A not-uncommon media narrative about finance goes like this: naive institutional investors have been tricked into putting their money into high-fee vehicles like hedge funds—even though these hedge funds don't outperform the market! This has been going on for a long time; here's a famous Businessweek cover from a decade ago, and here's the generally excellent Roger Lowenstein from just a few weeks ago.¹

The simple version of this argument compares two investment vehicles that people use to target roughly 10% annual returns:

1. They can write a large check to a multi-strategy fund, which will impose a long lockup. The fund will keep something approaching half of the gross returns generated. So on $1m in net gains, the investor is paying about $1m in fees. And the fund won't tell investors precisely what it's doing; they might offer a general breakdown, but investors will usually hear about big positions well after the fact. This fund may have separate vehicles for different investors, including some that are only available to the manager and a few select employees.

2. They can put the money in the Fidelity 500 index fund and pay a basis point and a half in fees.

The expected return for these two choices is actually pretty similar, at about 10% annualized. Some hedge funds run with a little more risk, or do a bit better; some do a bit worse. But 10% is not a bad anchoring point for thinking about their returns. As a result, whenever there's a bull market, the multi-strategy funds tend to underperform. In a bear market, mass media coverage of fund performance just doesn't get as much attention, because it's not as if the typical Vanguard investor can casually call up Millennium and ask to move their retirement account over.

This is a bit of a perpetual motion machine for financial media, because this situation is working out just fine for everyone involved, and so is unlikely to change. But it does look weird if the story omits key details.

The first detail, of course, is that there's a very big difference between a return of 10% +/- 5% and 10% +/- 20%. The 10% a multi-strategy fund aims for comes from, schematically, something like this: for every $1 of assets under management, lever up, say, 8x. On that $8, earn a gross return of 2.5%, for a total gross return on assets under management of 20%. This fund probably passes mosts of its costs through to investors, and those costs might amount to 1% of their gross assets under management (not a crazy budget for external research, external data, IT, back-office functions, and, of course, actually paying investment researchers, portfolio managers, traders, and the like). That modest fee looks like a much larger one when it's compared to equity, of course; it's 8% of investors' assets, every year. And then there's the performance fee on what's left. If that fee is 20%, a 2.5% gross unlevered return becomes a 20% gross return on assets under management, which shrinks to a 9.6% after-fee return.

Which investors are happy with, because, at least in theory, that's the return every year. That 2.5% is the spread between long and short positions that are hedged across asset classes, industries, and factors—not just a bet on stocks going up, or even a well-timed bet on oil stocks going up, but a portfolio that picks the best of the small-and-cheap oil companies and shorts the worst, and that similarly buys the best of the high-momentum large-cap oil companies and shorts the worst.² This is a labor-intensive job, and that's what the fund’s employees are being paid for.

Which is the right way to think about this broadly: asset management generates returns with some combination of labor and capital, and the rewards are roughly proportionate to those inputs. So questioning hedge funds based on their fees, rather than returns net of fees, is a bit like asking why anyone would invest in Accenture, which returns almost all of its revenue to employees in the form of salary, rather than buying shares of BP Prudhoe Bay Royalty Trust, which has zero employees and which distributes 98.2% of its revenues directly to shareholders.

We see this at extremes too, like HFT, where trading capital is a required input, but ends up being a small share of the cost. You can talk about an HFT strategy in terms of its returns on capital, but that’s measuring an input that just isn’t a constraint. It's a bit like looking at a company's revenue per megawatt: relevant for an aluminum smelter, crypto miner, or datacenter operator, but not a very meaningful number for an accounting firm, even though it's true that their revenue would drop if they didn't pay the electric bill.

The investors who prefer a steady-but-lowish return are typically endowments and pension funds (high net worth investors are in the mix, but they've been a shrinking factor over time). These investors expect continuous outflows of cash, so a big drawdown can be crippling: if an endowment is 100% in equities, and tries to spend a steady 5% of capital, then going through 2008 means either literally cutting spending in half—a school firing half its teachers and administrators, a pension defaulting on payments to retirees—or it ratchets its spending up to an unsustainable 10% of assets and has to dig its way out of the hole somehow. They'd rather accept a lower overall return if it didn't fluctuate much.³

If hedge funds are a better deal for these investors, where does that leave equities? Households saving for retirement and high net worth individuals are the ideal groups to take the risk of inevitable fluctuations in equity markets. If you're 45 years old, and you plan to retire at 65, a market crash that wipes out half of your retirement funds is very scary, but it's also something that inevitably happens every so often, and you can stay fully invested for the recovery. For high net-worth individuals, the general calculus is that 1) they are living well below their means, not because they're stingy but because they have so much money, and 2) if their source of wealth is a big investment in a company they work for or founded, they'll likely be overweight out of some combination of signaling and ego. It's fortunate for the world that the happiness-maximizing portfolio is not the profit-maximizing one, since it means that there will always be a cohort of rich people willing to make slightly suboptimal investing decisions; since alpha sums to zero before transaction costs, that's a source of upside for everyone else.

Read More in The Diff

We’ve covered this space in The Diff many times. For example:

• Here’s a piece on data indicating that high net worth investors get higher returns on their investments ($).

• We have a review of a great book on managing factor-neutral portfolios ($).

• Transaction costs push towards a world where most money is passive but active managers are more active. ($)

• And we’ve looked at the economics of “alpha capture,” another way big funds benefit from scale ($).

1. I owe Lowenstein in particular a debt of gratitude, because reading his Buffett biography in the summer of the year 2000 was a formative experience for me. Buffett has shown a praiseworthy aversion to fees, at least after he got enough capital. And it's true that when someone is worth that much, they will do just fine from capital appreciation without needing to layer on incentive-based compensation, lavish perks, etc. On the other hand, the story has been more complicated for a long, long time; one thing Lowenstein's book doesn't mention but that other Buffett biographies do is that when he launched his fund, he didn't include his personal assets, and put that money into less liquid and higher-risk investments. Nobody was complaining about a 30%-after-fees performance, of course, nor should they have. But Buffett, by his mid-twenties, had discovered something many other fund managers have since then: for a generally talented investor who finds opportunities at many sizes and volatilities, the wealth-maximizing approach is to put the higher-capacity trades into a vehicle with other people's money, while the capacity-constrained bets get made with the manager's personal funds.

2. In practice these exposures get hedged, imperfectly, across the portfolio, so it wouldn't be quite that narrow, especially because if you slice the categories with sufficient care, some of them have no members at all. So an energy and materials PM might be using a short position in a large-cap gas company to neutralize the large-cap exposure introduced by owning a large oil company. As long as the portfolios balance overall, it's a straightforward linear algebra problem, though there's a lot of work involved in getting to the point that everything is straightforward.

3. College endowments have another consideration, too: at a given size, they want to run a portfolio that's less risky than their peers. They can expect alumni donations to follow the market cycle, since their richest alumni will have a disproportionate amount of wealth in either literal equities or equity in private businesses. A school that loses less in a bad year can make opportunistic hires that pay off over time.

Join the discussion!