Une approche dynamique de l'allocation d'actifs (VIDEO)

Ben, thank you for being here.

Ben Inker: Very glad to.

Benz: Today, your talk was about standard glide paths for target-date funds. When you look upon them, you feel that there are significant shortcomings in the glide paths that target-date funds are using. Let's talk about what you think are the shortcomings, what are the risks of glide-pathing today.

Inker: I think the biggest shortcoming of the target-date funds, as they exist today, is they make this underlying assumption that the expected returns to assets are constant through time. And that turns out to be a really false assumption. Starting valuations really matter. That's most obvious when you are looking at fixed-income instruments: At the end of the day, if you own a 10-year bond for 10 years, you know what the return is going to be, and today it's 2%. And in 2000, it was 6%. To expect the same return from those two bonds is simply wrong.

We think the same thing works with stocks with regard to their valuation. And so the basic problem is it assumes that the right solution is only dependent upon your age. If you're 65, you should have 52% in stocks and 48% in bonds. We think if you're 65 and stocks and bonds are priced normally, that's true; if you're 65 in the year 2000 and bonds are cheap and stocks are expensive, you shouldn't have anything like that much in stocks. Rationally, you should probably have zero. And by the same token, if you're in 1981, you should have a ton of money in stocks. So, what we really think is missing from target-date funds is the recognition that the environment you are in and the valuations you are facing really should impact the portfolio you have today.

Benz: So specifically, GMO believes that valuation should be a key driver in what a person's asset allocation looks like at a given life stage. From a practical standpoint, can you talk about some of the inputs into GMO's valuation modeling and how individual investors might think about using some of those same ideas when positioning their own asset allocations?

Inker: You can get caught up in trying to come up with very specific models of valuation and, quite honestly, we spent a lot of time agonizing over exactly what the right valuation metric is, but there are some pretty simple ones that get you most of the way. So, with regard to bonds, the simplest thing is to ask what the real yield is today. What is the yield less expected inflation? And that's information that you can get pretty readily. With stocks, the cyclically adjusted P/E, or the Shiller P/E, is a simple valuation metric that you can access. Actually, Robert Shiller has a website where, every month, you can see what the Shiller P/E is, and that's a good basic valuation metric to help you understand whether stocks are expensive today or cheap.

And once you've got that basic information, you can try to adjust your portfolio in favor of whichever one is looking better at a given point in time. You don't have to move your portfolio that aggressively to have a material impact. You don't have to be that incredibly precise in your understanding of the valuation, because these valuation swings can be really big, and sometimes stocks are priced at stupid levels, and sometimes bonds are priced at wonderful levels. And moving even a moderate amount of your portfolio in the right direction can have big impacts on retirement.

Benz: So, the problem today for people looking at their asset allocations, if you look at bonds, as you've said, it's not an especially inspiring-looking return going forward. Equity valuations, certainly in the U.S., are not particularly attractive either. So, say I'm a 30-year-old person attempting to construct a sane asset-allocation mix for my retirement maybe 35 years from now. What should my asset allocation look like based on your bottom-up assumptions?

Inker: To a decent extent, it really depends on how broad of an opportunity set you have. If your choices are U.S. stocks and U.S. bonds, it's a problem because neither one of them is particularly attractive. Right now, we would say, even if you're 30 years old, having more than a normal amount in bonds makes sense because U.S. stocks are so overpriced that we think they are actually probably going to lose to bonds over the next seven years. But it would be much better to have the opportunity to go outside the U.S. where stocks are cheaper--in Europe, Japan, and particularly in the emerging markets, where people have really given up some of these markets for dead and the valuation show it.

So, we think there are some opportunities to do better, and we would say if you can get yourself to move outside of the U.S. in equities, then having a normal equity weight makes sense. It's just that you want those equities not to be in the U.S. If you're thinking just with regard to U.S. stocks and U.S. bonds, we'd push moderately in favor of bonds today.

Benz: How about an older investor, maybe someone who is 60 or 65, getting very close to needing to start tapping that principle for their ongoing cash flows? What should they be thinking about when they setting their asset allocation?

Inker: Well, a lot of the same things. The one key difference is that if you are getting close to retirement or if you are at retirement, your ability to save additional money is very limited. So, I think for those who, say, are at retirement today, we can look at the expected returns of stocks and bonds and recognize that both of them are lower than history.

So, the unfortunate truth for someone retiring today is if the old rule of thumb was that you could afford in your first year to take out 5% of the total corpus, that doesn't work today because that assumes a level of expected returns that just doesn't exist. So, the most important thing is to recognize in your spending plans that you're going to have to spend a bit less because the expected returns are lower.

Otherwise, judging purely from a U.S. lens, again, having moderately more in bonds than stocks today make some sense; but in order to try to generate more return, the best thing one can do now is to take money out of U.S. stocks and put them in non-U.S. stocks.

Benz: Discussing the drawbacks of a dynamic asset-allocation strategy during retirement and leading up to retirement, can you talk about what you think are some of the biggest drawbacks, potentially?

Inker: The biggest drawback to trying to be dynamic is the fact that there will be times where you're going to look stupid relative to the traditional portfolio. So, I was saying U.S. stocks look lousy versus non-U.S. stocks; I could have said that a year ago. In fact, I did. And in 2014, U.S. stocks were up 15%--14% for the S&P 500--and non-U.S. stocks were basically flat. So, in the year 2014, I looked stupid.

Benz: So, you'll be early sometimes.

Inker: That's OK, because the nice thing about saving for retirement is you've got a very long time horizon. In one year, almost anything can happen; over 10 years, much less weird stuff can happen. And in over 20 years, it's even less so. So, the key problem with being dynamic is while you may have long-term goals for your saving, it's very difficult to get away from taking a short-term view of how you're doing. And when you are dynamic, the obvious thing to look at is what would've happened if you hadn't done this. There are going to be some times where not being dynamic will have done better. And if you're not prepared to stay the course, you shouldn't do it in the first place. What tends to happen to people who start down a path that they don't really believe in is that, when it goes wrong, they abandon it.

And so, for example, if you had underweighted U.S. stocks last year, and you say, "Well, that didn't work--I'm going back to the U.S." Then, you lost money in 2014 relative to doing nothing, and you're never going to make it up because you've forgone the potential return of owning what are now even cheaper non-U.S. stocks relative to U.S. stocks. So, taking more short-term volatility against whatever you are going to be measuring yourself to than you can stand is going to be a problem. Even though you may be thinking in terms of having 40 years to worry about or 50 years to worry about, if you are going to spend your time worrying about what happened last year, if you are going to be dynamic, you want to be dynamic at the margin.

Benz: So, behavioral factors are certainly in the mix. Ben, thank you so much for being here. Such an important topic. We appreciate you sharing your insights.

Inker: Thanks very much.