The Case Against Valuation-Informed Indexing

The problem with the conventional indexing approach (developed by Vanguard Founder John Bogle) is that it does not call for adjustments in the investor’s stock allocation at times of extreme valuations. The historical stock-return data shows why conventional indexing works well during bull markets; stocks perform so well in bulls that it is not necessary for investors to take on the risks associated with trying to pick stocks effectively to obtain solid returns. It also shows why conventional indexing is not likely to prove effective in a prolonged bear market; indexers are hit hardest during a prolonged downturn (they cannot avoid the hit by picking companies doing better than the market as a whole) and the losses sustained in secular bear markets are so great as to make continued reliance on buy-and-hold strategies a practical impossibility for most middle-class investors.

Dangers of Valuation-Informed Indexing

The purpose of my Valuation-Informed Indexing approach is to provide a means by which middle-class investors can realistically expect not only to talk the buy-and-hold talk but also to walk the buy-and-hold walk. Valuation-informed indexers lower their stock allocations by about 25 percent at times of extremely high valuations (a P/E10 level above 20) and increase their stock allocations by about 25 percent at times of extremely low valuations (a P/E10 level below 12). Valuation-Informed Indexers appreciate the futility of short-term timing. Their expectation is that stocks will perform in the future somewhat as they always have in the past and that long-term timing (changing one’s stock allocation with no expectation of seeing a benefit for doing so for a time-period of up to 10 years) will continue to provide a substantial edge. Valuation-informed indexers are concerned with the value proposition represented by their index purchases and are not willing to commit the same percentage of their portfolios to stocks at times when the 10-year return is likely to be low as they are when the 10-year return is likely to be high.

This article summarizes 20 arguments that have been made against the Valuation-Informed Indexing concept, and offers my brief responses to those arguments.

Argument #1 Against Valuation-Informed Indexing — The Equity Premium Has Recently Been Greatly Diminished.

James Glassman put forward this argument in his book Dow 36,000. The idea here is that middle-class investors have only in recent decades discovered the long-term appeal of stocks, that this discovery will cause prices to go higher than most now believe is possible, and that the higher prices will permanently reduce the long-term return on stocks provided from that point forward.

Glassman is arguing that stocks are going to perform differently in the future than they ever have in the past. Valuation-Informed Indexing is rooted in an assumption that things will likely not turn out too different this time from how they have always turned out before. Glassman might be on to something. The reality is that the one constant in investing is surprise.

Still, I doubt that the future is going to be entirely different from anything we have seen in the past. I think that there are some reasonable arguments that can be made that stock prices will go higher than the historical data alone would lead us to believe. It’s a mistake to get too carried away with this kind of thinking, however. Some upward adjustment in the numbers obtained by looking at the historical data makes sense for those with an optimistic bent. That said, I think it would be fair to describe Herman Hupfeld as every bit as much of an investing expert as James Glassman. And Hupfeld, who wrote the famous song used in the film Casablanca, argues forcefully that “the fundamental things apply as time goes by.”

Argument #2 Against Valuation-Informed Indexing — Economic Growth May Be Greater in the Future Than It Has Been in the Past.

What Is Valuation-Informed Indexing?
This is another way in which the future could turn out different than the past. Those who believe that economic growth will be greater should make an upward adjustment in the numbers generated by The Stock-Return Predictor (see tab to the left).

Argument #3 Against Valuation-Informed Indexing — Economic Growth May Be Less in the Future Than It Has Been in the Past.

Those who believe that economic growth will be less in the future of course need instead to make a downward adjustment to the numbers generated by the Return Predictor.

Argument #4 Against Valuation-Informed Indexing — The One Thing We Can Count On Is That There Will Always Be Surprises.

We look to the historical stock-return data for guidance as to how stocks may perform in the future. I think it is a big mistake to ignore the lessons taught by history. The other side of the story is that, while future events often rhyme with accounts pulled from the history books, things almost never play out in precisely the same way twice. We should expect to be surprised, at least a little bit and perhaps a lot.

Argument #5 Against Valuation-Informed Indexing — Today’s Numbers Are Just Too Shocking To Be True.

Many react to the numbers reported by The Return Predictor with shock. Too much funny talk about efficient markets has led some to believe that stocks always offer the best long-term value proposition.

This I do not buy. I do not see how any asset class can always be the best choice. We bid stocks up to insane price levels in the late 1990s. Where did the money come from? It was borrowed from the investors of today.

