A Retirement Allocation Starter Kit

Note: We're refeaturing this video as part of Morningstar's January 2017 Guide to Saving for Retirement.

Christine Benz: Hi, I'm Christine Benz for Morningstar.com. It's Model Portfolio Week on Morningstar.com. Joining me to discuss Morningstar Investment Management's approach to asset allocation is Brian Huckstep--he is a senior portfolio manager for Morningstar Investment Management.

Brian, thank you so much for being here.

Brian Huckstep: Thanks for inviting me.

Benz: Brian, one thing that I've done in creating all of these model portfolios on Morningstar.com is that I've relied on Morningstar's Lifetime Allocation Indexes to inform the asset-class positioning of the various portfolios. I wanted you to be here to sit down and talk about how you put together those indexes because you're very involved in asset allocation for Morningstar Investment Management, as well as those indexes. So, let's talk about the overall philosophy, at a very high level, that guides those indexes' asset allocations.

Huckstep: The lifetime indexes glide from a more aggressive equity stance to a more conservative one--as do many target-date glide paths. We spend a lot of time working on the intra-asset-class allocations within there. So, not only do they glide from high equity levels to low equity levels as they get closer to the retirement target date, but within there, we're really thinking hard about those intra-asset allocation mixes and what the most optimal mixes are.

We're using a number of different optimizers. We use a traditional optimization approach with mean-variance optimization with forward-looking expectations for each asset class. It's taught in every portfolio-management class today in business school--trying to maximize return for each unit of risk. At the equity level, it's selective for the glide path. We know the expected volatility, and so we select those asset classes that maximize return. But then we go further. We have [mean-conditional value-at-risk] optimizers--is what we call it. We're trying to minimize the drawdown--the tail events. So, we have optimizers that do that, and they're looking at concepts like skew and kurtosis--those are the tail event.

So, if you think about 2008, we want to minimize the impact of those events on the portfolios. And then there are number of other optimizations we're doing. We do liability-driven optimization, which is a way to consider the liability in retirement. So, when people retire, the typical U.S. investor is spending his or her dollars in the U.S. They might go outside the U.S. or buy things from other places, but the majority of the money is going to be spent in U.S. So, we think about inflation and how that liability is changing over time. If inflation is high, that liability is going to be higher, in nominal terms. So, we are thinking about ways to hedge against that. So, liability-driven optimization is a way for us to build the best model we can to hedge against that inflation risk in the U.S.

<TRANSCRIPT>

Benz: One key piece of research that informs how you position these allocations is the role of human capital. Let's discuss, for most investors, how human capital should affect their portfolio's positioning?

Huckstep: When we talk about models, we don't want to just think about the financial assets. That's why we're helping people invest. Typically, I have a 401(k). I have some outside savings, and I need to think about how to invest that money. We'd like to think about the indexes we are providing as a great way to do that. But when you think about how to allocate those assets, you want to think about the assets that aren't there also. You want to think about what we call human capital--the majority of that is made up of money you haven't made yet. So, I'm 45, I have another 20 years to work, and I expect that salary. And that salary doesn't change a whole lot with the market. It's very bondlike.

So, when I think about that human capital and the money I need when I retire, I like to consider that asset. And we like to think of it, for the typical U.S. investor, as about 30% stocklike and 70% bondlike. So, when I think about the whole picture then, I include the human capital with the financial capital to think about that target allocation. Our team surveys investors, and we think about that level of human capital at each age. We're thinking about other things, too. We're starting to incorporate more and more of a total-wealth approach that includes the home, if you have a home. So, maybe because you have exposure to real estate there, maybe we reduce the allocation to real estate just a little bit when we consider that home and how the average U.S. investor is exposed to that asset class. That holistic approach helps guide where the financial assets are going to go, so it's got a big impact there for the typical U.S. investor.

Benz: You mentioned for the younger investor--the 45-year-old investor--they have very bondlike human capital, so they can expect relatively stable earnings throughout their lifetimes until they retire. But that human capital becomes a little more equitylike, then, as people get closer to retirement. That means that the fixed piece of the portfolio needs to step up in importance.

