Bogle: The Past Is Not Prologue

Don Phillips: You made a point about international diversification, and I questioned that you had thought all you needed was the U.S. stock market, total market portfolio. You didn't need all the international funds. And your point was that while statistically it lowered your standard deviation, it really didn't lower your risk, because when you really get into a crisis, everything goes down.

John Bogle: Yeah.

Phillips: I think that's a lesson that you were citing in 2000 that a lot of investors had wished they had taken hold of before last year.

My question to you, though, is a little different. Is there another bit of conventional wisdom that you sort of bucked the convention on today that you think investors don't know that may be helpful for them to know in the future?

Bogle: Yeah. If I tell you I'm deeply into a subject and I'm going to give a major speech on it somewhere down the road, that's usually the kind of speech it's going to be. And I have one in mind.

That is, I think we give far more credence to past returns in the stock market than they even remotely deserve. The past is not prologue. The stock market is not an actuarial table. As has been mentioned, and John Maynard Keynes told us back in 1936--and I put a lot of numbers on this--it's the sources of the returns, the kinds of things that I was talking about a moment ago, that determine future returns, not the recurrence of an ongoing event that comes up on a standard probability distribution curve.

<TRANSCRIPT>

And of course we've all heard about the ends of that curve in the form of "black swans" as they're called, totally unpredictable events that come out at the very tails of those standard distribution curves, that throw everything out of kilter.

So rely on reasonable expectations for the future based on the sources of stock and bond returns and not the past. If you agree with me, there goes the Monte Carlo simulation, because it's based on past returns.

But think with me about this for a minute. That Monte Carlo simulation has a 7.5%--let's call it divided by 12, every month that we put that data into the computer--it has about 4/10 of 1% of yield. What if the yield on stocks is 1%. Well, that 0.4% per month, or 4.5% a year, is irrelevant. It's gone. You're using a new, much lower number. And yet in the Monte Carlo it's as if the yield is still 4.5%. It's not. And a lot flows from that.

The growth of earnings will probably be similar to the past, I agree on that. And finally, the P/Es are really quite unpredictable. Except there's something about PEs I think we know, and that is if they're under 10, they're much more likely to go up than down. And if they're over 25, they are much more likely to go down than up.

And it's on that basis that I would kind of construct the returns that I was trying to have my client understand for the future.

So I'm going to get this all in a speech, and I've said a lot about it, but this is going to be really a good one. This is going to be kind of a new way of looking at it. And I'm going to call it after a quote from Samuel Taylor Coleridge: "History Is But a Lantern Over the Stern." It shows you where you've been but not where you're going.