Felix Salmon

Paul Volcker famously says that the ATM was the only financial innovation he can think of that has improved society. Well, I can think of another one: the target-date fund. Virtually every honest personal-finance expert will agree that the best investment strategy, if you’re saving for retirement, is to buy up some index funds, hold on to them, and maybe rebalance occasionally. But doing that is non-trivial, which is where the target-date funds come in: they do it all for you, and if you choose a good vendor, like Vanguard, they do so at an astonishingly low cost.

So far so good — but the fact is that ATMs have improved steadily over time, and there’s no particular reason why automatically-rebalancing retirement investing shouldn’t improve over time as well. Because for all that target-date funds are wonderful things, there are a couple of weaknesses with them.

Firstly, despite the fact that target-date funds are designed to be increasingly risk-averse as the investor approaches retirement, they can still lose a lot money near that target retirement date. During the financial crisis, for instance, the Vanguard 2010 fund fell 26% in 12 months, and other funds fell much more. That’s partly because no one is agreed on how much stock-market exposure investors should have when they retire: in a world where retirement can last for decades, it’s surely substantial — but that also means a significant risk of large losses.

Secondly, target-date funds generally have only two components: stocks and bonds. In terms of maximizing risk-adjusted returns, that’s problematic: the more factors you get to choose from, the more likely you are to be able to reduce your risk without reducing your returns. And more to the point, we simply don’t live in a world any more where stocks are the risky things and bonds are the not-risky things. With interest rates where they are, and sovereign finances where they are, there’s a strong case to be made that at some point between now and your retirement, bonds are going to prove significantly more risky than stocks.

Recently I talked to a couple of people who reckon that they’ve managed to improve on what until now has been the gold standard — the Vanguard target-date fund. The first was Vineer Bhansali, who runs Pimco’s target-date funds, which it calls RealRetirement. The idea behind these funds is that they diversify according to risk, rather than according to asset class: they take a bunch of different asset classes, disaggregate the various risks, and then work from there. For instance, if you want exposure to commodities or real estate, and buy stock in mining or housing companies, then you’re getting equity exposure at the same time.

At the same time, the funds actively try to minimize what they call drawdown: the maximum amount that a fund is likely to fall, from peak to trough. Here’s a slightly scary chart from one of Bhansali’s papers:

The equity beta of the stock market is 1, so these lines all represent fractions of what you would see if you had all your money invested in the stock market. Take the olive-green line representing an equity beta of 0.6: that represents a portfolio with three-fifths the volatility of the stock market as a whole. If you follow a standard 60/40 strategy of 60% stocks and 40% bonds, you’re probably not far from there.

What the green line shows is that if you hold that kind of portfolio for 30 years, you can expect your maximum drawdown to be north of 40%. That is, at some point over the course of your investing years, your portfolio will fall by 40% from its highs.

Now we all just went through the financial crisis: we remember exactly what it feels like to lose 40% of your money. It’s not nice. And so Pimco spends some serious money — somewhere between 15bp and 50bp per year — on tail-risk hedging. As a result, says Bhansali, when stocks fall by 50%, his funds will only fall by 20% or 25%. Which is a key level, because research shows that when people lose more than 30% of their money, they tend to panic and sell. The Pimco funds are designed with the idea that you should never lose more money than you can afford to lose, even on a mark-to-market basis, and that your fund will never fall so much that you will end up doing that very human thing, selling at the bottom of the market. Pimco can’t force you not to sell, of course, but the tail-risk hedging will do its bit to help you stay the course during the crisis or two that will inevitably come along at some point.

The way that Pimco does that changes over time — just as the composition of the funds does. When short-term TIPS had a 1% real yield, they were a fantastic place to put risk-averse money: it was guaranteed to grow at more than the inflation rate over time. Now that the real yield on short-term TIPS is guaranteed to underperform inflation by more than 1%, however, they’re much less attractive, and the asset allocation has moved elsewhere.

I do like a lot of what Pimco is doing here. For one thing, it’s concentrating on risk rather than asset classes, which makes a lot of sense to me: as bonds become riskier, it’s silly to stick to an investment strategy which assumes they’re largely risk-free. And it also has a large dollop of behavioral economics: it knows where investors’ pain points are, and endeavors not to hit them. The point is to concentrate on investor returns rather than investment returns: there’s no point in embarking on a strategy which will give you a great payoff in 30 years’ time if you panic when a crisis hits at Year 20, and sell the whole thing at a multi-decade low.

