Wednesday, May 06, 2009

Passive Investing is for Extremists: The Critique

Yesterday Rob of a Rich Life posted what I thought was a thoughtful and well written guest post on this blog, making the point that passive investing is a fool's game. I would like make a few comments regarding this post and its underlying assumptions. 

First, it's important note that the term "passive investing" can be defined in a number of different ways. I think of passive investing as investment in Index funds vs. stock picking or investments in actively managed funds. I have written numerous articles on the subject and academic studies indicate that for the vast majority of investors, and over the long run, index investing yields much better returns than do the actively managed alternatives. However, Rob is not differentiating between these different stock investment strateiges. His main claim relates no so much to how you invest in stocks, but rather to the percentage of your portfolio that is invested in this asset class, regardless of which stocks or stock funds you put your money into. I think that it is more correct to say that Rob is against passive asset allocation, than he is against passive investing as I understand it.

With that in mind, let's discuss the main point. The underlying assumption of Rob's post is that when stocks are overpriced, investors are better off aggressively under weighting stocks in their portfolio, and supposedly the opposite is true when stocks are under-priced. Essentially Rob is advocating a form of long-term market timing. While this may make sense in theory, I am not clear that it can be accomplished in practice. For one thing, short term market timing is notoriously difficult to get right. In fact, my post tomorrow will deal with exactly that issue. Why make the assumption that investors will be better at predicting the long term peaks and troughs in the market than they are able to predict short term ones?

Another important question that needs answering is what metric is used to determine whether stocks are overpriced or whether they offer good values. The price to earnings ratio is commonly used for this purpose, however that indicator is far from straight forward. For example, at the peak of the economic cycle, just as stocks are getting set-up for a fall, earnings are often at their highest levels, sometimes making stocks appear modestly priced. The reverse is true at the bottom of a cycle when earnings can be dismal. Valuing stocks by more complex models (cash flows, CAPM, or whatever else you may favor) adds layers of complexity and murkiness to the analysis - not to mention making it inaccessible to most of the population.

Oh, and one more thing. There is an entire industry of investment advisors out there that use sophisticated (and presumably meaningful) quantitative tools to make investment decisions. These professionals can move assets between stocks, bonds and often other asset classes. Some of them can even bet against the stock market by taking short positions. Do these guys do any better than the rest of the investment community? Nope. The fact that you have a quantitative model doesn't mean that it has any predictive power. Ask some of the hedge fund managers who got crushed in the recent meltdown.

I am not ruling out Rob's theory, although I think that he has a heavy burden of proof to overcome. I think that there are clearly very appealing aspects to his theory. However, I am not convinced that this theory can be put into profitable practice when everything is taken into consideration: transaction costs, potentially increased tax burdens, and most importantly the ability to correctly gauge whether the market is correctly valued. At the end of the day, investing in stocks over the truly long-haul will get you about a 6% premium above what you can get for putting your money into treasuries.Maybe you can come up with a better theory that will yield higher returns, but I am betting it's unlikely. I'll be glad to be proven wrong.

Here are a few other investment related posts I found around the PF community:

All Financial Matters is commenting on the performance of target date funds in light of a new SEC investigation. GenX Finance also looks at target date funds and shows off some numbers (Yikes).

My First Million is betting that the stock market rally will stall, he says "sell".

Weakonomics tells his readers to ignore the financial media and uses the current rally as an example. Also on Weakonomics, catch the latest Carnival of Personal Finance.

Enjoyed this post? Please consider subscribing to Money and Such by free RSS Feed or by email. You can also follow me on Twitter.


Rob Bennett said...


I've been writing about these issues on the internet on a daily basis for seven years now. It has been a rare event in that time for me to see someone coming at these questions from the other side to respond to my presentation of them in a manner as intelligent and balanced as yours. I am not a flatterer. I say that in all sincerity. I am very impressed.

First, you agreed to run the guest blog entry even though you had doubts about the validity of the argument. Second, you proposed writing a critique of the argument after it appeared. That was just wonderful. It is of course possible that I am wrong in all that I say (it's been known to happen). Still, there is a critical need for these questions to be discussed. Even if I am wrong, we all need to find out WHY if we are to continue to have confidence in our strategies. To do that, we need to talk things over in a spirit of mutual respect and affection. By writing a critique, you encourage further discussion. Third, your critique is well-argued and, more importantly, FAIRLY argued. There is not one word that you put forward that I take exception to.

You are absolutely correct that the thing that I am finding fault with is Passive Asset Allocation. I love indexing. I myself advocate a strategy that I call "Valuation-Informed Indexing."

You are also right (in my view) to criticize short-term timing. I personally believe that the case is very strong that short-term timing does not work.

Finally, you are right to be skeptical about long-term timing. I believe very strongly that long-term timing works. But it would be a mistake to buy into the idea (which has not been widely explored as of today) too easily. You are putting forward the proper mix of cautious openness to new ideas and appropriate skepticism as to whether those ideas stand up to scrutiny.

I am grateful for what you have done to expose these ideas to your readers.

If you have an interest, I would be thrilled to continue the dialog. I have explored these issues in great depth. I would love to have an opportunity to put forward additional guest blog entries on The Case for Long-Term Timing and to have you continue to critique them from the perspective of a fair-minded skeptic. I of course understand if you do not want to proceed with such an arrangement. But if you have an interest, I think we could get some interesting discussions going and help people on both sides to sharpen their thinking re these matters.

In any event, thanks for what you have already done. It cheers me to see someone respond to the ideas in a manner as intelligent and fair-minded as you have.


Rob Bennett said...

I reread the critique and thought that I should flesh out my proposal just a bit by pointing out that I would address each of the four main points made in your critique in follow-up guest blog entries:

1) Why believe that long-term timing works better than short-term timing?

2) What valuation metric should be used?

3) How can the average investor expect to do better at this than most of today's professionals?; and

4) Is the 6 percent real long-term return expectation of those following the Passive Asset Allocation Model realistic?

These are all important questions. The best way to address them is to analyze each of them separately and invite community members to join in the discussions and offer their own takes. My thought is that breaking things up in this way would help us all Learn Together.

I have seen great interest in these questions at many places that I have posted. It is my belief that if we could manage to have a civil and reasoned discussion of these questions, we could help not only readers of this blog come to a better understanding of the issues. We could also help other blog owners, who would be able to turn to the series of discussions of the different topics at play to help them form their own assessments of the overall question of whether taking price into consideration when setting one's stock allocation makes sense or not.


Shadox said...

Let's go for it.

Rob Bennett said...


I'll send you a Guest Blog Entry for your review on the topic of why long-term timing is more likely to work than short-term timing within the next few days.