Uncategorized
A rebound in manager performance through May
admin | June 5, 2020
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
More than half of CLO managers outperformed the broad leveraged loan market in the three months ending in May. The average CLO leveraged loan portfolio lost 9.59% and, after adjusting for broad market exposure, or beta, still underperformed the market by 0.04%. Managers with strong track records, better loans and good liquidity at the end of February delivered more excess return than their peers through the volatile months of March, April and May. Palmer Square came out on top with excess returns over the index of 1.38%.
Loan returns plunged in March but rose sharply in April and continued to rise in May (Exhibit 1). After accounting for the various reporting dates of managers, the S&P/LSTA Index lost 9.37% between March and May. These managers held portfolios with an average beta of 1.02, which meant the average manager should have lost 9.56%. With the actual average performance at -9.59%, the average manager underperformed the index by 0.4%.
Exhibit 1: Loan returns plummeted in March but bounced back in April and May
Source: Bloomberg, Amherst Pierpont Securities
Secondary market volatility in leveraged loans—as measured by the standard deviation of daily returns on the S&P/LSTA index—spiked between March and May (Exhibit 2). At the peak, the rolling 30-day volatility reached more than 10 times higher than the average volatility before March. Although the market volatility has retreated since the peak, it remained elevated in May.
Exhibit 2: 30-day volatility spiked between March and May
Source: Bloomberg, Amherst Pierpont Securities
More than half of the 66 managers with at least five active deals tracked by APS led the index, a significant uptick from the figures reported by APS last month where only 23% of the managers led (Exhibit 3). The other half of managers lagged the index significantly, however. More than 15% of the managers trailed the index by at least 1% while only 4% of managers came ahead of the index by more than 1%.
Exhibit 3: Half of the managers outperformed the index between March and May
Note: data shows excess return only for active deals. Source: Amherst Pierpont Securities.
The credit quality and tail risk of managers portfolios coming into the spring had clear impact on recent excess returns (Exhibit 4). Managers with highly priced loans and low exposure to ‘CCC’ loans delivered more excess return on average. Managers who had a history of consistently producing excess returns—measured by their information ratio—also tended to do better in recent months. Finally, liquidity as gauged by the average portfolio bid depth, helped managers adapt to the volatile market conditions.
Exhibit 4: Correlation of portfolio or manager features with recent excess return
Note: data shows the correlation of manager or loan portfolio attribute with managers’ excess return or alpha from March through May only on active deals. Portfolio attributes measured as percentiles. Source: Amherst Pierpont Securities.
The following managers delivered positive alphas despite the tough environment between March and May (Exhibit 5). Palmer Square ranked in the top five according to APS reports since March and was a clear winner this time. The others in the Top 5 include York, Triumph, Goldentree and Napier Park..
Exhibit 5: Alpha leaders in CLO portfolio performance March-May 2020
Note: Performance for managers with five or more deals tracked by APS. Performance attribution starts with calculated total return on the leveraged loan portfolio held in each CLO for the 3-month reporting period ending on the indicated date. CLOs, even with a single manager platform, may vary in reporting period. The analysis matches performance in each period to performance over the identical period in the S&P/LSTA Leveraged Loan Index. Where a deal has at least 18 months of performance history since pricing and no apparent errors in cash flow data, the analysis calculates a deal beta. The deal beta is multiplied by the index return to predict deal return attributable to broad market performance. Where no beta can be calculated, the analysis uses the average beta across each manager’s active deals weighted by the average deal principal balance over time. Any difference between performance attributable to beta and actual performance is attributed to manager alpha. Source: Amherst Pierpont Securities.