Capital Ideas^™ 
Investment insights from Capital Group

Diversification is often referred to as a “free lunch” for investors. But it requires a nuanced approach in a defined benefit pension plan. Sponsors seeking to mitigate risk or boost returns by diversifying their liability-driven investing (LDI) program should perform an in-depth assessment of any potential new asset classes. In this article, we detail five key considerations, and propose an alternative approach to diversification focused on the plan’s current asset managers.

1. Diversification should be viewed at the total portfolio level, not just relative to the LDI program

Defined benefit pension plans generally include a hedging portfolio that utilizes LDI, and a return-generating portfolio. Nuances exist, but overall, this framework has been effective in helping sponsors evolve their plans over time. Plans typically cement funded status improvements by increasing the allocation to the LDI portfolio while trimming the return-seeking portfolio.

When plan sponsors consider a new asset class to diversify the pension risk management program, it is important that they do so in the context of the entire plan. Focusing too heavily on the hedging portfolio in isolation can significantly overstate a diversifying asset’s positive impact and miss important relationships with the return-generating portfolio. For example, adding a diversifying source of credit risk to the LDI portfolio may look attractive given that component’s focus on valuing the plan’s liabilities using a discount rate based on high-quality corporate bonds. But at the total portfolio level, credit risk inherent in equities and other credit-spread-sensitive assets may lead to a less positive, or even negative, impact – particularly in periods of funded status stress.

2. The diversifier should move the needle on risk and/or returns by enough to justify incremental fees and complexity

Plan sponsors will often add a diversifying asset class to reduce total portfolio risk, enhance returns, or both. Risk can be defined in several ways, including the probability of the plan’s funded ratio falling below 100%. Whatever the reason for seeking to reduce risk, the diversifying asset should target the specific risk metrics that the sponsor is focused on and, importantly, the impact should be significant. If a plan is replacing 5% of an investment-grade (BBB/Baa and above) credit portfolio with a 5% allocation to an alternative source of credit risk, the sponsor should ensure both qualitatively and quantitatively that the new allocation can materially reduce risk at the total portfolio level. 

On the return side, “material” ultimately needs to be defined by the plan sponsor. Staying with the 5% reallocation example above, suppose the diversifier is expected to produce 150 basis points (bps) of annualized excess return over a market cycle. This equates to just 7.5 bps (5% x 150 bps) of expected excess return at the total portfolio level.

It is also important to factor investment management fees into the analysis because they can have a material effect on outcomes. Particularly for private markets and more esoteric diversifiers, investment management fees can be several times higher than those associated with more liquid and traditional LDI implementation approaches. 

3. Don’t lose focus on strategic flexibility or liquidity

Under some circumstances, diversification can impair a plan’s strategic flexibility and lead to liquidity challenges. This risk was starkly illustrated in 2022, which, arguably, delivered a “1-in-20-year” de-risking opportunity for corporate pension plans. Inflationary pressures coupled with Federal Reserve policy tightening drove nominal interest rates sharply higher across the yield curve. This led to higher liability discount rates that produced lower liability values and sizeable increases in funded status for most plans despite double-digit declines across major stock and bond indexes.

Given this funded status surge, did plans significantly ramp up de-risking? Not exactly. Owing to the high correlation between their liquid assets (equities and fixed income, broadly speaking), many plans experienced massive drawdowns in their available liquid capital. Some plans’ asset allocations became meaningfully detached from their strategic targets. This effect was magnified for sponsors who continued to hold private assets that are not marked to market. In some cases, sponsors with interest rate overlays had to meet capital calls that exacerbated portfolio imbalances and further weakened liquidity.

This episode underscores the importance of protecting strategic flexibility and liquidity when considering LDI diversification. Plan sponsors should ensure:
 

• Liquidity to make benefit payments. Even with asset portfolio drawdowns coupled with lower market-based liabilities, benefit payments don’t change as a result – and they can increase as a percentage of assets.
• The ability to rebalance. Strategic asset allocation rebalancing is important, particularly in periods of market volatility. Maintaining your liability hedging objectives and a reasonable range around your strategic allocation to return-generating assets is critical to not compromise the integrity of specific plan objectives.
• Flexibility to execute the glide path. Higher funded status is great, but the ability to source and move capital is imperative to continue de-risking as funded status increases.
• The capacity to execute any risk transfer activity. Lump sum payouts and annuity purchases require liquidity, both for the transaction and to ensure that the plan’s post-transaction asset allocation is not compromised. This has become even more important as insurance companies increasingly want cash rather than in-kind transfers.

4. Don’t count on expected outcomes, correlations or even 1-standard-deviation funded status volatility

Expected outcomes can’t be ignored. But they often reflect one economic path out of thousands, and they rely heavily on normally distributed inputs such as individual asset class returns, risk and correlations. Particularly for plans that want to protect a strong funding position, the normal distribution of outcomes is less important than the potential downside outcomes that can trigger a multitude of negative consequences, including cash contributions from the plan sponsor and costly variable rate premium payments to the Pension Benefit Guaranty Corporation. 

