Vanguard's Davis: Expect Muted Returns for Bonds

Joe Davis: Thank you, Christine.

Benz: So you and your team arrive at expected market returns for various asset classes, and I'd like to talk about some of them, but first I want to talk about how you arrive at those expectations, what do you look at?

Davis: The process – well, it's a mix of quantitative approach, as well as qualitative approach. The key to understand for your listeners is that we produce a range of outcomes: there's no point forecast to this. It's very statistical, it's actually driven by various risk factors and some historical relationships, but we also build in a great deal of uncertainty. Statistically speaking, things such as fat tails and other things that we know historically can lead to, in times of stress, correlations going to one.

And if anything, we go well and beyond historical record to see what potentially could have happened that did not. So again, we're doing the best we can to incorporate uncertainty, but that said, we also look very closely at the central tendencies as well to get some sense of the implications they may have for portfolio construction.

Benz: So I want to start by talking about fixed income. Your team has relatively muted expectations for fixed income. Let's talk about what you're looking at there? And I'd also like to home in on TIPS [Treasury Inflation Protected Securities] and what you're thinking about for TIPS?

Davis: Sure. Just broadly, first, fixed income. One of the key attributes of our simulation returns is that they incorporate initial conditions, which traditional Monte Carlo or simulation tools explicitly do not, and that's a key distinction.

Now when you do that, we all know that interest rates, generally speaking, are much lower today than they have been, and it's those initial conditions that matter. And the biggest reason why we have below historic-like returns, I mean, yield to maturity is actually a pretty decent predictor of future bond returns.

And we also do incorporate some higher yields over a long period of time, and so because in part of that reversion, you tend to have below yield like returns, at least as the central tendency. That's broadly for all asset classes, whether its TIPS, as well as just Treasuries or corporate bonds.

Benz: Okay. So from a practical standpoint, investors looking at this feel like they're between a rock and a hard place. They know they need the security of fixed income, but you've got rock-bottom yields and the prospect of higher rates. How should you navigate?

Davis: Yeah. Well, I mean again, I think it depends upon how big of a percentage of fixed income one has in the portfolio and really is it: a) that the key reason for fixed income is income generation, or is it b) more diversification from a more volatile asset class such as stocks. And depending upon the answer to that, as well as one's aversion to potential losses in the near-term principal volatility, that's where you have to work through that exercise.

I think, broadly speaking, though, regardless of whether one answers a) or b) on that spectrum, I think, we all as fixed income investors are going to have to become a little bit more comfortable with principal volatility if we even hope to get close to the returns we've had over the past 30 or 40 years. I mean, the past 10 years, we've had equity-like returns in fixed income as interest rates have proven so. Unfortunately, that's the place that we don't stand today.

I don't think we will be at those levels permanently for a 10-year basis, but that is a headwind from a nominal return perspective. Real returns is a little bit different.

Benz: Okay. So, on the more positive side, though, for equity market returns, you are actually fairly positive, expecting returns in the range of historical norms?

Davis: Which is – again, that can take some by surprise given the economic headwinds we're facing here, consensus expectations are fairly low. I mean bottom line is, we were actually projecting as a central tendency a somewhat higher-than-average equity risk premium, which means expected returns of stocks over bonds or even cash.

Biggest reason for that is the one factor that matters, particularly over five-, 10-, 15-year horizons, is valuations. Any other predictive signal, whether it's the steepness of the yield curve, credit spreads, realised volatility, them and a number of indicators that have some influence on near-term returns, they all drop out. And the only thing that matters is the price you pay for growth not expected growth per se.

And so, when you do that, the central tendency is historic-like; why? It's because valuations broadly speaking are also near their historical averages. They are not grossly high like they were in the late '90s nor grossly depressed like the '70s and early '80s. It's more in the middle. And so that shouldn't be a surprise as to that's where the formula and approach leads to a more historic-like return.

Benz: And you had mentioned to me that even though the markets have run up a bit over the past few months that you are still comfortable with sort of that 8% to 12%?

Davis: Yeah. Again, the 8% to 12%, that's a central tendency. I mean statistically that's the most likely outcome, but the actual – of the 10,000 simulations we run – it's about 25%. So there is a wide distribution to that, but that said, it is still – what would change that, would be if valuations really start to meaningfully deviate at the present time from where they are, and unless that happens, I won't expect material changes in that sort of expected risk premium for investors.

And I think, if anything, it's ironic that the highest returns historically for stocks have been in years when the unemployment rate is high rather than when its low. Again, it gets back to – it's not growth per se, it's a price you pay for growth.

Benz: So, I want to talk about two other categories, commodities and commercial real estate, and you have noted that you are expecting sort of below-average returns for those categories?

Davis: For commercial real estate, the prospects have improved a little bit, and by that we believe that over long periods of time, the return on a broad REIT Index will be very commensurate with commercial real estate, which again some investors may have exposure to through direct private property.

But if we look at REIT as an asset class for commercial real estate, the valuations have improved so much and stabilised so much so that the valuations look more historic-like closer to where they have been with respect to US or international equities. So, the prospects there improved a little bit.

For commodities, where we stand there is, we believe that investors should expect the positive risk premium to commodities futures.

Benz: Okay.

Davis: Commodity futures and this is not spot price. I think there is a lot of confusion out there in the broad marketplace. Now that said, we recognise that it's actually very difficult to have as much confidence and what that expected risk premium should be versus, say, even other asset classes, which already have a wide distribution. Biggest reason: not really a valuation metric to tie down commodity future returns.

So, if anything, we take a more conservative approach. We are unsure if the changes in the institutional marketplace and investor behaviour may influence that expected return – again, unclear.

So in part because of that and also the fact that we've seen a rise in correlation over the past five years, markedly so, between commodity futures as well as stock returns, if anything, we would be just more conservative on that front. And those have actually the widest distribution of future returns of all the asset classes that we consider.

Benz: Okay. Well, Joe, thanks so much for sharing your insight. It's very helpful.

Davis: Thank you, Christine.

Benz: Thanks for watching. I'm Christine Benz for Morninstar.com