Huckstep: Yes. The human capital, in general, goes down. We have a graphic we can share that shows age on the horizontal axis and, on the vertical axis, it shows kind of the level. You can see the human capital goes down. So, the amount of "bond" that an investor has in their back pocket--that salary that they can earn in the future--it goes down. So, my overall exposure to bonds typically decreases over time, if it is the case that my human capital is 70% bondlike. We think about that when we build the portfolio. So, I need to ramp up my allocation to bonds as I get older--that's the upshot there. So, the whole portfolio changes over time. Typically, an investor has maybe more exposure to a home also, so we think about that when we consider REITs. It's a little part of the intra-asset class that has a small impact, but it's exciting to think about.

Benz: For those Lifetime Allocation Indexes, I noticed that you have aggressive, moderate, and conservative versions. So, if I'm looking at those indexes and thinking that this target date approximately matches my own retirement date, how then would I decide whether I should be more aggressively positioned for someone in my age band or less aggressively positioned? What are the key factors that would play into that decision-making? Or maybe it would be that I'm looking at these asset allocations and they're maybe not at all appropriate. Let's talk about the factors that would nudge me up or down in terms of my equity exposure.

Huckstep: It's complex question. But it's one that is important to tackle. Something that might nudge you up to make the aggressive model maybe more appropriate for you would be understanding that the aggressive model is more equity-centric. So, maybe if your human capital is more bondlike or maybe if you have a pension. We think about things like pensions. Maybe you used have a job before and you have a pension that's due to you--maybe you are in the military--and you know that money is pretty dependable. It's very bondlike, very steady. In that case, you would want your financial assets to be more equity-centric. So, you should bump yourself up to that aggressive glide path.

Maybe you've been a terrific saver. Maybe you've saved 20% of your salary every year, and maybe you've got a nice surplus built. So, if you have a target financial level of your savings and maybe you're ahead of that--maybe you have a surplus built up--that gives you the license to be a little bit more aggressive in your equity, also. So, you have the capacity to take on more risk, and maybe you just have preference to take on more risk.

Benz: I know that maybe in previous bear markets, I've been very placid--maybe I've even added to the unloved asset class. That's an indication that maybe I'm someone who can, from a risk-tolerance standpoint, take a little more equities in my portfolio.

Huckstep: Yes. Buying low is generally a good strategy. People who are able to do that, who do have that capacity or preference to take on more risk, typically can put themselves in a more aggressive model. And someone who, maybe in 2008, as soon as the market went down and they got a couple of quarters of those losses on their 401(k) statement or savings, found themselves saying, "Maybe I just don't have the stomach to stay in the market"--a conservative glide path is going to give them a much smoother ride. It's going to make them much more comfortable.

It's still a terrific model to get into. Over the long run, it has a slightly lower expected return, but it comes with that lower volatility and lower downside. It's just a smoother ride than the moderate or the aggressive. We built the moderate model for the typical U.S. investor. So, when we survey things like savings rates, financial balance, and the type of job you have, we think the moderate models are appropriate for most people. But certainly, there are people who are better served by the conservative or aggressive models.

Benz: I want to follow up on your comments on the intra-asset-class positioning. We've been talking about just setting your baseline stock/bond mix. But then when I drill into, say, the stock piece of my portfolio and the bond piece, your research--and certainly the indexes' allocations--would indicate that those elements, too, should change a little bit in terms of their complexion as I get closer to retirement. Let's talk about some of the key findings there. I noticed that Treasury Inflation-Protected Securities really aren't present at all in the portfolios for young accumulators, but they really step up for people who are getting close to retirement.

Huckstep: There are two parts that drive the majority of the differences there between that underlying asset mix. The first part is the liability-driven investing. In retirement, you want to hedge out inflation risk. So, if we have a period--which we hope we don't--where we go through something like the '70s again, in which we have double-digit interest rates and high inflation, if you're exposed to things that move with inflation and are going to act favorably in an high inflationary environment, you're going to glad you have those. So, as you get close to retirement, it's more important to have more of those things--like TIPS and commodities. Real estate is another one that moves relative to inflation. It kind of sticks with it. So, if you have high inflation, you get higher returns typically in those asset classes. As you get closer to retirement, it's more important to have those because you don't have multiple market cycles left to come back from that.

