Shiller-Bashing

Scott Sumner writes,

I distinctly recall that Robert Shiller did not recommend that people buy stocks in 2009. That made me wonder when Robert Shiller did say it was a good time to buy stocks.

Barry Ritholtz writes,

By one metric — Yale professor Robert Shiller’s cyclically adjusted price-to-earnings ratio, or CAPE ratio — stocks are especially pricey. This has become the bears’ favorite valuation measure. But beware of cherry-picking any particular metric that rationalizes your position. Indeed, over the past 20 years, the CAPE measure has pegged U.S. equities as “overvalued” 85 percent of the time.

But for me, the most interesting Shiller-bashing is in the book I am reading by Duncan Watts, Everything is Obvious. He reproduces a chart created by David Pennock and Dan Reeves, using option prices to derive the probability distribution of future stock prices. The chart shows clearly that the uncertainty about future stock prices is much higher than the variation of past stock prices. That is exactly the criticism that I made of Shiller’s famous “variance bounds” estimates when he first published his work on that topic, and which he told the journal editor to reject. I still think that I was right. I should note that Watts does not make the Shiller connection in his book. However, I think that Watts gives us plenty of reason to be cautious about making statements like “Shiller called the housing bubble.”

I wish that more economists were aware of Watts.

This entry was posted in books and book reviews, financial markets, Scott Sumner is Coherent. Bookmark the permalink.

12 Responses to Shiller-Bashing

  1. Andrew' says:

    “Indeed, over the past 20 years, the CAPE measure has pegged U.S. equities as “overvalued” 85 percent of the time.”

    Because they have been overvalued. That’s why we had a moderation and then a depression, in some humble opinions. The period in 2009 to grab up stocks at good value was something on the order of a month to months. Ritholtz is being a trader, which is great for Ritholz.

    • Andrew' says:

      I should make clear by emphasizing the fact that the time to grab stocks was to a rough estimate one month at the exact time (some) people think Ben Bernanke was saving us from the end of the world.

      In real terms, stocks have returned close to zero for coming up on 15 years, if they reach a permanent plateau, in exchange for a roller-coaster risk ride, so what is incongruous with saying they’ve been overvalued for a large part of the last 20 years? We don’t, after all, have to hold stocks. The good news for stocks, though, is we’ve managed to screw up most alternatives.

    • Al says:

      We had a moderation and a depression because stocks were overvalued? I don’t think you have thought this through.

      • Rick Hull says:

        Al, think in terms of a bubble popping. We rode the self-reinforcing mania (irrational exuberance, if you will) on the way up until it simply could not continue.

      • Andrew' says:

        We had a moderation which created an over-valued condition and a depression mainly because of Greenspan’s put, which caused the false moderation. That’s my story. Choose your own adventure.

  2. Patrick says:

    Hi,
    Are you accounting for the fact that the distribution derived from option prices is the risk-neutral distribution? It contains the risk-aversion of the investor. So a wider spread in this distribution compared to the historical one can also be interpreted as investors who buy insurance against extreme outcomes

    Greetings
    Patrick

  3. F.F. Wiley says:

    Shiller’s variance bounds and CAPE aren’t perfect (nor does he claim perfection), his public policy positions are mindboggling (and not just for the interventionism – how do you reconcile a belief that markets are irrational and bubble-prone with a recommendation that government should subsidize financial advice?), and he may be a bit of a permabear (or not, I have no idea what he was saying in 2009 and, in any case, he’s not in the investment advisory business).

    That said, I’d like to balance things out a bit by saying that he’s put out a lot of interesting (and accurate) thinking about markets. I recommend both “Irrational Exuberance” and “Animal Spirits,” despite the unfortunate JMK tie-ins in the latter. CAPE is a useful way to look at valuation when you think that earnings may be unsustainably high (I agree with Andrew – Ritholz is just talking his book). And, Shiller did clearly warn that both stock prices in the late 1990s and house prices in the mid-2000s were overdone. In other words, I’m taking the “most charitable view” on this one.

    • Al says:

      The matter of Rithholz’s motivations is worse than inconsequential in that it confuses careful reasoning with the dumb satisfaction of being able to categorize a pundit in a self-serving manner. CAPE is as useful as any model of expectations, but it is used bluntly. Like Andrew, many people simply look at points above trendlines, nod their heads, and announce “overvalued”. There are some accounting adjustments to be made, like standardizing goodwill amortization across periods, and the intelligent user must still have a theory of why CAPE shows what it shows. What people like Rithholz and Sumner point out is that the dumb use of CAPE shows poor utility.

      • F.F. Wiley says:

        Since you mention the accounting change issues, a guy at Hussman Funds wrote a good piece recently on this and on CAPE generally:

        http://www.hussmanfunds.com/rsi/cape.htm

      • Andrew' says:

        I don’t even know what you are suggesting, so I’m pretty sure I didn’t do that.

        • Andrew' says:

          I am not putting forth any estimation of the degree of over-valuation of stocks. I just know that I am not required to hold stocks when they are in an upper band.

          What is the total return on the stock market since 1994? Why can’t the market be relatively over-priced for 20 years?

  4. Floccina says:

    Haven’t the returns from capital been falling since the industrial revolution?
    Wouldn’t that mean that this should be factored in to expectations about stock prices?
    Wouldn’t that mean that a fixed ratio is insufficient but rather the target ratio should rise slowly over time?

Comments are closed.