Morningstar Advisor - August/September 2009 - 33

Leverage. This is where you bring in your active strategies to express a view. Tactical could be a hedge fund play or a go-anywhere manager. The insurance is Vineer Bhansali’s tail-risk hedging strategies at PIMCO, or it could be an advance life-delayed annuity contract for a retiree. My point is that if you want to express views, it’s OK. We’re human. Our clients, ourselves, the PMs we give money to, all have biases built into their decision process. But we should try to sequester the lion’s share of our portfolios away from these biases. Don’t let your views overarch the core. Kaplan: Moshe Milevsky of York University in Toronto has a book called Are You a Stock or a Bond? What Milevsky is saying is that whether your human capital—the present discounted value of all of your future income—is more correlated with stocks or with bonds depends on the nature of sources of income. A tenured professor is a “bond” because like a bond her future income is at a known fixed level. In contrast, a stock trader is a “stock” because the value of his career is highly correlated with the stock market. Is MPT a general-enough framework for us to deal with that adequately? Fox: To the extent that you can define something in terms of measurable return and risk, it can fit into the MPT framework. As soon as you introduce multiperiod structures and cash-flow requirements, you’re stressing its limits. But in general, to get a broad, adequate picture of what a portfolio should look like, sure there’s room. Kaplan: What about for a one-over-N’er? Falk: Human capital is critical to the equation. Falk: All depends on how you create the inputs and the value of those securities. It’s one thing to model Social Security, which is fairly consistent. If we’re considering an individual a stock or a bond, what happens when the person loses his or her job or changes career paths or starts a company? You’ll have to make changes in how you calculate that. Kaplan: Nevertheless, when those changes occur, would you change the asset mix or not? Falk: Typically, I don’t change the asset mix. I use it as a tool in the rebalancing argument, because the reality is, we’re not selling anywhere nearly as much bond to buy stock as clients think we are, because their bond asset is significantly larger than they think it is. Fox: Did I hear you right? Did you say that Street. Because again, we’re talking about something that we think we can count on. It’s a challenge. The human capital asset can be very big, and the accuracy of forecasts about its value is critical due to the inordinate impact they can have on the allocation. The allocation framework is where you get to a lot of the art of this business. Kaplan: When we do asset allocation, we’re always looking for what else can we put in a portfolio that’s not going to go down at the same time that stocks and bonds go down. We’ve just gone through an event where pretty much most of the asset classes went down. Is there really such a thing as a non-correlated asset class? Fox: An uncorrelated asset is a myth. It doesn’t exist. An example I use is oil and large-cap equity. It goes through fits and starts where correlation goes between plus 0.5 and minus one. It depends on which window you’re looking at, as to how big that number is. Falk: Not only doesn’t it exist, but if it’s ever actually found, we would ruin it. Think about this for just a second—the pursuit of alpha, assuming that actually does exist, versus beta. We’ve moved into esoteric asset classes. We’ve moved into potential illiquid asset classes that the early adopters could extract additional returns, alpha. Once those asset classes have reached a level of popularity and have become more liquid through other investment vehicles—ETFs, ETNs, etc.—the pursuit becomes beta. It is no longer alpha. Fox: Your ability to gain returns, disproportionately to the risk inherent in it, goes away. Falk: Yes. Kaplan: On that point of agreement, let’s you don’t change a portfolio allocation when a client’s circumstances change? Falk: One-over-N is the core. Why would you reweight a purposely equally weighted allocation? The explorer portion of the portfolio is where you will reshape a portfolio based upon clients’ changes. The challenge here, as we think of classic risk tolerance, is how we then produce allocations. Risk tolerance is two things: risk preference and risk capacity. Risk preference may be consistent, other than the fact that investors are loss averse. Risk capacity is what we’re talking about when clients’ goals change. Do we change the portfolios? Do we change the definition of the goal? How much should they be saving? This gets into managing expectations. Fox: Because the university professor who When you expand the asset picture from paper assets, stocks and bonds, and real assets, you think of the human capital, Social Security, and pension assets. What we find is that people have a much bigger portfolio than they think they do, and it is heavily weighted towards fixed income. Kaplan: Is that a problem for MPT? goes from MIT to Wall Street thinking he can be a trader suddenly moves from being a bond to a stock. I think that should fundamentally shift your perspective of what the portfolio allocation should be. That’s an easy translation to make, in terms of the structured decision you’d get from MPT. Falk: if you think he’s going to stay on Wall conclude the session. Paul D. Kaplan, Ph.D., CFA, is Morningstar’s vice president of quantitative research and a frequent contributor to the magazine. MorningstarAdvisor.com 33

http://www.MorningstarAdvisor.com

Morningstar Advisor - August/September 2009

Table of Contents for the Digital Edition of Morningstar Advisor - August/September 2009