New Study Affirms Value of Credit Counseling

A study released in March 2002 by the Credit Research Center at Georgetown University's McDonough School of Business documented a significant improvement in consumers' borrowing and payment behavior in the years following their receipt of credit counseling. The results provide statistical evidence that credit counseling has a positive impact on many borrowers, and will bolster the appeals for additional funding by those agencies that offer full-service counseling. This would be a positive development in the marketplace as the revenue streams for many full-service agencies are being squeezed because rates are being competed down by counseling agencies who skimp on the counseling and put clients directly on debt management plans. The continued provision of full-service counseling is currently in jeopardy.

Background:

In 2001, between 2.0 and 2.5 million Americans felt themselves under sufficient financial pressure to seek advice and other assistance from a credit counseling agency, sometimes prior to bankruptcy but mostly as an alternative to bankruptcy. Member agencies of the National Foundation for Credit Counseling (NFCC), the nation's oldest and largest network of non-profit credit counseling agencies, counseled over 880,000 new clients in 2000, more than twice the number counseled in 1990. Clients counseled by NFCC agencies receive a comprehensive budget review and written action plan. For about one-third of all NFCC-agency clients, the agency actively intervenes with creditors to arrange a Debt Management Plan (DMP) and creditor concessions in the form of reduced interest rates, fees and minimum payments. For the other two-thirds of counseled clients, the agency provides a budget review, education, advice, possibly referrals to social-service agencies or other institutions to solve specific problems, and recommendations for specific changes in the client's behavior.

Both the agencies and the clients' creditors closely track the progress of individuals who agree to a DMP. However, there have been no empirical studies of the impact of counseling on clients who do not end up in DMPs, the large majority of all individuals served by NFCC-member agencies. Historically, the revenues received by agencies from creditor contributions (typically calculated as a percentage of DMP payments) were sufficient to subsidize the counseling provided to the larger group. However, the competitive pressures in today's market require agencies to demonstrate the effectiveness of non-DMP counseling. Without such evidence, creditors will not continue to pay to subsidize the counseling services delivered to clients who are not on DMPs.

A further motivation for evaluating the impact of credit counseling is the growing reliance on mandatory financial education as a policy tool for remedying perceived problems of credit markets. The US Department of Housing and Urban Development already requires homeownership counseling in a variety of its affordable-lending programs. Recent federal and state legislation regarding predatory lending problems as well as pending bankruptcy reform bills also contain counseling provisions.

To address the need for empirical evidence, the NFCC encouraged an objective, academic research study of the impact of credit counseling (in the absence of a debt management plan) on the subsequent credit usage and payment behavior of clients. The resulting report, co-authored by economists Michael E. Staten and Gregory Elliehausen (Georgetown University) and E. Christopher Lundquist (Lundquist Consulting), examined the impact of one-on-one counseling delivered by NFCC-member agencies to approximately 14,000 non-DMP clients during a five-month period in 1997.

Study Design:

Credit bureau data supplied by Trans Union, LLC provided an objective measure of credit performance for clients over the three-year period (June, 1997 to June, 2000) following the initial counseling session. These data were compared to similar data for a large randomly selected comparison group of individuals with risk profiles and geographic residences similar to the client group in 1997 but who were not identified by the five participating agencies as having been counseled. The study uses regression analysis to isolate the impact of the credit counseling experience, holding constant other observable factors that may also influence subsequent borrowing and payment behavior.

The statistical model also corrected for any selection bias that may arise because receipt of counseling is the result of choice rather than random selection. That is, borrowers experiencing financial difficulties may choose counseling because they are more motivated to take corrective steps to improve their credit profile than otherwise similar borrowers in the comparison group. The analysis took this factor into account in isolating the impact of counseling itself.

The study examined ten different measures of borrower behavior subsequent to counseling, including (1) summary measures of creditworthiness such as credit bureau scores, (2) specific dimensions of credit usage such as number of accounts with balances and total amount of debt, and (3) payment performance.

