More construction loan monitoring ultimately decreases loan default, according to a new FDIC Center for Financial Research working paper.
The paper, “Bank Monitoring with On-Site Inspections," will be presented later this month at the Community Banking in the 21st Century Research and Policy Conference. It claims to be the first empirical study of bank monitoring within non-syndicated loans to test long-held theories that banks’ ability to monitor borrowers is a key advantage.
While it doesn't necessarily reflect the views of the FDIC, the paper includes preliminary findings from research by FDIC staff and an FDIC Visiting Scholar. It is based on 30,000 multiple-draw construction loans as well as examinations of:
- The frequency of loan monitoring by the bank via staff or third-party on-site inspections
- The contents of the inspection reports
- The borrowers’ actions over the life of the loan.
“We provide a comprehensive analysis of the determinants of construction default, the first study of its kind, as a preamble to analyzing the incremental effect of monitoring,” authors Amanda Rae Heitz, a Visiting Scholar from Tulane University, FDIC Senior Financial Economist Christopher Martin, and FDIC Senior Financial Economist Alexander Ufier wrote.
“After implementing an instrumental variable framework and controlling for relevant determinants, we find that loans with more on-site inspections are less likely to default, suggesting that, in line with theoretical predictions, monitoring ultimately improves loan outcomes and adds value to banks.”
The researchers controlled for a broad range of project, borrower, and loan characteristics the bank could choose in making the loan and setting terms, including the draw schedule, to best measure the marginal effect of more inspections on default. They found that increasing inspections from two in a 100-day period that a loan was active to three inspections in a 100-day period would lower the probability of default by 3.63 percentage points.
“As the default probability is approximately 5 percent for these loans, this is a meaningful improvement in default probability,” said the authors, who previously circulated the paper under the title, “Bank Monitoring in Construction Lending.”
The researchers also found that banks are more likely to deny draw requests when projects have negative inspection reports, supporting the idea that information collected during monitoring is important to banks’ decision-making.
In addition, researchers showed that construction loan monitoring is less frequent for loans where the bank has a prior relationship with either the borrower or the project contractor. This suggests that banks may be “transferring information across projects,” the paper’s authors wrote, adding that reduced monitoring costs might be a resulting benefit.
Managing construction loans can be complex and time-consuming, especially if financial institutions rely on manual processes and spreadsheets to track budgets, inspections, due dates, and draw information. Automating construction loan management makes processes more efficient across the board so that financial institutions can monitor both isolated and concentrated exposures in the portfolio.