Tuesday, August 11, 2009

The Value of The Efficient Market Theory

I was an official economist for a while, back at KeyCorp, and would sometimes have to give little presentations on the standard macro variables. I remember once in early 1995 having a little 5 minute presentation to the board of a large hardware supply company (a small 'Home Depot' type chain). About 20 suits in a room, and they were all very interested in where interest rates were going. At that time, rates had rise a lot, the yield curve was steep, and they were very concerned. Now, given they had a lot of fixed rate long term debt, they could lose a lot of money if rates fell (rates fall, bond prices rise, and they were short bonds). If rates fell, their liabilities would go the other way. As rates were moving a lot, they were very concerned.

I told them in general, when the yield curve is really flat, rates tend to fall. They did, but that's not the point. Indeed, I should have told them I have no clue, but I knew they wanted a 'view', and I was trying to be helpful (we were a 'full service' bank). They debated back and forth for 20 minutes, and though I left, I had the sense they kept discussing it. I never found out what they actually did. The point is a large amount of management time was spent discussing something where they had no alpha, no comparative advantage, no value as management. They all assumed markets were 'wrong', so the question to them was clearly which way. If instead they thought bond prices incorporated all available information, they would have chosen a neutral position, say by having floating rate debt (it was feasible to swap into this).

A lot of resources are wasted by people who think it is imprudent to act as if markets are rational. They then basically take random gambles that on average lose money via fees paid on various transactions. Most importantly, they lose their focus on where they conceivably have alpha.

11 comments:

Anonymous said...

well, what do you tell your clients now, inflation or deflation?

Eric Falkenstein said...

Heh. I don't advise on such matters, because I don't know. I guess like most economists, I think inflation will be very low for the next couple years, but could be very high after that.

jsalvati said...

Since you frequently have something to say about EMH, I humbly ask you to comment on this: http://www.economist.com/blogs/freeexchange/2009/08/lucas_roundtable_the_emh_must.cfm. Are you familiar with this literature?

Anonymous said...

Someone please share this with Congress

Eric Falkenstein said...

I don't see any reference for Smithers presumption that EMH has been proven to have a negative autocorrelation inconsistent with the EMH. I know markets have negative autocorrelation at longer frequencies (3-5 years), but it is extremely weak, not statistically significant.

Ivo Welch did a neat out-of-sample test of aggregate stock market predictors, and they did horribly. Hindsight is 20-20.

Anonymous said...

So then I assume you are also against the notion of a Central Bank?

Eric Falkenstein said...

I don't see why that follows? A clearinghouse of some type is needed for the big banks. If they target a Taylor rule, I don't think that is inconsistent with the EMH.

cityislander said...

"A large amount of management time was spent discussing something where they had no value as management."

Not only do the best mind have little predictive powers, the majority of analysts are going with the flow or concocting their own stories.

One solution, is to hedge against what you don't know (interest swap).

A very long term solution is to teach to recognize one's cognitive biases in our education system.

Anonymous said...

By setting short-term interest rates and/or buying assets is not a Central Bank simply saying it is smarter than the market. If the short-term interest rate is always "less good" than a market-produced rate shouldn't that produce all sorts of negative outcomes?

Anonymous said...

http://www.ritholtz.com/blog/2009/08/six-impossible-things-before-breakfast/

Anonymous said...

Re: Inflation or Deflation --
Anonymous economist agrees with Eric. There's been a change in psychology towards consumption that will persist for a few years. That's going to mean some excess capacity for the next few years and weak prices. But then it will be forgotten, and the cycle will start all over again.