Ken Fisher on Building a Well-Diversified Portfolio

Sometimes people ask me how do you build a diversified portfolio correctly and the answer presumes that already you've concluded you want to build a diversified portfolio. It steps over the issue of why a diversified portfolio is important versus not.

So some people will tell you that it's not, and right now I’m not going to get into that argument. When you want to build a diversified portfolio, you step back to the core concept of modern portfolio theory as created originally by Harry Markowitz in 1951 that looks at the notion of what's called mean variance optimization, which is a jargony term that really means to have things that have similar long-term return expectations, but where those components zig and zag against each other so the total portfolio has lower volatility relative to that long-term return. That's the construct. The construct is to blend together items that have short-term negative correlations but similar long-term returns so that you've got things that are zigging and zagging against each other this month providing greater portfolio stability, but that over the long term tend to get you to the same place as opposed to, let's say, having all tech stocks, which you know is not diversification, that tend to have okay returns in the long term but tend to have wild volatility in the short term. Or all railroad stocks, or all commodities, or all anything which gives you more short-term volatility even though it might get you to the long-term return, the purpose of the diversified portfolio is to have it be more stable toward that long-term goal.

So what does that mean? That means broader. What does broader mean? It means you don't just have one sector, you don't just have one industry, you don't just have one stock, you move off to the broadest indexes. Most people traditionally have thought about that as what's the broad accepted, in America, US index which would be of course the S&P 500 for most people's thinking. You could even go a little farther than that to some broader cap-weighted indexes for the United States. But even further, in my thinking, you say to yourself: In the very long term—and this is central to thinking about equities correctly—no major category can have superior returns to any other major category because eventually, as there's a perception that that category has a superior long-term return, companies and investment bankers will create enough new supply of that category to overwhelm that perception of superior demand, and that supply will meet the demand to bring the pricing back into line with other categories.

So if you look at very long-term returns like 30-year returns, categories tend to neutralize within a very, very tight bandwidth. Very tight between countries. I don't mean some little third world country that might do this but I mean between the broad US versus foreign, US versus Europe, broad categories. Returns tend, over the long term, to neutralize between sectors tend to neutralize not about a single stock not even necessarily a simple subset industry within a sector like buggy whips, but the broad sectors in that the broadest is to just think about the whole world and own the world index. It gives you the most things that zig when something else zags that keep you going toward that same 30-year return. So if you're building a diversified equity portfolio, the world index becomes the thing you build against and you match to it. Then you may choose to overweight and underweight certain sectors based on you thinking they will do better, and that's fine. You may even choose ones that you think have more zig and zag to them than some of the others.

But the central point is to build that zig and zag in, and yet with an index, whether it's in America only with the S&P 500 or whether it's with the broader world as a whole, the concept’s not to try to get a higher return. It's to get more things that are zigging and zagging against each other against a very long-term return goal. In that you then get to the simple point that you say: “Well if I’m going to increase these parts to decrease those parts, what's the basis for that? Why would I increase these parts and decrease those parts versus the index?” And the answer is: You think somehow, some way, you know something that isn't already priced into the stock market. Now that in and of itself is a fairly close to arrogant argument because, for the most part, when I ask myself: what do I know today that everybody else doesn't know? Most of the time the answer is nothing. And if you don't know anything that other people don't know the theoretical argument for passivity is almost perfect. You just buy the index, sit on it, you've got the diversified portfolio.

It's easy. It's easy and you will get the index return. You have to stop and ask yourself: Do I know something other people don't know? And the basis in portfolio theory for making an active bet is always and everywhere only that you do believe somehow, some way, you know something other people don't know it. And therefore this bet will pay off because if everybody else knows that they've already pre-priced it and therefore it won't pay off. So to build the diversified portfolio where you don't actually believe you know something other people don't know, passivity is the rule. Get the whole world, do it passively. It's cheap, it's easy, done. The rest of the time to try to improve on that and get a higher return relative to risk, relative to the benchmark of the world, or in America if you're doing America only the S&P 500, you have to go back and say: what do I know other people don't know? And you better be modest about answering that question because otherwise you get yourself into trouble.

Because in capital markets, overconfidence is the biggest killer as the Great Humiliator is always waiting to try to humiliate you, me, and everyone else for as much money as possible over as much time as possible because that's what the market does. So your final question becomes, and I reiterate: what do you know other people don't know? Otherwise passivity with a broad index is the easiest way to build a diversified portfolio.