Finance and the equity risk premium: A solved puzzle?

I am taking the opportunity of Professor Robert Barro's recent article entitled "Rare Events and the Equity Premium Puzzle" (pdf) to test a new service that I subscribed to.

First a few words on the so-called equity premium puzzle. In a nutshell, it says that the long run performance of equity markets exhibit a remuneration of the equity risk that is to high to be compatible with traditional pricing models such as the CAPM (you are the lucky one if your grand parents and parents were wise enough to have invested in the stock market!).

Numerous researchers have tried to modify the model to make it more complex: Habit formation (what I have for lunch does depend on what I had for breakfast and/or I want to keep up with the Joneses), labor income risk, generic background risk etc... Some others have suggested that the data that have been tortured until they confess are "biased": They mainly refer to the stock market that has indeed been the most successful, namely the US stock market, hence a survivor bias etc...

Here comes Barro who resurrects an old piece by University of Iowa Economist T A Rietz. In this paper Rietz suggests that the puzzle can be explained if one takes due consideration of low probability disasters. What are these disasters? In Barro's own terms:

"Well, “rare events” in this context are low probability, large disaster events. As a U.S. example, you think immediately about the Great Depression. However, war has been historically more important for most countries. I have in mind particularly the economic devastation of World Wars I and II for many countries, including much of Western Europe and Japan. From a U.S. perspective, one thinks about the world wars as times of good economic performance, but that outcome is unusual from a global perspective. For the U.S., one has to go back to the devastation of the Civil War in the South to find something comparable."

What are the consequences of such events? Again in Barro's terms:

"Suppose that you have potential events with, say, a 1 percent annual probability, where you lose half of your capital stock and GDP. This possibility seems to be enough to get something like the observed equity premium. Moreover, this mechanism has implications for a lot of other variables, not just for the excess of the average return on stocks over the return on government bills. For example, it can explain the very low “risk-free” rate and low expected real interest rates during most U.S. wars back to the Civil War. It can also explain some of the evolution of price-earnings ratios for the U.S. stock market."

Remember that the puzzle is about the equity risk premium being too high, in other words (for the CAPM aficionados) the empirical (observed) slope of the Security Market Line being too steep. The line starts at the risk-free rate where equity (beta) risk is zero and then goes linearly upward at a rate given by the equity premium (so that you get compensated for the extra equity risk you have to carry). The puzzle is also named the risk-free rate puzzle as the observed risk-free rate is low compared to the theoretical one. Well, according to Rietz and Barro, it could well be that people do not discard rare events that much or, more precisely, that their perceived disaster probabilities over time are higher than has been assumed in the literature (e.g. Mehra/Prescott). Hence the demand for the safe harbor investments is higher and the risk-free rate should indeed go down (Hint: How do you get to estimate these perceived disaster probabilities? Well, by looking at option prices and if you go deeper in the past at insurance prices).

So much for complex explanations! Two intertwined things worry me though. First, the Rietz/Barro argument sounds like the quantum leap debate in physics (disclosure: My field is not physics!). A lot of the literature in the economics of risk and uncertainty has provided evidence that people usually underestimate low probability events (from Howard Kunreuther to Nassim Taleb's famous Black Swan, click here for a video of Taleb). People have a hard time moving from the "normal" to the "rare" and back. How come that things (chief among them behaviors) are not the same in the small and in the large? So, who's right? Those who believe in the Black Swan or those who don't? More work is needed and it does not relate to finance only (think of climate events etc...).

Second, University of California Philippe Jorion and Yale University William N Goetzmann have shown that when you gather more data from more markets (outside the US) you get an empirical premium that is lower than the one that has been estimated on the US market only. Question: How do you reconcile the two sets of observations and arguments? In other words, is the premium too low or too high, is it vanishing and what kind of premium level should we expect in the future?

Maybe the real puzzle is with researchers themselves who seem to be as puzzled as the Equity Risk Premium itself!

Now for those who'd like to know where finance was in 1999 here is a piece entitled New facts in finance. by Cochrane, John H. source: Economic Perspectives, September 22, 1999.
via: HighBeam™ Research
COPYRIGHT 1999 Federal Reserve Bank of Chicago.

(This is the reason why I wrote this rather lengthy note to check how HighBeam research is working.)

If you need further bibliographies on the ERP puzzle here is one and another one.

Comments

[avatar]

Well dont know but have a read here
http://richerich.blogspot.com/

[Pete]

re: "people usually underestimate low probability events"
Doesn't it matter whether the event has a positive or negative outcome? I thought the literature showed that people underestimate *negative* low-prob events but overestimate *positive* ones (e.g. winning the lottery).

[Pierce Wetter]

Or maybe CAPM is just broken.

(Note that I work at Marketocracy, so I've conclusively convinced myself that this is true...)

[Jeroen van Bommel]

As a student at the vrije universiteit amsterdam currently cramming for corporate finance as well as investment exams I am pleased to discover this most excellent blog/resource center on interesting finance stuff. Keep up the great work.

ps. CAPM broken is partly due to the fact that the model fails because of a flawed beta...from post 1964 (according fo Fama&French) the model is simply not accurate. Perhaps we can attribute it to the globalization of the world economy and perhaps a little bit of bretton woods decline/termination? what do you think?