Yes, it is “irrational” that stock owners are being paid less for owning a risky asset than investors in Treasury Inflation-Protected Securities (TIPS) are being paid for owning a risk-free one. But this irrationality is just the flip side of the irrationality that saw the P/E10 value rise to an unprecedented 44 in January 2000. We brought into effect the irrationality of today’s stock performance the old-fashioned way — we earned it.

Argument #6 Against Valuation-Informed Indexing — If You Are Willing to Wait 30 Years, the Arguments for Conventional Indexing Hold Up.

The Return Predictor puts the most likely 30-year real annualized return for S&P stocks purchased today (this article was posted in November 2007, when the P/E10 level stood at 29) at 5.2 percent. That beats the return available through any of the safe asset classes by a big margin. It makes some sense to invest in stocks that portion of your portfolio for which you can truly wait 30 years to see an appealing return. Only you know what percentage that is for you.

Argument #7 Against Valuation-Informed Indexing — Economic Change Is Coming So Rapidly That the Historical Data Cannot Tell Us Much.

The Case Against Valuation-Informed Indexing

I see this as a reasonably good argument. I use the historical data as a guide because I am not willing to put money on the table without making reference to some sort of objective guidance and I view the historical data as the best source of objective guidance available to me. But we are indeed living in a time of rapid change. That cuts against placing too much reliance on any historical-based guidance.

Argument #8 Against Valuation-Informed Indexing — Investors Lack the Emotional Maturity to Follow This Approach.

Some of the “experts” who argue for maintaining the same stock allocation at all valuation levels make this argument. It could be that there’s something to it. We won’t know how people implement the principles of Valuation-Informed Indexing until a large number give it a try.

I have more confidence in my fellow middle-class investors than the experts putting forward this argument evidence by doing so. The biggest problem faced by most middle-class investors is that their common sense is telling them one thing (that prices must matter) and most “experts” are telling them something different (that there is some efficient market pixie dust that causes stocks to perform differently than any other asset that is bought and sold). I don’t believe that most investors are today emotionally prepared to execute Valuation-Informed Indexing successfully. If the “experts” regularly put forward more realistic advice, I think that many investors would over time come to grow in confidence that a buy-and-hold approach really can work in the real world.

Argument #9 Against Valuation-Informed Indexing — Investors Lack the Intelligence to Follow This Approach.

Lots of investors are making a terrible mistake to maintain a high stock allocation at today’s prices. Are they dumb? I see no evidence of it in the Post Archives of The Great Safe Withdrawal Rate Debate. My take is that investors are being misled by poor advice from the “experts,” which is put forward as a means of appeasing the small group of investors that is emotionally invested in the rationalizations that became popular during the wild bull market.

Valuation-Informed Indexing is in an important sense easier to follow than the conventional indexing approach. In a surface sense, it is slightly more complex because those following it need to make decisions as to when to make adjustments in their stock allocations. The other side of the story is that conventional indexing puts investors at war with their common sense while the idea of buying more stocks when they offer a stronger long-term value proposition is in accord with how we make purchases of all other assets. The new approach causes less emotional turmoil and thus is actually easier to understand for investors who have come to see the holes in the Efficient Market Theory.

Argument #10 Against Valuation-Informed Indexing — Investors Will Be Tempted to Take Extreme Positions.

This is a legitimate concern, in my view. The historical data makes a case for investors going with a zero stock allocation at times of high valuations. I think that is too extreme a position; stock prices can continue to go up for a long time even after reaching very high levels. Advocates of Valuation-Informed Indexing need to stress the need for investors following it to take moderate and balanced positions. We know that stocks perform poorly in the long term when purchased at times of high valuations; we are not able to make good predictions of when prices will return to reasonable levels.

Argument #11 Against Valuation-Informed Indexing — Most Investors Lack the Patience to Wait 10 Years for Their Expectations to be Realized.

Again, this is a legitimate concern. We know that long-term strategies pay off. We also know that they are hard to pull off. We need to do more to encourage investors to tune out the media-generated noise reporting on the short-term performance of stocks and to focus on the long-term realities. The first step to getting to where we need to go is reporting the long-term realities (including those that make stocks look like a bad deal at extremely high prices) in clear and understandable and actionable terms. The more that investors know about how stocks work in the real world, the more able they will become to handle the emotional ups and downs of long-term investing.

Argument #12 Against Valuation-Informed Indexing — The Experts Do Not Endorse This Approach.

The Over-Confident Investor

Most middle-class investors have limited time to devote to learning about investing. Most rely on the advice of “experts.” Most of today’s experts have let us all down with their unwillingness to address the valuations question squarely. We need a new kind of expert, one possessing a richer and deeper understanding of the role played by emotions in the setting of stock prices.

