Unmasking Performance Problems in Otherwise Strong Loan Portfolios May Involve `Listening to Static'

OVERLAND PARK, Kan. – A well-managed and mature loan portfolio can thwart loss rates for credit unions, but some CUs probably aren’t aware of potentially hidden concerns that may spill over into other areas. A strong portfolio can mask effects on credit performance of changes in underwriting standards and in the targeting of marketing efforts, said Brad Miller, senior portfolio strategist with CNBS. It can also provide a false sense of security for rapidly growing loan portfolios as the loss rates of newly issued loans are typically a fraction of the loss rates of more seasoned loans. The remedy might be static pool analysis, Miller said. A static pool is a group of loans issued at approximately the same time, typically during the same month or quarter. Time-series data is a set of data elements that is collected for each period being analyzed. For example, historical monthly balances and interest payments would be used to determine both interest income and the rate earned on the account, Miller explained. “My perception is that static pool analysis is infrequently used,” Miller said. “The ability to capture the necessary data out of lending systems probably wasn’t as efficient 10 years ago as it is today. Only in recent years has changing technology made it practical for credit unions to perform static pool analysis.” The need to understand how loans perform has increased in recent years as the lending market has become increasingly competitive, Miller said. Now that credit unions have the access to data to perform static pool analysis, he anticipates its use may gain speed. In order to perform the analysis on closed-end loans, Miller said credit unions need the following elements: beginning balance; average balance; scheduled principal payment; actual principal payment; delinquency status; charge off; interest payment; and fees. Financial performance, credit risk and prepayment behavior are also factored in. The analysis has several different applications for analyzing loan performance and can shed light on how lending strategies can be changed to improve performance, Miller said, including providing an early warning for loan strategies that are performing poorly or that expose the credit union to excessive credit risk. For example, teaser rates are often used on home equity lines of credit, credit card loans and variable rate mortgages to try to increase volumes. “The credit union will utilize these rates knowing that the loans will not be profitable during the time that the teaser rate is in effect but will become profitable after the `go-to’ rate goes into effect,” Miller explained. Where static pool analysis comes into play is providing the ability to look into month-to-month profitability and alerting the credit union to the extent of any loss during the teaser period and how much time elapses after it expires before the loan becomes profitable, Miller said. “Armed with the knowledge of how teaser rates affect profitability over time, a credit union can adjust either the level or life span of the teaser rate and/or the level of the `go to’ rate to modify the performance of future loans,” Miller said, as well as make informed decisions about its ability to match teaser rates offered by competing institutions. Static pool analysis can also be used to gather information about how various types of loans prepay, Miller said, including the measurement and management of liquidity, interest rate and option risks. Prepayments are determined by comparing the amount of principal received that exceeds scheduled amortization to the principal balance. By obtaining information on how various loan types prepay through various interest rate and economic environments, the credit union is able to make estimates of how loans will prepay in similar future environments, he added. Credit unions wanting to take a critical look at credit performance evaluation especially if their loan portfolios have recently seen changes in either underwriting standards or targeting of marketing efforts, can use static pool analysis to uncover gaps. “The delinquencies of recently booked loans are compared to those of loans booked at some point in the past,” Miller said. “Comparisons are made of the delinquency rates for early months that each pool of loans was on the books.” If for example, in month six the recent loans have a higher delinquency rate than the month six delinquency rate from loans booked one or two years ago, it could indicate potential looming credit problems, Miller said. “This can provide a much earlier warning than waiting until write-offs begin to accelerate before adjusting underwriting standards,” he explained. Loans can also be separated and analyzed by credit score range, direct versus indirect, loan-to-value range and by APR price range. Miller said it may be possible to issue loans at a particular price to those in a particular score band – for example, from 600 to 609. But loss rates in a lower score band “may be sufficiently higher such that they require either higher pricing, or perhaps can’t be made at all.” “Performance is often affected by the source of loans and directly sourced loans may have lower loss rates than indirectly sourced loans,” Miller said. Loan origination and marketing costs can also affect profitability and these can be taken into consideration in order to determine their effects on loan profitability through the analysis tool, Miller said. -