Positioning a Multi-Manager Portfolio for Long-Term Success

DB plan sponsors need to ensure their liability-driven investing strategies are diversified and well-positioned for years to come.

Defined benefit (DB) plans’ funded statuses have reached levels not seen since the end of 2007. The challenge for plan sponsors now is how to lock in those gains.

Plans have already allocated significant assets to their liability-driven investing (LDI) portfolios, with the average plan allocating 50% of its total portfolio to LDI, according to the Milliman 2021 Corporate Pension Funding Study. We argue that, from here, it is less about the size of the LDI allocation and more about the overall structure of the multi-manager LDI portfolio.

By continuing to only increase allocations to existing managers, plans are exacerbating what was already high manager concentration risk and a portfolio that could only do well in a credit bull market with below normal downgrades. We believe LDI portfolios are under-diversified and ill-positioned to thrive across a range of potential environments. It is more important than ever for plan sponsors to properly diversify the LDI strategy, similarly to how they painstakingly diversify their return-seeking asset portfolios.

In our work with clients, we have identified five potential risks many plans face in their current LDI portfolios.

Risk 1: Substantial overlap in sector allocation and credit exposure among managers

Fundamental analysis of managers’ holdings helps identify which active bets managers were taking versus the benchmark and the degree to which these changed over time. From 2002 to 2021, many managers gravitated to the same areas of the market in order to generate alpha. For example, most managers were consistently underweight Treasurys in favor of off-benchmark sectors, e.g., high-yield, non-agency mortgage-backed securities (MBS). When constructing a multi-manager LDI portfolio, it is important to understand these biases and avoid concentrated risks.

Active positioning largest LDI managers 2002-2021

High degree of correlation in active weights vs Barcap Long Gov/Credit index

Source: Source: eVestment. Top 14 LDI Mgrs. by assets plus Schroder Long Duration Value. Based on quarterly holdings data from 2002 Q1 through 2021 Q2 where available for each manager. Past performance is no guarantee of future performance.

Risk 2: Managers tend to be “risk on” at all times

Maintaining a persistent underweight allocation to Treasuries and an overweight allocation to BBB-rated- and A-rated bonds allowed managers to generate alpha through exposure to spread or “beta” relative to a AA discount rate. However, managers should be able to pivot in risk-off environments. Based on the data, managers appear unwilling or unable to do this. For example, coming into two risk-off periods (the fourth quarter of 2018 and the first quarter of 2020) only one of 15 long-duration bond managers was meaningfully overweight Treasurys and underweight BBB-rated bonds.

Risk 3: Portfolios are not sufficiently alpha-seeking

LDI portfolios face liability drag of 1% to 1.5% per year due to the uneconomic treatment of downgrades in the discount rate. Plans must overcome this drag by earning sufficient alpha in LDI strategies. Aiming to modestly exceed a long-duration index will ensure persistent underperformance of the liability. Of the 15 managers observed, only four generated annualized alpha of 1% or more over the past 10 years.

Risk 4: Manager size limits access to the full credit universe

The Bloomberg Barclays Long U.S. Corporate Index is dominated by smaller debt issuers; 73% of borrowers have less than $5 billion of debt outstanding. Mega LDI managers have problems accessing the portion of the credit universe beyond the biggest benchmark issuers. A simple hypothetical makes this clear: Imagine a $50 billion manager takes a $500 million (1%) position in an issuer with $5 billion in total debt outstanding. The manager would have to purchase 10% of that issuer’s outstanding debt. While that’s perhaps appropriate for an event-driven hedge fund, it’s not appropriate for an LDI portfolio. The ability to access the full corporate universe may afford smaller managers opportunities to diversify downgrade risk and pursue alpha through opportunistic credit selection.

Number of issuer tickers by outstanding debt

Barclays US Long Corporate Index

Source: Bloomberg. Data as of June 30, 2021

Risk 5: Regime dependency in manager performance and correlations

Plan sponsors look for managers with repeatable alpha and low correlations to each other. Over longer periods, the performance of an LDI portfolio may reflect this. However, the correlations among managers and their performances fluctuate wildly under different market environments, such as in times of scarce liquidity or widening spreads, making it challenging to build an “optimal” portfolio without a more powerful framework that aims to secure true diversification.

To wit, in times of market stress many managers perform similarly to each other, due to their issuer concentration and underweighting of Treasurys. The diversification thought to be in the LDI portfolio might prove to not exist. It would be a shame to lose these gains in funded status if economic events were to simultaneously drive equities down, increase downgrades and reveal the under-diversification present in most LDI portfolios.

For plans to achieve the best outcomes, they should ensure their group of managers takes active decisions in all environments, including “risk-off,” and in the aggregate are nimble enough to navigate the full range of the credit universe beyond the large benchmark issuers. The task for plans now, given the increase in funded status, is to review their LDI programs and ensure they are truly diversified among alpha drivers and seeking sufficient returns to offset the liability drag.

 

Ryan Miller, CFA [Chartered Financial Analyst], ASA [Associate of the Society of Actuaries], is a solutions manager at Schroders.

This feature is to provide general information only, does not constitute legal or tax advice and cannot be used or substituted for legal or tax advice. Any opinions of the author do not necessarily reflect the stance of Institutional Shareholder Services Inc. (ISS) or its affiliates.

«