For now, it is a reality that the big-name experts do not endorse this approach (although a good number do endorse the principles on which Valuation-Informed Indexing is based).

Argument #13 Against Valuation-Informed Indexing — This Approach Has Not Been Tested in the Real World.

This is a good point. A measure of caution is certainly appropriate when considering a new investing strategy.

Argument #14 Against Valuation-Informed Indexing — The Market Has Remained at High Valuations for Over 10 Years Now.

This blows people’s minds. It certainly is a legitimate point; stocks have been at very high valuation levels for over 10 years. The historical data shows that it sometimes takes 10 years or even a bit longer for prices to return to reasonable levels. Still, prices have stayed higher for longer this time around than ever before. That reality provides some legitimate grounds for holding doubts about the Valuation-Informed Indexing approach.

Argument #15 Against Valuation-Informed Indexing — What Sounds Too Good to Be True Probably Really Is Too Good to Be True.

It sounds wonderful to be able to form reasonable assessments of how stocks will perform 10 years out and 20 years out. My sense is that a good number of investors see our claim that it is possible to predict long-term returns with a reasonable degree of accuracy as a claim that is too good to be true.

Investors who see these claims as too good to be true need to study the data and and the research and the articles backing this new approach to investing. Those who do not understand the reasons why predicting long-term returns is possible will not possess enough confidence to stick with the strategy when it does not appear to be working.

Argument #16 Against Valuation-Informed Indexing — There is a Potential Opportunity Cost Associated with Being Out of Stocks for a Long Time.

If stock prices remain high enough for long enough a time, it is possible to imagine a scenario in which it would have been better to have maintained a high stock allocation despite high prices.

My sense from my study of the historical data is that the odds against these scenarios playing out in the real world are long indeed. John Walter Russell’s research has provided valuable insights re this point. But I would like to see a greater amount of informed discussion of this point than I have seen to date.

Argument #17 Against Valuation-Informed Indexing — It Is Suspicious That This Approach Relies on Use of a Rarely Used Valuation Metric (P/E10).

Worst-Case Scenario

We use the P/E10 valuation metric to determine when stocks offer a strong long-term value proposition and when they do not. Most people have never heard of P/E10. This naturally makes them reluctant to convert.

Those who study valuations have known of the defects of the P/E1 metric for a long time. I view it as a scandal that many “experts” regularly cite P/E1 to this day, despite its known flaws. P/E10 is no doubt an imperfect tool. But I have yet to hear an argument that I find compelling for why it should not replace P/E1 as the most commonly cited valuation assessment tool.

Other reasonable valuation metrics (Tobins’s “Q” and the Gordon Equation, for example) generally also show stocks to be dangerously overvalued today. That lends credibility to the P/E10 tool, in my view.

Argument #18 Against Valuation-Informed Indexing — The Leading Advocate of This Approach Has Been Banned from Several Internet Discussion Boards.

That’s me!

In ordinary circumstances, being banned should be viewed as a bad sign. The circumstances that apply here are extraordinary. While I have been banned from several boards, I have never put forward a post that is in any way, shape or form abusive. The fact that I have been banned for the “crime” of reporting accurately what the historical data says about the effect of valuations on long-term returns should be viewed as an argument in favor of Valuation-Informed Indexing rather than an argument against it. The fact that advocates of conventional indexing feel so unable to formulate an effective argument for their positions as to feel required to seek bans on honest and informed posting on this topic reveals them to be holding an extremely weak hand.

Argument #19 Against Valuation-Informed Indexing — Few Investors Invest Solely in the S&P Index.

The S&P index is often used as a proxy for U.S. stocks in general. It is certainly true that investors who are not invested solely in the S&P index will see results different from those that will apply for the S&P. There is more data available for the S&P index, so researchers often must focus on that index. Knowing how the S&P is likely to perform provides useful but highly imperfect guidance as to how other stock indexes are likely to perform.

Argument #20 Against Valuation-Informed Indexing — If This Worked, Surely We Would Have Heard About It A Long Time Ago.

What's Wrong with Indexing?

It truly does seem that someone other than me should have come forward with these ideas a long, long time ago. From one way of looking at things, a good number did so. For example, Benjamin Graham, author of Security Analysis, put forward similar ideas decades ago. So perhaps there really is nothing new under the sun. My take is that the popularity enjoyed by the Efficient Market Theory during the wild bull of the 1980s and 1990s caused most experts to “forget” the essentials of common-sense investing for the exceedingly strange stretch of stock-market history that we happen to be living through today.

My contribution was to apply ideas that long had been viewed as nothing more than common sense to Bogle’s indexing concept and thereby to transform it into strategy far more likely to pay off in the real world.