessarily apply to institutions that can print money.  But that's something we could resolve by separating the two functions.

The Theory of Moral Sentiments

That's all about the macro-economy.  What about bank regulation?  I'm very unconvinced that Dodd-Frank or any of the other changes since the financial crisis will actually help in the long run.  Congress routinely delegates legislative power to bureaucrats who can make rules for the legislation of the industries they're charge with.  We also have the problem that the number of pages of law regulating the financial industry grows but the number of people charge with enforcing those laws does not.  And so the enforcement grows ever more arbitrary.

In the short term this isn't a problem.  All the bureaucrats in charge of enforcing the rules remember the financial crisis.  It looms large in memory and for the next while we shouldn't worry about lax enforcement.  But the traders and executives on Wall Street also remember the crisis where many of them lost everything.  Those memories will last for a number of years and while they come quickly to mind I wouldn't be surprised if we barely needed any regulation at all in practice and that fear of those frost giants who in 2008 descended from the north and tore down the towers of the Lehman Brothers wouldn't be enough to keep bankers mostly in line.

But as time passes memory fades and the children of summer who have never known great hardship come to take more and more responsibility.  The lessons of the financial crisis will fade and a new generation will come who want to take greater risks.

The problem is that at the same time the memories of the regulators will fade too.  Young bureaucrats who weren't there in 2007 will rise up.  They will want to do the right thing but like most psychologically normal humans they will want to get along with the people they interact with every day.  And many of those people will be the new banker chafing under the restrictions placed on them.  The laws run into their hundreds of pages of vague categories and it will be very easy for the sympathetic ear to say "Yes, I'm sure that extra risk won't cause any problems."

That's how we got into the original crisis in the first place after all.  The Basel Rules admit many categories of financial instrument based on how risky the regulators think they are.  With the tranches of subprime mortgages the regulators agreed that they were probably quite safe and let banks leverage themselves very heavily on them.  The savings and loans crisis of the late 1980s had grown dim in the minds of both the bankers and the regulators so stuff like that didn't seem very dangerous.  But it was.

A good response to the crisis would have been to recognize that deciding which investments are risky and which are safe is too hard a problem to expect people to solve reliably.  We could just expect or banks to stay below some fixed amount of leverage and leave it at that.  But instead of created a large number of very complicated rules that will probably be interpreted strictly for a while but more and more loosely as caution fades.

Then this will happen all over again.

Thursday, December 3, 2015

(Late) November Links

I should have put this up a few days ago but I was somewhat distracted starting my new job at RightHand Robotics.  There are some pretty cool videos on our front page.  But on to a curated list of nifty things I found on the internet in the last month!

Nature has a good list of some interesting unsolved mysteries in physics and cosmology.

There are a lot of nifty things that can only be made in microgravity like foamed metal and antibiotic crystals.  Now we can add metal glasses to this as well.

A lot of the time we think of conservatives as being in favor of "original interpretations" of the Constitution and liberals as being more supportive of the "living constitution" approach.  There are good reasons for that but it's worth remembering that many of the things that the Black Lives Matter protesters and others criticize about how our criminal justice system handles police immunity were "judge created law" as conservatives usually put it.  The recent ruling saying that policemen can't be sued for using deadly force against fleeing suspects is just another example.

There's been a lot of ink spilled on the problem of replicating experiments in psychology recently.  Here's a comic that basically shows how you get the problem.  Here's a nice article on how you can find it via analyzing papers without having to do a study yourself.  And here's on possible solution.

There was a kerfuffle at the MFA recently with their Kimono exhibit which let museum patrons try on kimonos.  A diverse group of people protested saying that this was cultural appropriation and the counter-protest of a group of Japanese and Japanese-Americans wasn't able to prevent the museum from closing the exhibit.  And of course back in Japan people mostly interpreted the protesters as just being anti-Japanese.

Some people are thinking of mining undersea vents.  I think that article did a pretty great job of covering the pros and cons.

The New Yorker had a pretty excellent piece on Nick Bostrom and some of his ideas.  I've been following him since 2001 or so and I was happy to see it.

And speaking of the New Yorker they also got Randall Monroe to do a piece on the 100th anniversary of Einstein's theory of relativity.

On average about 56 million people die each year so roughly 4.6 million died in the last month.  Wikipedia has a list on the few hundred who are the most famous.  One particular one, Joseph Engelberger, was "the father of robotics" which struck a bit close to home (though he actually died Dec 1st).  Every one of those millions of deaths is a tragedy and it's worth remembering that when we consider the various people killed in terrorist attacks this month.  The 9/11 attacks killed 3,000 people directly.  The extra road deaths they indirectly caused through people not flying as much, through fear of terrorism or TSA hassle, now stand at about 7,000.  But in the end both of these numbers are tiny compared to the vast ocean of suffering that exists in the world.

We're trying to make our SARS-2 tests better than we should

Ok, that's a somewhat provocative title but I think it's basically accurate.  During this pandemic the US in particular has had a pr...