CPR Report: SBA 7(a) Prepays Decrease in March
March 21, 2014
By Bob Judge
Editor, CPR Report
In March, prepays fell below 7% for the first time this year, having decreased by 17% from February.
Defaults just barely stayed below 2%, making it seven months in a row of sub-2% default prepay speeds. This continues the longest such stretch in our database, which goes back to 1999.
As for the detail, overall prepayments fell 17% to 6.70% from 8.10% in February. In comparing prepayment speeds for the first three months of 2014 to the same period in 2013, we see that this year is running 9.93% above last year, 7.63% versus 6.94%.
As for the largest sector of the market, 20+ years to maturity, prepayment speeds fell by 20% to 6.37% from 7.91% in February.
Turning to the CPR breakdown, the default CPR rose by 25% to 1.98% while the voluntary prepayment CPR fell by 27% to 4.73%. For the record, this is the first sub-5% voluntary prepayment reading since last April.
Preliminary data for next month suggests that prepayments will be largely unchanged, staying below 7% for the second time this year.
Turning to the default/voluntary prepayment breakdown, the Voluntary Prepay CPR (green line) fell to 4.73% from 6.51%, a 27% decrease.
While the VCPR remained above 6%, the Default CPR (red line) increased by 25% to 1.98% from 1.59% the previous month.
Prepayment speeds fell in five out of six maturity categories. Decreases were seen, by order of magnitude, in the <8 year sector (-64% to CPR 6.32%), 16-20 (-35% to CPR 4.54%), 20+ (-20% to 6.37%), 8-10 (- 7% to 8.15%), and 10-13 (-5% to CPR 8.10%).
The lone increase was seen in the 13-16 sector, which rose by 32% to 5.52% from 4.18%.
After two months of +8% prepayment speeds, we have settled back below 7% for at least the next two months, pushing us back toward the speeds we saw last year.
While we continue to expect higher prepays in 2014 versus 2013, perhaps it will be only a modest, single-digit percentage increase year-over-year.
For further information on the terminology and concepts used in this article, please refer to the “Glossary and Definitions” at the end of the report.