PART FOUR:

FOCUSING ON FUNCTIONS

MOVING BEYOND TRADITION

Our journey through the asset allocation process began with a fundamental idea: by spreading wealth across a wide variety of investment types, we create a portfolio with a higher return and lower risk than its component assets. This simple, yet powerful concept is rooted in common sense and life experience, making it highly relatable and intuitive.

We also examined the mathematical basis that guides the choice of investment types to include in an efficient portfolio - one that maximizes return while minimizing volatility risk. Our goal is a stable portfolio return that remains within an acceptable range as it moves through market cycles. This is essential to meeting each client’s monetary goals: providing sustainable withdrawals while preserving the portfolio’s real value so that future withdrawals have a high degree of certainty.

The mathematics behind diversification is relatively simple: combine complementary assets that tend to offset each other’s periods of disappointing performance. The ideal pairing of assets includes one asset that earns above-expectation returns while another experiences below-expectation results. These assets are known by their low correlation to each other, and they make ideal diversifiers. We apply this concept across all the portfolio’s assets, deriving a correlation between each asset and the overall portfolio. In this portfolio context, correlation is the percent of individual asset risk that remains in the diversified portfolio. We found this statistic to be relatively high for global equities, very low for high-quality bonds, and somewhere in-between for alternative assets. Using these insights, we created an efficient strategy by equalizing each asset’s contribution to return and to risk (subject to the unique constraints of the client.)

Could there be a weakness in this approach? Not in terms of the underlying mathematics – but what about our approach to grouping our investments into broader asset categories (i.e. stocks, bonds and alternatives?) It turns out that the objective factors that drive diversification and efficiency (i.e., return, volatility and correlation) are unrelated to how we group these assets. Might our arbitrary groupings be preventing us from achieving even better results?

In this chapter of our asset allocation series, we examine the effects of grouping assets by how they function, and by what benefits each asset grouping provides to the client. In this client-centered, goals-based approach, our efforts are directed toward maximizing favorable outcomes while minimizing harmful ones. This suggests taking a “functional” approach to how we evaluate the segments in our portfolio, grouping assets by how they behave, rather than what they are called.

STARTING WITH CLIENT GOALS

What benefits does a client expect from the investment portfolio? Designating these benefits as our goals is the first step in identifying the assets that we should include and how we should group them. We suggest the following portfolio goals:

• Liquidity: every client has immediate funding requirements, and the portfolio must provide easy access to stable amounts that match planned withdrawals (“spending”)
• Growth: this is the other end of the “time barbell” because future withdrawals depend on a growing portfolio value
• Inflation protection: the rising cost of living is one of the key challenges to a sustainable portfolio, and real investment growth depends on a return that exceeds inflation plus the spending rate
• Illiquid alpha: while asset allocation assumes only passive returns, there are a few unique market segments that provide an additional risk premium for the non-public nature of their longer-term commitments of capital

ALIGNING ASSETS WITH GOALS

We will leave liquidity for last, because this requires the alignment of portfolio outflows with planned withdrawals. (Plus, it is a fun process, and we are “saving the best for last!”)

Our view of assets is now focused on the benefit each provides. Each segment within the three traditional asset classes – stocks, bonds and alternatives – is “matched” to its function within the portfolio, finding its way into one of our four groups. These assignments may appear surprising – at first – but we will demonstrate their “fit” into this client-centered asset view.

OUR FUNCTIONAL GROUPS

Growth assets are distinguished by their high expected returns, high volatility, and relatively high correlation to each other and to the overall portfolio. This is especially true for global public equity. We also see four segments from the traditional bond group find their way into the growth category, bringing their unique returns and risk premia. These help to create a “package of risks” that provide growth along with some diversification benefit. These diversifying segments include high yield and emerging market bonds, which reflect a high degree of credit risk and equity-like, idiosyncratic behavior. Foreign bonds bring currency and geopolitical risk, each of which is a source of return, volatility and diversification. Securitized bonds being a risk premium from credit risk and the uncertainty of prepayments in response to interest rates. Using traditional asset groupings, our growth segment resembles a “70-30” balanced portfolio focused on growth. 

Inflation protection assets are essentially “the stuff that inflation is made of.” Over the long term, we can expect these assets to benefit from inflation, essentially acting as a hedge against this key risk. Treasury Inflation Protected Securities (TIPS) increase the payoff amount to compensate investors for unexpected inflation that occurs during the bond’s life. Global real estate protects against inflation in commercial buildings and housing. Commodities include agricultural products, livestock, mining and energy, and we separate gold as a hedge against both currency declines and geopolitical uncertainty. We note that these inflation-protection assets would typically be assigned to the categories of bonds and alternative investments. This shows the arbitrary nature of that traditional perspective, and we contrast this with the highly-intuitive nature of our assignment to their function within the portfolio, and the significant benefit these assets provide to the investment client.

Illiquid Alpha assets are essentially collections of stocks, bonds and strategies that may not be publicly available. These are active products that carry both a market return plus an excess return to compensate the investor for both the long-term commitment of capital to these funds, plus the benefit of the investment decisions of the fund managers.

Liquidity is provided through a Cash Matching portfolio segment made up of high-quality bonds (typically Treasuries and corporate bonds) with maturities spanning the number of years of required liquidity. In our example, the client goal is to “pre-fund” five years of expected payments starting with 5% of the portfolio’s value, with these withdrawals increasing with inflation (assumed to be 2.5 percent.) By combining short and intermediate-maturity bonds, this asset segment creates a “cash flow” schedule that matches the expected withdrawals. The money for these withdrawals comes from interest payments on the bonds, plus maturities over the 5-year withdrawal period. (In our example, the long Treasury segment is not used, since its bonds extend beyond the longest withdrawal.)

