CLO Portfolio Overlap
CLO portfolios, even from CLOs issued in different years, tend to have a lot of underlying loan borrowers in common. This is the result of loan repayments causing CLO managers to continually be in the market buying loans for their CLOs. Thus, portfolio differences due to CLO vintage are muted. CLO managers’ practice of allocating loan purchases across all their CLOs causes additional borrower overlap among CLOs managed by the same manager.
The average “name” or “credit” overlap among the 32 CLOs we study in this chapter is 45% of par. The average overlap among CLOs from the same vintage is 44%, and the average collateral overlap among CLOs managed by the same manager is 81%. We also look at CLO and collateral vintage. We find that different vintage CLOs have similar collateral vintage distribution. For the 2003 to 2007 CLOs that we study, 2007 collateral was within a tight range of 55% to 58%, and 2006 collateral ranged from 24% to 29%. At the same time, small loan allocations, and the necessity of filling several CLOs with collateral, keep single-name concentrations within individual CLO portfolios small. The average number of separate credits in the CLOs that we study is 250.
We first present several measurements related to collateral overlap and single-name concentration. We look at collateral overlap in CLOs between individual CLOs, between CLO managers, and between CLO vintages. We also look at collateral vintage across CLO vintage. Next, ...