The Pimco funds might not give you the very best returns, then: by cutting off the lower tail, they necessarily give up a certain amount of performance. But there’s more to investment than maximizing returns, and in fact the Pimco funds are looking pretty good right now, since they generally outperformed during the crisis.

The other company doing interesting things with passive investment strategies is Wealthfront, which today announced that Burton Malkiel (yes, that Burton Malkiel, and yes, he’s still going strong at 80) has joined as Chief Investment Officer.

Wealthfront doesn’t offer investible funds: instead it’s an investment manager, which puts your money into a mixture of six different asset classes, allocated according to all manner of wonky Modern Portfolio Theory algorithms. (They love nothing more, for instance, than to talk about the finer points of the Black-Litterman model, which they fully sign on to.) Their threshhold-based rebalancing is about as sophisticated as rebalancing gets, and the way they allocate your money is smart too: they know that people nearly always overestimate their own risk tolerance, and they adjust for that when investing.

Wealthfront is great for small portfolios, because it doesn’t charge any fees at all on your first $25,000; it’s great for larger ones, too, since it provides free tax-loss harvesting once your portfolio goes over $100,000. This is a sophisticated strategy which really can help your after-tax returns: the idea is that if your stocks, say, go down while the other bits of your portfolio go up, you can sell your stocks to claim a loss which offsets the gains you have elsewhere. What that means is that you sell the S&P 500 ETF you’re invested in, which you don’t really mind doing at all: because all index funds are much of a muchness, you just buy a different S&P 500 ETF instead.*

Wealthfront claims that by using six asset classes rather than two, it can boost returns by 45bp per year on average, and that by layering on tax-loss harvesting as well, you can add another 100bp to that. Both of which make Wealthfront’s maximum fee of 25bp look decidedly modest; it’s certainly lower than anything offered by traditional investment managers.

The Wealthfront service is mainly being targeted at young, male Silicon Valley geeks in the first instance.* These are people who, it turns out, tend not to be saving for retirement: their needs are more short-term, and they’re more interested in when they’ll be able to buy a house, or whether they have enough money to quit their job and take a year or two off. With those kind of risk profiles it’s very hard to pick a traditional target-date fund: after all, target-date funds are designed explicitly for retirement funds rather than for invested funds you might need long before retirement. And in general, an investment manager like Wealthfront is always going to be more personalized than a faceless fund from Pimco or Vanguard.

That said, Wealthfront is emphatically not an old-fashioned investment manager who will sit down with you and talk to you and build up a human friendship over time. There’s a good reason that when traditional brokers move from one company to another they tend to take their accounts with them: it’s that the investor’s relationship is with the individual broker, rather than with the broker’s employer. At Wealthfront, that’s not the case: they don’t have highly-paid brokers, and the relationship, insofar as it exists, is basically with a website. You can see why they’re starting in Silicon Valley.

Both Pimco and Wealthfront, then, are offering intriguing products which in some ways improve on the stripped-down simplicity of the Vanguard ETF. Complexity is never a good thing in and of itself, and both of them are using financial technologies which the end investor will almost certainly not understand. I don’t have a lot of faith in those technologies: when Wealthfront CEO Andy Rachleff launched into his spiel by telling me how deeply his firm’s philosophy was rooted in Modern Portfolio Theory, I didn’t exactly light up with enthusiasm.

But the fact is that individual investors are not very good at managing their risks, and that both Pimco and Wealthfront are likely to be better at it than they are. There’s nothing at all wrong with a Vanguard target-date fund: it’s a wonderful product. But no financial product is right for everybody. And I’m very glad that now we’re seeing the arrival of other financial services which can credibly claim to be just as good if not better.

Update: I had the tax-loss harvesting strategy a little bit wrong: you don’t swap from one ETF into another based on exactly the same index, since the IRS doesn’t consider that to really be selling your position. You just move into a similar-but-not-too-similar index for a month or so. As for the young males, I’m told that some 20% of Wealthfront’s customers are now female. Which is still low, but the number’s rising.