Certainty of funding is produced by the contractual nature of the payments of interest and the repayments of the loans at their maturity dates. Our client is now “immunized” against the risks of changes in interest rates, as well as from downside volatility of the growth segments in the portfolio. Over this 5-year period, the client can allow the riskier, growth-oriented assets to be left alone to earn their expected returns, without fear of needing to sell them during a market downturn. The client can be more risk-tolerant, knowing that the next 5 years of withdrawals are unaffected by downside market volatility.

A simplistic visualization of this cash match segment is a “bond ladder” – in this case, one zero-coupon Treasury bond (“STRIP”) for each payment, with each bond’s maturity value equal to the withdrawal amount it must fund. But this “set it and forget it” approach is the least-efficient way to accomplish the funding task. In practice, interest-bearing corporate bonds provide the required cash match without needing to allocate as much capital. We found that our 5-year spending schedule could be funded with about 22% of the portfolio’s assets.

Is this Cash Match portfolio a “free lunch” when it comes to liquidity? We think so.

• The bonds used are typically part of most diversified bond portfolios,
• they provide a substantially-higher return relative to holding cash,
• they provide significant diversification benefits, resulting in lower portfolio volatility.

In terms of portfolio efficiency, the cash match segment is a powerful diversifying asset.

Our simple example uses par bonds, where the coupon rate equals the required yield on each bond, and where the market value and par value of the bond are equal. We purchased a set of five individual corporate bonds, spending about 22% of our hypothetical $10 million portfolio.

We see that the cash match portfolio provides a bit more than the required amounts on each of the spending dates. If you look closely, you also see that we funded the initial payment of $500,000 with only $390,000 invested in our 1-year bond. Where did we get the additional $207,000 of cash flow? In addition to interest on the 1-year bond, we also collected the first-year interest payments on the other four bonds in the portfolio. This “coupon waterfall” lowered the amount we needed to invest in the earlier years of the withdrawal schedule. This shows how interest-bearing corporate bonds are superior to STRIPS.

This raises an important question:

“Why invest in the typical constant-duration bond portfolio that has its own price volatility, when you can hold a “double-duty” bond portfolio that prefunds your withdrawals with certainty, without losing any return or diversification benefit?”

CREATING THE FUNCTIONAL PORTFOLIO

Having created our functional asset groupings, we can now use them to create a portfolio that meets the specific goals of each client. What return do they require? How much liquidity do they desire? What unique constraints might they want to introduce? Our approach provides the flexibility to alter the mix within each of the functional groups, and then to allocate assets across these groups to meet our client’s goals.

We allocated 40% to “safety assets” that include Cash Match and Inflation Protection, and 60% to “opportunity assets” that include Growth and Illiquid Alpha. We can see how our asset allocation would appear if it had been presented in terms of the traditional asset groupings. With the traditional approach, we would have no idea how well our strategy addressed the key risks and opportunities that we hope to manage.

EXAMINING THE RESULTS

Given our mix of investment segments within each functional group, and our allocations to the groups, we developed this forecast of the returns and volatility risk for each functional group and for the overall portfolio.

In terms of asset efficiency, we see a strong relationship between risk and return across the functional groups. Taken together, almost 99% of the portfolio’s return is explained by the volatility risk of its functional groups. We also see that the portfolio enjoys diversification benefits, with its risk-adjusted return lying above the trend line of its assets.

The key to portfolio efficiency is the joint attribution of both return and risk, so that we can compare the contributions of each asset group to the total return and risk of the portfolio. An efficient portfolio is achieved by equalizing these contributions, so that every asset “pulls its own weight” (subject to any investor constraints.) As expected, our growth segment is the least-efficient of all, since it is heavily weighted to global public equities, which are highly-correlated to each other and to the total portfolio. The result of these high portfolio correlations is a higher retention of these growth assets’ individual volatility. Fortunately, this inefficiency is offset by the “super-efficiency” of the Cash Match segment, which contributes disproportionately to return, relative to its lower risk contribution. This is a double benefit, as the cash match segment provides essential liquidity, while also contributing significantly to overall portfolio efficiency. The inflation protection and illiquid alpha segments are perfectly efficient, lying directly on the “Efficiency Line.”

Our essential message is this:

Using functional groupings lets us manage risks that threaten client success, while earning returns that compensate us for regular volatility risk. This approach links our strategy to client goals, as we focus on “risk from the client’s perspective” in this initial phase of the investment process.

LOOKING AHEAD

Our fifth-and-final segment of this asset allocation saga will be a radical departure from the typical approach to this topic.

So far, we have followed the well-worn path of assuming that we have forecasting skill regarding future returns and risk in the investment markets. These forecasts are the result of substantial efforts by knowledgeable and experienced investment practitioners, market strategists and economists - but they are still only opinions, and opinions can be wrong.

Our next paper answers this question:

“How do we proceed if we don’t have forecasting skill, or if we lack confidence in our forecasting abilities?”

We will introduce a practical and intuitive approach to asset allocation that uses broad, long-term relationships and empirical evidence from nearly 100 years of market behavior. This common-sense, definitional method is as robust and mathematically-sound as the forecast-driven, speculative approach that most firms adopt. 

.

.

.

Written in partnership with Stephen Campisi