Wednesday, January 27, 2010

Don't Listen to Me. I Don't Listen to Me.

Earlier this week I spoke with a friend and happened to mention that I think we are due for a substantial correction in the stock market. I am one of those who was expecting a fairly strong economic recovery to take hold, but over the last month or two I have been less optimistic. For one thing, I think that the stock market rally is overdone. For another, I feel that things are getting somewhat shakier, but I can't quite place my finger on the reason for this. It feels like there is another shoe out there somewhere and it may be getting ready to drop. Sovereign debt default somewhere in the world (Greece? Iceland?); an overheated Chinese economy cooling down? Continued weakness in the US housing sector? Or is it simply the fact that the stock market bounced up too sharply and too high over the past 10 months? I have no idea. I just know that I don't feel all that well about the prospects for the economy in the short term.

I previously shared these concerns with my friend, but to my surprise he seems to have acted on my doubts. He told me that after he recently lost his job, he rolled his 401K into an IRA but is holding the money in a money market fund, waiting for whatever pull-back in the stock market to occur before putting it back to work.

I was aghast. "Are you insane?" I asked him. Why would he listen to me? I don't listen to me. Yes, I talk about my feelings regarding the economy. I speculate about investment strategies and about the state of the stock market, but I never follow my own advice. My money is invested in index funds, precisely because I don't trust my own judgment and hunches enough to put my money where my mouth is. In fact, if he were to listen to me about anything it should be about the following statement: you can't beat the market, and market timing is a particularly bad form of guessing. The investment strategy that I follow is a simple one: small, regular purchases into index funds that cover my desired asset allocation.

After our talk, my friend decided to divide the money in his newly rolled out IRA into 12 equal parts, and to move one part per month into his chosen asset allocation. Now that's a smarter decision. Seriously people, I am not qualified to give investment advice. Almost no one is.

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

8 comments:

Edwin said...

It's quite a tough issue and not one person could ever grasp the economy and all it's workings.

I tend to be very pessimistic about the future and expect no real recover (more jobs) for at least a year. The stock market was mainly boosted by inventories and the stimulus which had it's biggest effects in 2009. So I wouldn't place to much belief in it recovering because of that.

Rob Bennett said...

There are lots of people who say what you say here, Shadox. They say that no one knows how stocks are going to do and that therefore the best thing to do is to put a set amount into stocks regularly and just stop thinking about what stocks are going to do.

My view is that this is the most dangerous of all possible strategies.

The problem is suggested by the story you tell in your post. You don't believe that stock prices can be predicted. Yet you cannot help mulling over in your mind the possibilities. You have no confidence in your conclusions. Yet you continue generating them. Why?

I think it is because there is a part of your thought process (it's called "common sense") that rejects the idea that stock returns cannot be predicted. You KNOW that they cannot be predicted (all the experts say this). And yet you also on some other layer of consciousness know that they CAN be predicted. If stocks prices were 100 percent unpredictable, stock investing would be 100 percent gambling. So. as strongly as you believe that stock prices can never be predicted, you ALSO believe that stock prices can always be predicted. You are caught in an endless loop of confusion and doubt and fear and uncertainty.

The only way out (in my view!) is to figure out why it is that just about all of us believe these two totally contradictory things (that stock prices are predictable and that stock prices are not predictable). We believe them both because there is a mountain of evidence for both propositions. Monkeys have been doing as well as stock pickers since the beginning of time. And those who ignore price have been seeing the wealth of a lifetime disappear in a series of crashes since the beginning of time. How can both of these things be true when they are opposite things?

I believe that the explanation of the puzzle is that short-term predictions never work because in the short term stock prices are determined almost entirely by emotion and long-term predictions always work because in the long term stock prices are determined almost entirely by the economic realities. I look forward to the day when we have a national debate on these questions, leave aside the guesses that led to development of the Buy-and-Hold Model and move on to the development of a model for understanding how stock investing really works in the real world.

You don't keep making these guesses because you are dumb, you keep making these guesses because you are too smart to really believe that stock prices are set by a random number generator in the sky. There are millions of us doing the same thing. And we need to get to a point where we accept that maybe it is the "experts" who got it wrong when they told us that stock prices are not highly predictable.

The question we should be looking at is not whether stock prices are predictable or not (they MUST be if they reflect predictable economic realities) but how to best go about predicting stock returns. We could all spend a lot of time productively exploring that question if only we could drop this idea that there is some sort of "study" somewhere showing that there are not (there has of course never been such a study and never can be one).

Rob

Carlyle said...

I think your friend did a very wise thing, gradually invest in stocks to mitigate the pain involved should there be a short-term drop in stock prices. No one can predict stock prices in the short-term. Everyone is counting on long-term stock prices to be higher X numbers of years from todays price.

Both you and your friend should read Bill Schulthesis' Coffeehouse Investor blog, paying particular attention to The Coffeehouse Credo. The important thing is to be diversified across several asset classes and rebalance periodically.

Shadox said...

Carlyle - I completely agree.

Edwin said...

Interesting thoughts Rob, and I totally agree. I think the reason Shadox didn't bring it up is because he would have to make a post as long as your reply to really get into it. And it just isn't realistic to inflate a blog post multiple times just to make sure you include all contingencies.

Surely the stock market isn't 100% unpredictable, or 100% predictable. The more information and knowledge you have, the closer you can come to predicting stock values.

The efficiency market hypothesis itself is based almost solely on the belief that a certain amount of information concerning a company's value is built into the stock price.

Of course this just isn't realistic. As you discuss in your most recent blog post about the buy and hold model, it's silly to think that markets are efficient, perfect, and the like.

