So far as I understand the conceptual basis for a lot of theorems in finance, one of the ideas seems to be reasonably straightforward: if some type of investment — domestic equities, foreign equities, bonds, housing, whatever — were systematically higher-yielding than some other type of investment, then everyone would just invest in the higher-yielding category and wipe out the difference in yield. So in the long run, you’d expect yields across asset classes to equilibrate.

Yes, this is based on assumptions, which might well be false. But let’s assume that it’s roughly true. Then the usual argument says that you can’t beat the market, and you’ll never do better than to diversify your portfolio. But note carefully what “diversify” means here. It *doesn’t* just mean “invest in all 500 stocks in the S&P 500”. There’s a whole lot more market out there! That is, there are a lot more asset classes than just large industrial stocks of the sort that the S&P traffics in. Even within the class of U.S. stocks, there are larger indexes like the Wilshire 5000. Or there’s the set of stocks tracked by the Vanguard Total Stock Market Index Fund. And then there are foreign stocks. And then government bonds. And municipal bonds. And corporate bonds.

But there’s *still* a lot more market out there. Some asset classes are harder to invest in than others, like houses. Wouldn’t it be cool if you could buy housing across many worldwide markets? It’s a little weird to imagine exactly how that would work, though large property owners do own a shocking amount of property across many cities. So in our imagined perfectly diversified portfolio, we’d have a bunch of housing. And we’d also have a lot of unlisted stocks. And we’d have some private equity. And we’d own some mines. Because again, the principle behind the diversification is that if everyone already knew that some asset class yielded outsized returns, they’d already be investing in it. The only way to beat the market in such a case is to know something that others don’t — to avoid some asset class that you know yields lower returns than the rest of the world thinks it does, or to invest in an asset class that others are systematically avoiding. In principle, I can’t see any reason why your buddy Doug’s new venture doesn’t count as an asset class for the purposes of this argument.

So here’s my question: how do I, J. Random Small Investor, get my portfolio fully diversified across all possible asset classes? One of Piketty’s observations in his masterwork is that the wealthy are able to obtain systematically above-market returns, in no small part because they can invest in asset classes that you and I don’t have access to. And in one example he gives — the Harvard endowment — it helps that the Harvard Corporation spends $100 million every year to manage $30 billion in assets. You and I could probably also do very well as investors if we spent all day every day managing our investments, and if we had a staff to do it, and if we had enough money to play with that we could offset some losing bets with more winning bets.

But that argument isn’t convincing to me, because we *do* have that ability; this is why we hire mutual funds. Apparently the Vanguard Total Stock Market Index Fund has $190 billion in assets. Granted, that particular fund won’t be investing in obscure corners of the asset universe, but why doesn’t Vanguard set up a fund that’s truly diverse across all asset classes, draw many billions of dollars from investors like me, and earn the same returns as the Harvard endowment or Bill Gates?

One possible answer is that regulation forbids them from investing in risky asset classes (like hedge funds or complicated swaps) if non-rich guys like me are on the other end of the trade. Is there some other reason I’m missing why wealthy people *must* earn higher returns than mutual funds do?