Valuation-Informed Indexing #226

by Rob Bennett

There are some stunning words that appear on Page 37 of my copy of Jeremy Siegel’s Stocks for the Long Run. Siegel states that: “The  recommended equity allocation increases dramatically as the holding period lengthens. The analysis indicates that, based on the historical returns on stocks and bonds, ultra-conservative investors should hold nearly three-quarters of their portfolios in stocks over 30-year holding periods. This allocation is justified since stocks are safer than bonds in terms of purchasing power over long periods of time. Conservative investors should have nearly 90 percent of their portfolio in stocks, while moderate and aggressive investors should have over 100 percent in equity. This allocation can be achieved by borrowing or leveraging an all-stock portfolio.” The table that appears on the same page as these words indicates that risk-taking investors with a 30-year holding period are best served by a stock allocation of 131.5 percent, according to the historical return data.

Could this be real?

Most of us are investing primarily to finance our retirements. So we have a 30-year holding period. We’re not all risk-takers. So perhaps the recommendation of a stock allocation of 131 percent does not apply for us. But Siegel’s book shows that even investors with only a moderate risk tolerance (who have a 30-year holding period) should be going with a stock allocation of 112.9 percent. I don’t know anyone who has a stock allocation even remotely that high. And I doubt that even Siegel would recommend that real-life investors borrow to provide for stock allocations in excess of 100 percent. But he obviously is a skilled academic. And he is saying that this is what the numbers show. So this much be what the numbers show.

Still, the question remains —

Could this be real?

I am confident that the numbers show what Siegel says they show. But I don’t believe that the story he tells is real. I think that the idea that a stock allocation of more than 100 percent makes sense at all times is a mirage.

The problem is that Siegel’s numbers report what returns the market provides rather than what return investors obtain. Stock valuations don’t move upward and downward in a random pattern. They head generally upward for about 20 years and then generally downward for about 15 years. Most investors go with much higher stock allocations when prices are high than they go with when prices are low. So most investors earn returns much lower than the market returns cited by Siegel. His claims are theoretically true but in a practical real-world sense untrue.

That said, his findings raise a compelling question.

Does it make sense that stocks could always be the superior asset class?

I do not believe that it makes sense. Yale Economics Professor Robert Shiller’s research shows that valuations affect long-term returns. That means that stocks offer a better long-term value proposition in some circumstances than they do in other circumstances, that stock investing risk is variable rather than constant. If that’s so, there is no one optimal stock allocation. The optimal stock allocation might in some circumstances indeed be in excess of 100 percent. But in other circumstances the ideal stock allocation might well be as low as 20 percent, or perhaps even for some investors as low as zero.

The optimal stock allocation for those who believe in Shiller’s research varies with changes in the P/E10 level.

The highest P/E10 level in U.S. history is the 44 that applied in January 2000. The historical data shows that the most likely 10-year annualized return at that time was a negative 1 percent real. It doesn’t follow that the optimal stock allocation was zero. Short-term returns are unpredictable (because they are governed by emotion). So I would say that the optimal stock allocation for the typical investor was 20 percent in 2000. Going strictly by the numbers, the optimal stock allocation at the time was 0 percent. But Shiller’s research shows that investor emotion is the controlling factor in achieving long-term investing success. It never makes sense to ignore emotional considerations. The typical investor would experience regret if he went with a zero stock allocation and stock prices soared in the following year. For most of us, something in the neighborhood of 20 percent stocks made the most sense in 2000.

The lowest P/E10 value that we have seen in recent times was the 15 that applied in 1982. At that price level, the most likely 10-year annualized return is 15 percent real. A stock allocation of over 100 percent certainly makes sense in a theoretical sense when returns are that high. One could finance a retirement in a shockingly short amount of time while earning 15 percent real per year if he were invested 200 percent in stocks. In the real world, middle-class people should not be buying stocks on leverage. The real-world answer to the question is that the optimal stock allocation in 1982 was 80 percent for most and perhaps 90 percent for young investors inclined to take on a higher level of risk than most.

So I don’t think that Siegel is right that the optimal stock allocation for those with a 30-year holding period is a number in excess of 100 percent. That is the case in a theoretical sense when prices are low. When prices are sky-high, even investors with long time horizons are better off going with a low stock allocation until prices drop and only then moving to a high stock allocation. Doing it that way permits the investor to invest a much greater number of dollars in stocks at times when the long-term value proposition attached to owning them is strong.

The amazing thing is that both the Siegel analysis and the analysis that I have done based on Shiller’s research find that stock allocation in excess of 100 percent often make sense. Isn’t that strange? Isn’t it counter-intuitive that any asset class would ever offer returns so great that allocations in excess of 100 percent would be in order? Wouldn’t you think that investors would flood in to take advantage of the outsized returns and thereby pull the available return downward to more reasonable levels?

I think that the issue here is that our understanding of how the market works remains primitive today. There are some who suspect that stocks offer a great deal when prices are low. But there is not yet enough research available to give them the courage they need to go with stock allocations in excess of 100 percent. There will come a time when such outsized returns will no longer be available. They are available today because our knowledge is not great and thus emotion plays a big role in our decision-making process. The market may be moving in the direction of efficiency but it has a ways to go before the destination is achieved.

Rob Bennett recorded a RobCast titled Buffett is the Scarecrow, Shiller is the Tin Man, Bogle Is the Cowardly Lion. His bio is here.