Anyway, after some rambling my main point is that the more information you can get about something that is involved in a stock's value, the more accurate your predictions can be, both in the short and long term.

For both short and long term you have to deal with an insane wealth of information that can clutter up your analysis. Everything from the intimate workings of the company to the politics affecting the company and its suppliers.

The long term becomes even more problematic as you have to include even more unpredictable variables like the state of the company's economy and political environment to even random weather occurrences that themselves can affect a bunch of things.

I just want to highlight that while it is possible to spend your whole work life collecting this information and trying to model for it (as stock an analysts do) but even then you might be lucky to match the reality of it by even 50% (arbitrary number).

Rob Bennett said...

The long term becomes even more problematic

Thanks both for your kind words and for your illuminating comments, Edwin.

The words quoted above are the one point re which I think you and I might disagree. I believe that it is far easier to predict long-term stock returns than it is to predict short-term stock returns.

Many people think of stock return predictions as being like weather reports. We can do a good job of predicting what the weather will be tomorrow. We cannot do a good job of predicting what it will be two weeks from today. A lot of us are inclined to believe that, if it's not possible to predict where stock returns will be one year from today, it's got to be even more impossible to predict where stock returns will be ten year from today.

But it's not so! Ten-year predictions are highly doable. Ten-year predictions done using reasonable means always work! There has never been a single exception in the historical record.

How could that be?

It's because the stock market works more in the way that a slot machine works than it works in the way that weather patterns work. Does the house always win when a gambler pulls the lever only 10 times? It does not. Payout percentages are actually pretty darn high in most casinos. If you pull the lever only 10 times, you have a decent chance of ending up ahead. But the odds favor the house in the long run. Pull that lever 1,000 times and it is a virtual certainty that you will end up losers.

That's how stock investing works, according to the entire historical record (we have data going back to 1870). Lots of people think that they can invest their retirement money in stocks at the prices that prevailed from 1996 through 2008 and have at least a decent chance of either making money or not losing money in the long run. After all, prices are not predictable in the short term. If prices are not predictable, there cannot be that much risk in investing in stocks even at times of insane overvaluation, can there?

No! The risk is off the charts when you buy stocks at those sorts of prices. Never in the history of the U.S. market have those investing in stocks at those sorts of prices ended up doing okay or well. Why? Because when stocks are selling at those sorts of prices, the odds are just so much against the stock investor that he cannot hold a realistic chance of beating them long term. Thinking that you will end up okay when you buy stocks at those prices is like a gambler thinking that maybe he will get lucky and beat the slot machine by playing it all day. This sort of thing is an extreme long-shot bet. It's gambling, not investing.

The primary cause of today's economic crisis is that we have been teaching middle-class investors precisely the OPPOSITE of what works in stock investing for over three decades now. We have been telling people that timing isn't necessary or even that timing might be a bad idea. The reality is that timing is REQUIRED for those hoping to have some realistic chance of long-term success.

The confusion came about because of the finding in the 1960s and 1970s that short-term timing doesn't work. When they learned that, the academics jumped to the hasty conclusion that long-term timing might not work either. The research on long-term timing was not done until 1981 and that research shows that long-term timing ALWAYS works. But The Stock Sellling Industry had already spent millions promoting Buy-and-Hold and the thought caught on that perhaps it would be better not to let middle-class investors know the full realities.

Today's economic crisis is the long-term result of this very, very bad decision on the part of those who make their living selling stocks to us and who are often cited in the press as "experts" in this field. The true experts are those who do not permit the financial rewards that come from telling people not to worry about overpricing to stop them from reporting the full realities, in my assessment.

Rob

Edwin said...

Rob, we don't disagree at all on my statement. I was just not as good at communicating it as I could have been. I'm working from a more theoretical standpoint of "perfect information.

I mean that in the short term, most of the information affecting a stock can potentially be knowable. While in the long term there are more events that are just impossible to predict that could affect the stock.

Of course you are right that in reality over the long run stocks can be expected to rise due to rises in the productivity of an economy. But as you say, this is no excuse to promote buying a stock at whatever value it holds. That's just lunacy.

Rob Bennett said...

I mean that in the short term, most of the information affecting a stock can potentially be knowable. While in the long term there are more events that are just impossible to predict that could affect the stock.

I think it depends on whether we are talking about an individual stock or a share in a broad index fund, Edwin. With a single company, you're right that someone who did the proper research would have a good take on the company's short-term prospects but perhaps would be less clear about long-term effects.

It doesn't work that way with index funds, though (all of my work is focused on indexing). With index funds, you have no way of knowing how the index is going to do in the short term (because stock prices are set almost entirely by investor emotion in the short term and investor emotion is not predictable). But in the long term the price of an index fund is always going to reflect the economic realities. It has to or the entire market would collapse.

The strategic implications of this are huge. This means that those investing in index funds can know in advance when stocks are worth investing in and when they are not. For example, at the top of the bubble (January 2000), the most likely annualized 10-year return on stocks was a negative 1 percent real and Treasury Inflation-Protected Securities were paying 4 percent real. You could have increased your annual return by 5 full percentage point for 10 years running (that's a combined gain of 50 percent of your initial portfolio amount!) just by being willing to ignore the advice from The Stock Selling Industry to follow a Buy-and-Hold approach (to keep your stock allocation constant at all times).

I am grateful for the back-and-forth, Edwin. You are bringing up great points. If you want to know more about Valuation-Informed Indexing (the opposite of the Buy-and-Hold approach), there's lot of materials at my site. I'm also of course happy to try to answer more questions, if you have more. Shadox has some Guest Blog Entries from me on this in his archives and also a few critiques that he posted in which he describes the parts he agrees with and the parts that he has doubts about.

Rob