Sunday, November 21, 2010

Me on Risk, Return, and Alpha

This was filmed while I was in Rotterdam in between krokets. My argument is that risk premiums do not exist within almost all asset classes, therefore, it is a reasonable approximation to say they do not exist in general. This does not mean you should not take risk, merely, that you should take risk for other reasons.

4 comments:

vonjd said...

Very impressive videos! You said you wanted to do something concerning the volatility risk premium - did you do it yet? Do you think that at least there will be a premium for taking risk?

Anonymous said...

Granted, there seems to be no risk premium (or certainly less than theory wants) within asset classes, but how do you explain the fact that in general risky assets outperform over time? That appears to be a positive long-term risk premium. I agree with the strong empirics that you're not rewarded for higher risk (vol or beta) within most asset classes, but would you really tell investors you're not rewarded for total risk on a diversified buy and hold portfolio long-term?

If you put together a global portfolio of stocks, bonds, credit, etc., at low cost, and are a buy and hold investor, do you really think the expected return is cash? Or am I misunderstanding? If I'm misunderstanding I think you should make it clearer as it seems you're saying people aren't rewarded at all (ex ante) for risk taking. If I'm understanding correctly, then I don't understand... :)

Eric Falkenstein said...

Equities are a special case. Gold, bonds, currencies, do not show a clear 'risk premium' across asset classes. Indeed, think about gold, and whether it should be considered risky, or a hedge: it can be argued either way, highlighting that risk as a practical matter is ambiguous and an unsettled issues.

But take equities. Several issues make actual returns much less than the raw index returns--taxes, transaction costs, market timing--so that the average actual return is much less than the indexes imply. An efficient investor could have captured a sizable return premium in the 20th century being in equities, though there's the survivorship bias of the USA.

So,The standard equity risk premium assumption of 5%, I think, is 1) too high 2) even at 2-3% is applicable only towards the 'talented tenth' that avoid overtrading

BRM3 said...

The number of low or "managed" volatility equity mandates that have been undertaken in the public pension fund community in recent years argues that there is at least some adoption of the view that there is no real risk premium by the slowest-moving and least adept segment of the asset management industry.

While these mandates still total in the $billions, do you see a time at which low volatility equity strategies could actually produce a risk premium in equities?

I realize that a lot of funds are usuing low volatility approaches to risk-budget and then bar-bell'ing a higher risk emerging markets position (or something similar) on the other side but there are others (esepcially in parts of Europe) that have replaced nearly 50% of their traditional long only equity exposure with fairly passive managed volatility approaches.