Now, if you are a younger person, we kind of ignore those a little bit. We're more focused on just worrying about that total return when you're younger because you have multiple market cycles to come back from a negative inflationary event. So, we don't really overweight those at all in the models for 2050-60--even the 2040 model has light amounts there as far as the target date today. Like I said, we include those a lot later.

Benz: So, the inflation hedges are stepping up, and that makes a lot of intuitive sense. How about within the equity piece of my portfolio? Would I tend to want to own maybe more small and mid-caps when I'm younger and maybe reduce that bias as I get older?

Huckstep: That's right. Just the general aggressiveness. You can be more aggressive when you are younger. Like I said, you have a lot of time to come back. If the market cycle is maybe seven years typically and you are 20 years old and you have 45 years to retirement, you have a number of market cycles to go through. So, you have the license and you have the capacity to really get out there and invest in things a little more aggressively. But as you get close to retirement, maybe you have only a few years left to retirement, or maybe you're in retirement, maybe you retired five years ago, your life expectancy is still around 20 years.

So, even in retirement, we have allocations to some of those assets because you hopefully do still have one or two market cycles, but you have less time to come back from that. So, we really start to think about the chance of shortfall. We have some Monte Carlo simulation tools. Monte Carlo is a technique to forecast out multiple events to really look at the downside and approximate the probability of a loss. As you get closer to really needing that money--needing it to be in that account--we don't want to put people in a position where they may have to withdraw money when the market is really down.

So, you are in more bondlike, safer assets; large cap generally has a lower standard deviation and lower volatility than small cap. High-yield bonds are more risky than short-term bonds. So, we move you to the more conservative asset classes. We think they are more appropriate as you get closer to needing that money because you don't want to have that shortfall. So, when we think about the amount of money you're putting at risk, that conditional money, or the potential for shortfall, we have multiple tools and optimizers to measure those. They lead us to be in the safer assets as you get closer to retirement.

Benz: We've been speaking mainly in the realm of strategic asset allocation. So, that's about setting my asset-allocation framework and then maybe gradually getting a little more conservative as I get older. When you look at asset allocation when you're putting together these indexes, do tactical factors affect the asset allocations in play at all?

Huckstep: Tactical is a word that has a lot of meanings. Slightly. When we build these models today, we have on the shelf multiple expectations for the market. We have 10-year expectations, 20-year expectations, and our infinite long-run expectations for each asset class. When we build these indexes, we're using a 20-year expectation for the market. As an example, today, for the S&P 500, our expectation is right around 7.7% over the next 20 years. Our long-run return is a little bit higher. Because valuations are a little higher than average today, we've taken that 20-year expectation down just a little bit. And for each asset class, that does impact how much we allocate--on the margin.

So, it's a slight valuation adjustment that we're focused on within the model. As far as turnover in these target-date models as you get older, that's really impacted by the glide path, as you glide it down. But year over year, we'll make small 1% or 2% tweaks to try to keep them on the efficient frontier--to try to make them as optimal as we can. So, we're not making tactical bets. I think a lot of people, when they think about tactical, they think 10% or 20% or 30% turnover: Maybe I'm trying to time the market--maybe I'm going to really reduce my long-term bonds today because I think rates are going to go up.

We're staying more strategic, and that's not a bad thing. I think we've put them almost right in the middle. So, there is a little bit of turnover--a couple of percent each year based on valuations--but not a lot. We're not really trying to time the market.

Benz: Brian, thank you so much for being here. I've been talking a lot about these indexes over the years. It's really great to hear your insights on how you construct them. Thanks again for being here.

Huckstep: Thanks.

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

* Disclosure
Morningstar's Investment Management group includes Morningstar Associates, LLC, Ibbotson Associates, Inc., and Morningstar Investment Services, Inc., all registered investment advisors and wholly owned subsidiaries of Morningstar, Inc. All investment advisory services described herein are provided by one or more of the U.S. Registered Investment Advisor subsidiaries. The Morningstar name and logo are registered marks of Morningstar, Inc.

The information, data, analyses, and opinions presented herein do not constitute investment advice; are provided as of the date written and solely for informational purposes only and therefore are not an offer to buy or sell a security; and are not warranted to be correct, complete or accurate. Past performance is not indicative and not a guarantee of future results.