Executive Summary
In the standard framework of portfolio management, changing a client's exposure to risk is essentially analogous to changing their overall exposure to risk assets. Want conservative growth? Invest in a portfolio with 40% equities and 60% fixed. Want a more moderate growth portfolio? Increase to 60% equities. More aggressive growth? Allocate your portfolio further towards an equity tilt. At its core, the proposition is pretty straightforward: increase your overall portfolio allocation increasingly towards risk assets to increase the overall risk (and hopefully, return) profile of the portfolio. But what if there was another way to increase overall risk? What if, rather than increasing overall risk by adding a little risk to the whole portfolio, the risk was increased by adding a lot of volatility to a very small portion of the money?
The inspiration for today's blog post comes from a recent email correspondence I've had with Professor Meir Statman (author of What Investors Really Want), and some concepts tying in with Statman's Behavioral Portfolio Theory work with Professor Hersh Shefrin. In particular, Statman posted a very interesting a basic question about risk management and exposure from a (Fidelity) risk questionnaire that I found fascinating:
If you could increase your chances of having a more comfortable retirement by taking more risk, would you:
a) Be willing to take a little more risk with all of your money?
b) Be willing to take a lot more risk with some of your money?
The standard process by which most planners increase client risk is essentially option (a) - to increase the overall equity allocation of the portfolio, taking a little more risk based on the overall allocation. (Of course, in practice this may involve rotating a portion of the portfolio into equities, but imagine for the time being the advisor implements the change by selling a single conservative growth balanced fund to buy a moderate growth balanced fund.)
Yet in practice, when the average consumer is surveyed about how they would rather take on risk, the response is actually strongly in favor of (b). The reason seems to anchor around our tendency for mental accounting: to compartmentalize buckets of money into separate groups and account for them separately, even though they are ultimately all part of the same personal balance sheet. For instance, when we think about our brokerage account as a separate asset from the account with grandma's inheritance, and investment them differently, it's because we're mentally accounting for grandma's inheritance in a different manner than the brokerage account, even though ultimately both are theoretically interchangeable accounts and/or could be merged at any time.
In the context of risk investing, the tendency for option (b) is a reflection of this mental accounting behavior. We feel more comfortable knowing that the majority of our money is "safe", and it makes us more comfortable to take greater risk with the small remainder - even if we ultimately end out with a portfolio that, in the aggregate, has the exact same risk profile as a more balanced moderate risk portfolio.
Yet despite our behavioral tendency to prefer portfolios of this nature, it's often not how we invest for clients. Arguably, the concept of "core and satellite" investing is somewhat similar - we have a steady core that we account for in one manner, and then we invest more creatively and with more risk in the satellite positions, taking potentially much greater risk with a very small portion of the portfolio. However, even in the context of core and satellite investing, if a client wanted to increase the risk of the portfolio, most advisors would likely increase the amount of risk in the core, not just buy even-more-aggressive small satellite positions.
In truth, to really invest in this manner could entail a dramatic change in portfolio construction. Instead of holding balanced portfolios of fixed and equity/risk assets, the client might hold an overwhelming majority of assets in cash, and complement it with "ultra" risky assets, such as the most volatile equity investments, or perhaps even options or leveraged exchanged-traded funds or mutual funds (e.g., 2X or even 3X leveraged funds).
This is not to say that all clients should go out and buy ultra-risky investments immediately. But the underlying point is still there - if the behavioral research suggests that clients actually prefer portfolio risk management approach (b) over (a), how might you invest your portfolios differently for clients?
So what do you think? Have you ever seen your clients express a preference for (b) over (a), perhaps implied with some of their investment decisions? If this was the "preferred" way of investing, how might you guide clients differently about their investments?
Michael. Great post. De-risking is a concept that your readers might find useful when looking to alter the risk profile of the portfolio while still making it something the clients can stomach. My post is a little long winded, but I’ll describe how advisors can reduce the risks in the fixed income allocation to open up appetite for risk on the equity side.
As you know, Asset Dedication espouses the concept of mental accounting. In our strategy we split the portfolio into 2 sub-portfolios, Income and Growth, each dedicated to a specific purpose. The Income Portfolio is an ultra-safe, fully immunized bond portfolio matched to the client’s financial plan. The Growth Portfolio is made up of equities and other assets with long-term higher expected return.