Results:

The empirical analysis found that borrowers who received full-service financial counseling (i.e., counseling that followed the NFCC methodology) improved their credit profile over the subsequent three years, relative to observationally similar borrowers who did not receive counseling. Highlights of the results include the following.

• Holding other factors constant, financial counseling had a significant and positive impact on summary measures of borrower creditworthiness—Trans Union's Empirica and Empirica bank card risk scores—over time. The effect was greatest for clients who had lower Empirica scores at the outset. Over half of counseled borrowers experienced increases in their Empirica scores relative to borrowers in the comparison group over the three-year period. Borrowers with initial Empirica scores at the lowest 10th percentile who were counseled experienced an average net increase of 36.3 points in their Empirica scores over the three year period, relative to borrowers with the same initial Empirica score in the comparison group. This 36-point increase in the Empirica score implies a 38% reduction in the predicted frequency of a chargeoff/repossession/bankruptcy.

• Smaller benefits were found for clients with higher initial Empirica scores. The inverse relationship between initial Empirica score and counseling's ability to lift that score over time appears to stem from at least two factors. First, it seems reasonable to expect that the effectiveness of counseling would depend upon the borrower's initial ability to handle debt. Borrowers with lower ability are likely to obtain greater benefits from the counseling experience than borrowers with higher ability. Borrower's with higher Empirica scores at the time of counseling have apparently managed their debts better than borrowers with lower initial scores, and would be less likely to benefit. Second, the decision to seek counseling reveals important information about a borrower's likely future credit performance that may not be captured in the Empirica score available to creditors at the time of counseling. Clients who seek counseling reveal the existence of an underlying financial problem. Clients with higher scores, by definition, have experienced fewer financial problems in the past. Their scores have farther to drop as a consequence of a new crisis. Consequently, even with counseling, clients with higher initial scores spend the next three years digging out of the hole into which their scores have dropped, but at the end of that period they still lag significantly behind their counterparts in the comparison group.

• Across a range of specific credit usage characteristics—number of accounts with positive balances, total debt, consumer debt, revolving debt, number of bank cards with positive balances, and bank card percentage utilization—counseled clients generally experienced improvement relative to the comparison group. For many credit attributes, there is evidence of improvement for counseled clients even when their Empirica scores did not improve relative to the comparison group. Relative to the control group, all clients except those with the very highest initial Empirica scores reduced the number of accounts they owned with positive balances (total and revolving), total dollar amount of debt, and total dollar amount of non-mortgage debt. Over three-fourths of clients with bank card debt reduced their utilization of bank card credit limits.

  The large majority of the estimated changes are consistent with the recommendations of counselors to reduce debt, refinance more expensive debt, and close unnecessary accounts. These changes signal borrowers who are actively making changes to improve their financial circumstances. That they occur across a broader range of the client distribution than does the improvement in Empirica score reinforces the idea that counseling triggers demonstrable (and positive) behavioral changes.

• Three years after counseling, delinquency experience—as measured by the reduction in 30+ and 60+ day delinquencies in the last 12 months—is substantially better for counseled clients, relative to the comparison group. Most clients experienced fewer delinquencies in the last 12 months of the observation period relative to the comparison group. For example, a borrower in the 25th percentile with respect to initial Empirica score had 9.0 fewer delinquencies of 30+ days in the last 12 months of the observation period, relative to comparison group members in the same percentile. As was the case for most other performance measures, the positive impact of counseling on delinquency experience diminishes for clients with higher initial Empirica scores.

Conclusions:

These results provide strong evidence that credit counseling that utilizes the NFCC agency methodology affects credit use and payment behavior in a positive way. Most counseled borrowers improved their risk scores relative to other borrowers with similar initial risk scores in the three-year evaluation period following their counseling. And, the large majority of counseled borrowers had significantly fewer accounts, lower debt, and fewer delinquencies relative to other borrowers, behavior that is consistent with the advice provided in credit counseling.