The most recent step in the evolution of portfolio construction practices has been a shift from an asset allocation–centered process to a more comprehensive risk allocation–based process. Cambridge Associates’ Risk Allocation Framework considers multiple dimensions of risk and return trade-offs when building portfolios and evaluates the consequences of risk allocation decisions during normal and stressed markets.
Investors have traditionally constructed portfolios by considering how much to allocate to different asset classes. Since the 1970s-style balanced fund allocated to domestic stocks and bonds, asset allocation has generally evolved as follows:
1. Replacement of balanced portfolios by separate asset classes, including asset classes such as foreign market equities.
2. Birth of the “style box” as investors hired separate managers to focus on large-cap, small-cap, growth, and value stocks.
3. Broader adoption of real estate and other forms of private investments (e.g., venture capital and buyout funds).
4. Addition of distressed securities, commodities, natural resources, and various kinds of hedge funds.
Yet the problem became that several of these more recently introduced “asset classes” actually have common risk factors that cross “asset class” boundaries. Examples include equity risk in distressed securities and natural resources equities, and illiquidity risk in hedge funds and commingled funds—particularly in stressed environments. Thus it became increasingly difficult to recognize, without significant analysis, just how much equity risk (for example) might be embedded in a portfolio that owned lots of assets not named “equities.”
To clarify matters, investors increasingly have constructed portfolios on the basis of the role they expected different kinds of investments to play in the portfolio (i.e., role-in-portfolio exposures), even if they still allocated investments to traditional asset classes.
“Cambridge Associates’ Risk Allocation Framework considers multiple dimensions of risk and return trade-offs when building portfolios and evaluates the consequences of risk allocation decisions during normal and stressed markets.”
The Risk Allocation Framework takes this evolution a step further by considering not only the role that different investments might play in the portfolio, but how and in what ways such investments contribute to or mitigate various forms of portfolio risk. The framework combines careful attention to risk allocation in the context of the risk sensitivities and limitations of a long-term investment portfolio (LTIP) given its role in the broader organization. Since risk exposures move over time, we monitor risk allocation and performance attribution dynamically.
There are four main components to the Risk Allocation Framework.
Enterprise Review: Simplify and clarify the process for identifying the degree of dependence on and integration of the LTIP in the total enterprise and identifying any constraints on portfolio construction. This facilitates informed decision making about appropriate fundamental exposures to include in the portfolio.
Policy Setting: Set top-down objectives to clearly express investors’ most timeless and fundamental risk tolerances and objectives. Policy reflects desired role-in-portfolio exposures (e.g., diversified growth, deflation hedge, and inflation sensitive), value-added performance objectives, and common risk factors of equity beta, illiquidity tolerance, and foreign currency risk. The policy is designed to provide those implementing portfolios with the appropriate guidelines for meeting long-term objectives.
Implementation: Determine allocations to most effectively implement investment policy objectives in the current environment. Implementation includes all decisions that result in differences between the actual portfolio and the policy portfolio, including more detailed asset allocation, manager structure, and manager selection. Measure and monitor implementation decisions to make sure risks taken are consistent with conviction about their potential value added relative to policy. Given the changing nature of implementation risks and potential rewards, the framework uses a dynamic approach to understand risk characteristics of portfolio positions in the current environment and put them in a historical context.
Ongoing Performance Monitoring: Use performance measurement and attribution analysis to understand the ways in which value has been added to (or subtracted from) the portfolio. Performance measurement serves as a feedback loop into continuous improvement in portfolio management.
Skeptics might question whether the Risk Allocation Framework is an evolutionary step forward or merely just the same old fellow, dressed in a new suit. Asset allocation—informed by rigorous valuation analysis—and manager selection remain important parts of the Risk Allocation Framework, so there is some family resemblance. In fact, this is by design, as we preserved the best practices of traditional portfolio construction (often referred to as “the endowment model” because endowments were the earliest adopters).
However, the Risk Allocation Framework’s primary merits are that it enables investors first to more rigorously construct portfolios designed to realize their investment objectives and second to understand much more clearly how best to manage such a portfolio dynamically to improve the likelihood that it performs as anticipated.