We de-risk the fixed income allocation in the Income Portfolio by using a pension fund style/liability driven investing approach where we match the bond maturities and coupon payments to the cash flows identified in the client’s financial plan. Here are some of the risks that are managed or mitigated: interest rate (portfolio is fully immunized); default (agency bonds and CD’s backed by the US Govt in IRA accounts and AAA rated munis in taxable accounts if clients are comfortable with the very low default rate); inflation (use the assumed rate in financial plan cash flows or hedge further using individual TIPS).
Assuming the portfolio is held as designed, the downside is limited to the weighted average of the YTM of the bonds (the portfolio’s IRR). Portfolio return is always positive and is known at the time of implementation. Even if the portfolio loses value along the way, the bonds are held to maturity so the losses are never realized. Unfortunately for bond fund investors, the mathematics of total return in a rising interest rate cannot protect the downside and funds will lose money or lag behind the YTM of the underlying bonds because the portfolios turn over (average is 175% turnover for high quality funds). No way around it. In a rising rate environment, bond funds bring added downside risk into the portfolio that can be eliminated by using an income-matching approach.
By de-risking the bond allocation, protecting the downside and delivering predictable income over the client’s chose time horizon, we have seen an increased appetite for risk in the equities in their Growth Portfolio. By keeping the purposes of the portfolios distinct and putting a time buffer in the Income Portfolio, clients seem to be able to ride out periods of stock market turmoil because they have set aside 8-10 years of “paycheck” in their Income Portfolio. They are not worried about where their Growth Portfolio is tomorrow and instead focus on where it will be in 8-10 years. This approach helps client say the course and stay invested so they can experience the recovery in 2009 and 2010 instead of bailing in 2008.
To use another behavioral concept, it’s all about framing. Income is the near-term portfolio and Growth is the long-term portfolio. Income is made up of bonds (not bond funds), which are predictable, but don’t have good long-term expected return. Growth is made up equities et al with higher long-term expected return, but lots of short-term downside risk. Find a comfortable time horizon of protected income for the near-term and clients can take on more volatility in the long run knowing that they have time for it to pay off.
Excellent post. I prefer that each asset in a portfolio meet a test for quality rather than have a mix of, for example, 95% cash and 5% futures contract deposit, or 80% A paper bonds and 20% junk bonds. I think if the bad stuff blows up the client would over-react emotionally. The client could get upset at the low risk part of the portfolio and want to take on too much risk or lose confidence with the adviser. Of course after testing each asset the portfolio’s components must be harmonious, meaning hopefully uncorrelated assets, with the proper total level of risk for the client and with many other strategic design concepts such as “asset location”.
How do you “design” something durable in a worldthat is changeable, where the future cannot be predicted with the kind of accuracy that we model for financial portfolios? I don’t think you can. But that should not stop us from trying. As much as I try to get clients to set goals so that I may “need-match their portfolios, so do they resist. So what do we do? We focus on our knowledge of them as people, their responsibilities, their fears, their dreams. Admittedly, the process is not overly logical but it works for us, I think. I think planners-managers tend to think in terms ofasset classes, as Michael mentioned, for risk on and risk off. Why? Because they use mutual funds and ETF’s. We use individual securities for the most part. Thus, for us, we can more finely tune risk to our perception of the market environment. Risk on means to ratchet up the beta and, perhaps, to diminish the market cap and liquidity of our portfolio. Riskoff means to find more stable equities and to raise the bond quality/lower the duration. But all of what I discuss is sausage-making. If it results in a predictably tasty sausage, our clients will keep coming for more – e.g. will be happy with our work. This said, we are constantly engaged in explanation of our activities and insights in an effort to “educate” and to bring clients along – most of whom have not a clue of the work we do. They just want to eat the sausage. “
Very Zvi Bodie, Michael. I like it.
I don’t favor this particular approach. But I see it as a very good sign that people are beginning to look at new portfolio allocation strategies. The first step to arriving at good answers is asking good questions.
Rob