The above chart describes pull-through for loans locked in CompassPoint™ during April 2019, as measured on July 31, 2019. See the notes at the end of this post for a description of our pull through statistics.
In our last post on this topic, we presented pull-through for Q1, 2019. As in that period, the April lock cohort shows a large lender peer group disadvantage vs. the smaller lender groups, especially in clean pull-through. Large lenders lag the medium and small lender groups by 7.6% and 8.2% respectively. Large lenders did improve total pull-through slightly vs. Q1 in the April lock data, but only by 2% and without moving the needle on clean pull through at all.
The small and medium cohorts remain very close to one another in both pull-through categories, though the small lenders showed modest improvements in both clean and total pull through vs. Q1.
Compass Analytics peer data also measures pull-through by business channel as shown in the following chart.
The channel data is also consistent with the Q1 data, Consumer Direct and Correspondent business models turned in significantly lower clean pull-through results vs. Retail and Wholesale. Total pull-through for Consumer Direct (CD) was lower than the other channels, though CD clean pull-through was better than Correspondent lender results. Correspondents reflected a significant clean pull-through hit (down 4.5%) vs. the Q1 report. Retail lenders were able to exercise a bit more control over locks in the April cohort – improving clean pull-through by about 1.5%. Wholesale lost some ground on both statistics, with clean pull-through down 2.5% and total pull-through down 1.6% vs. Q1.
Notes: Compass distinguishes between total pull-through (WA Pull Thru in the chart) and clean pull-through (WA Clean Pull Thru in the chart). Total pull-through is the percentage of locked loans that closed by the end of the measurement horizon, without consideration of changes to lock terms that may have impacted the lender’s hedge cost and total profitability. Clean pull-through measures the number of loans that closed under the original lock terms, with no pricing exceptions, lock extensions or re-locks.