Correlation analysis can be an effective starting point to evaluate whether a particular diversifier has a relationship to the underlying goals of the hedging portfolio or the total portfolio and/or a more explicit linkage to a liability discount rate. That said, while correlations are simple and easily understood, they have several limitations, including poorly capturing magnitude, imprecision, instability due to potential large and rapid swings that can occur under regime shifts or broader economic stress, and other statistical pitfalls such as nonlinear relationships and outliers.

Funded status volatility is easy to understand as we work through any asset/liability analysis. The typical 1 standard deviation covers about 68% of the expected (normally distributed) range of outcomes. As mentioned earlier, with downside protection arguably more important than upside participation, caution should be used in any analysis focused on normally distributed outcomes.

We strongly suggest any analysis of a diversifier should more heavily weight potential downside outcomes. This analysis should include at least historical and deterministic stress testing, risk factor decomposition, value-at-risk analysis focusing on potential worst-case scenarios and even stress testing the underlying capital market assumptions. While we can’t predict the exact parameters for the next period of economic stress, we can better prepare ourselves for a wide variety of scenarios that could result with the addition of a particular diversifier.

5. Be cautious in markets that are not as observable or marked to market

Diversification conversations often lead to considering private market allocations, which can exacerbate the liquidity challenges mentioned earlier. Further, it is not uncommon for private allocations to look “diversifying” due to factors such as portfolio values being driven by a significantly lagged appraisal process, and in some cases, less observable markets where mark-to-market valuation is either difficult or potentially misleading. Ironically, during periods of stress when limited secondary markets may dry up further and the realizable value of private assets comes into question, their values may appear to shine on paper from an unrealized or modeling standpoint.

There are methods to adjust returns data to reflect the impacts of lags and infrequent price discovery, but don’t expect them to be perfect. Correlations with other asset classes may remain understated, and diversification may remain overstated. Understanding the assumptions that are going into the analysis and focusing on the economic drivers of the diversifying market are critical in any analysis.

An alternative approach to diversification: Focus on the managers 

Embracing the five considerations above can help plan sponsors ensure that any diversifying asset classes add material value to the pension risk program and are effective across a wide range of portfolio outcomes. However, here we propose a different perspective on diversification that focuses on the plan’s asset managers. 

Reevaluate your current manager lineup

Particularly for larger hedging programs that utilize multiple credit managers, a lot of time and energy are focused on manager diversification. There isn’t a one-size-fits-all framework for multi-manager evaluation, but reviewing a few key questions could help improve what a plan sponsor already has.
 

• Is my manager focused primarily on security selection? Managers who drive outcomes primarily through security selection are more likely to provide an idiosyncratic (or uncorrelated) return stream relative to other managers. Additionally, they will likely have excess return outcomes that are uncorrelated to other macro exposures in the broader portfolio.
• Is my manager taking outsized macro positions, such as systematically overweighting credit? Credit risk has typically been rewarded over the long term. But the times when credit risk in a hedging portfolio has not worked were often the same times that many of the return-generating asset classes underperformed. Most plan sponsors intentionally have significant credit risk for both hedging and return-generation purposes in their portfolio, but having systematic credit risk with their active managers can exacerbate drawdowns in periods of economic stress.
• Am I taking a total portfolio perspective when evaluating my credit managers? We often rely heavily on metrics such as tracking error and excess return correlations to build a multiple-manager credit portfolio. These commonly utilized metrics can be valuable, but evaluating excess return relative to equity market drawdowns, credit spread widening and funded status drawdown can provide a more holistic, and likely more complete, manager evaluation. The best-case outcome for a pension risk management program is that its manager performs well when the rest of the portfolio is stressed – arguably the time plan sponsors need excess returns the most. 

Consider adding diversification by giving your current managers more flexibility, without changing the strategic benchmark

Another way to incorporate diversification into a hedging program is by allowing your current managers more degrees of freedom in their investment guidelines. This shifts the diversifying conversation from a strategic allocation decision to one that adds more flexibility into the current LDI program. Expanding the opportunity set, but still holding the manager accountable to the strategic benchmark (e.g., long credit), puts the manager in a position to own the decisions regarding the timing and sizing of the diversifying allocation relative to the strategic benchmark. The decision about a diversifying allocation relative to the broader opportunity set allows the manager to incorporate exposures that not only are diversifying, but also take into account the relative value and opportunity cost of the broader opportunity set.

Conclusion

Investors are drawn to diversifiers for enhanced returns, reduced risk, or both. If they believe they can achieve improved risk-adjusted returns, they should explore diversifiers further. This involves full consideration of whether any potential incremental improvements justify any added costs or fees, complexity, potential worsening of worst-case scenarios, loss of liquidity, slippage of the high-quality corporate bond discount rate hedge or other factors. Where diversifiers are incorporated, consider whether they are best governed as stand-alone mandates, or as out-of-benchmark allowances in portfolios fundamentally measured against more traditional liability-hedging benchmarks such as long credit.