Commentary: Future of portfolio construction lies in adaptive, not static, risk allocation

Risk-parity strategies are designed to fare well in different types of markets using a multiasset approach. In the years leading up to the 2008 financial crisis, many practitioners embraced the key tenets of risk parity and used them to develop new strategies. Ideas such as using leverage to harness the higher risk-adjusted returns of fixed-income markets, investing according to risk instead of capital and the diversification benefits of multiple risk premiums beyond equity have become common for sophisticated investors.

However, one area of risk allocation that has not been explored as commonly is the idea of adaptive rather than static risk allocation.

At a high level, risk parity seeks to balance the underlying risks of stocks and bonds and achieve a higher Sharpe ratio (i.e., it is a more efficient portfolio, delivering greater return per unit of risk).

To reframe the problem, if an investor's goal is to maximize Sharpe ratio, is a static risk allocation approach always the answer? Not necessarily. Looking at rolling three-year returns, one gets a different answer on how much equity risk is optimal given market performance.

Exhibit 1 shows the optimal equity risk weight in a portfolio on a monthly basis for the period January 1973 through September 2020. In this exhibit, we define optimal as delivering the highest possible Sharpe ratio.

Importantly, the average across this time period is a portfolio with 54% risk from stocks and 46% coming from bonds. Many would see this as validation for a static risk-parity approach — over time, stocks and bonds appear to reach near equilibrium, even if the actual historical allocation is markedly different at any given point. However, if we can say these sorts of swings can and should be identified ex-ante, then an investor could strive to achieve a better Sharpe ratio by adapting their risk allocation approach to different market environments.

Adaptive risk allocation is predicated on the idea that there are certain environments in which an investor should take more risk in a particular asset class or less risk in others and that these periods can be identified ex-ante. In this way, a portfolio manager can aim to protect the portfolio when required or take more risk when appropriate by adapting based on market conditions. This observation may seem simple, but it's not widely implemented.

For many, the rationale for not using an adaptive approach is that it introduces human fallibility (via market timing) into what is otherwise a mechanical process. While those concerns are valid, risk allocation strategies can be enhanced by using a relatively simple rules-based strategy to establish an optimal equity allocation.

An interesting development we've noticed in the first quarter of this year is a shift in equity and bond correlations. Inflation volatility has caused equities and bonds to be more positively correlated than in the past. While this trend may not persist, it's an important consideration for investors. A portfolio is usually composed of more bonds when in a defensive market state or heavily weighted toward equities in a highly bullish market state. A traditional risk-parity approach may only assess a portfolio's stock vs. bond holdings on a surface level, but this could put investors at risk if both assets move in the same direction — especially down.

An adaptive risk approach looks under the hood of individual stocks and bonds and allows for more diversification through varying durations. The traditional 60/40 portfolio is outdated and may present more risks than rewards in today's volatile market environment.

Using a standard efficient frontier analysis, we can observe that the (static) risk-parity portfolio is a more efficient outcome than any of the traditional capital allocation portfolios. But we can also see that this adaptive portfolio has significantly higher returns with a minimal increase of risk over the period from Dec. 31, 1969, through Oct. 31, 2020.

But the advantages are not just limited to more attractive traditional risk/return trade-offs as shown in our efficient frontier. A further benefit of the adaptive approach is more attractive drawdown characteristics. We would have seen a 31% drop for the 60% equity/40% fixed income capitally allocated portfolio, a 23% drop for the static risk-parity portfolio and 19% drop using the adaptive risk approach during drawdown periods between 1969 and 2020.

Risk parity has been arguably the most disruptive innovation in the history of the asset allocation industry. The idea to invest with an attention to risk as opposed to capital changed the landscape for diversified investors and has led to more advanced risk management methods for asset allocation portfolios, both traditional and alternative. We believe that by using an adaptive approach, investors can reap the benefits of risk-parity science, protect against the downside of market crises and have opportunities to participate in the upside during calm environments. By taking an adaptive approach, risk parity can continue to be a successful, relevant investment strategy for the uncertainties that the future holds.

Joshua Kutin is head of asset allocation, North America and senior portfolio manager at Columbia Threadneedle Investments. He is based in Boston. This content represents the views of the author. It was submitted and edited under P&I guidelines but is not a product of P&I's editorial team.