Highlights

• Almost 91 percent of the banks examined showed no material change in underwriting practices since their last examination.  Twice as many of the banks examined tightened underwriting practices (6 percent) as loosened (3 percent).
• Examiners noted “above-average” risk in current underwriting practices for new loans in just over 10 percent of the banks examined.  Among those banks, approximately 17 percent also failed to adjust pricing for loan risk, according to the examiners.
• “Above-average” risk in loan administration of the banks recently examined was a common concern of examiners in each FDIC region.

Purpose and Design of the Report

Beginning in early 1995, the FDIC implemented a survey to report on the riskiness of current underwriting practices at FDIC-supervised, state-chartered nonmember banks.  The survey is completed at the conclusion of each bank examination that the FDIC conducts.  By systematically collecting observations from examination sites, the survey is designed to provide an “early-warning”  mechanism for identifying potential lending problems.  

The survey focuses on three topics:  material changes in underwriting standards for new loans, degree of risk in current practices, and specific aspects of the underwriting process for major loan categories.  Questions about the third topic focus on underwriting weaknesses that have caused nationwide problems in the past.  Loan types covered in the survey include:  business, construction, commercial real-estate, consumer, credit-card, home-equity, and agricultural.

Examiners evaluate underwriting practices based on both their experience and generally-accepted industry standards.  In some cases, they rate the risk associated with underwriting practices of an institution as average or above- or below-average.  In other cases, examiners classify the occurrence of specific risky practices as “frequent enough to warrant notice” or, if more prevalent, “commonly or standard procedure.”

Results from the survey can be used to monitor underwriting practices of banks within the FDIC regions.  While interesting, comparisons across regions or different time periods must be interpreted with caution.  As noted, the survey is completed at the conclusion of each bank examination.  Thus, the sample during any given period depends on examination scheduling requirements such as the financial conditions of the banks, coordination with state regulators, and the availability of staff.  As a result, the sample is not random, and the banks sampled during a reporting period and within a region are not necessarily selected for examination for the same reasons as banks in another region or during a different period.  As such, one cannot draw valid statistical inferences from cross-region or cross-time comparisons.

Results: General Underwriting Trends and Practices

Reports received from examiners during the six months ending March 31, 1997, showed no widespread problems with either current underwriting practices for new loans or in existing loan portfolios.  Examiners indicated a material change in underwriting practices since the last examination in just over 9 percent of the  banks examined.  Twice as many banks showed tighter practices (6 percent) as looser practices (3 percent). Where standards were relaxed, examiners most frequently attributed the change to loan growth goals or increased competitive forces.  These results generally mirrored results from the April-September 1996 survey in which 90 percent of the banks examined showed no material change in underwriting practices.  However, a larger proportion of the banks examined during this most recent six-month period tightened standards than previously; conversely, a smaller proportion loosened standards.

Examiners indicated “above-average” risk in underwriting practices for new loans in 10 percent of the banks examined nationwide, about the same as the April-September 1996 period. Of the banks characterized as having “above-average” risk, approximately 17 percent also “commonly” failed to adjust price for loan risk.  In existing loan portfolios, examiners noted “above-average” risk in 12 percent of the banks, approximately the same as during the previous reporting period.

Additional information from responses received during the six months ending March 31, 1997 indicated:  “Above-average” risk in loan administration surfaced in roughly 13 percent of the banks examined.  Approximately 10 percent of the banks examined booked new loans that resulted in high concentration of loans to one borrower or industry “frequently enough to warrant notice;” and an additional 6 percent of the banks examined booked such loans “commonly or as standard procedure.”

Individual Loan Types

In the survey, examiners are requested to indicate the loan types that are both a significant portion of the bank's new lending and were also reviewed during the examination.  Some 84 percent of banks examined were active business lenders, and 72 percent were actively making consumer loans.  A majority of institutions were active lenders in other loan areas:  nonresidential commercial real-estate loans (67 percent) and agricultural loans (51 percent).  Only 48 percent were active construction lenders, 29 percent were active home-equity lenders, and 21 percent were active credit-card lenders.  Few deficiencies existed in any loan type.  

Business Loans.   The strength of the borrower and the repayment source are both important elements in business lending.  For asset-based loans, the monitoring of the collateral pledged is critical.  For the six months ending March 31, 1997, almost two-thirds of the banks active in business lending made asset-based loans.  

• Approximately 15 percent of the banks making asset-based loans failed to monitor the collateral pledged either “frequently enough to warrant notice” or “commonly or as standard procedure.”
• Of the banks actively making business loans, approximately 12 percent of the banks examined loaned to borrowers who lacked documented financial strength either “frequently enough to warrant notice” or “commonly or as standard procedure.”

Construction Loans.  Five questions in the survey cover the riskiness of construction loans.  Two deal with borrowers' repayment sources, an important concern since, typically, developers receive funds to repay the loan fully only upon completion of the projects.  The remaining three questions request information concerning whether the institution (1) is making speculative loans, (2) defers interest payments during the loan term, and, (3) fails to use realistic appraisal values. Responses received from the banks examined actively making construction loans during the six months ending March 31, 1997 yielded the following:

• Just under 22 percent of the banks examined funded speculative construction projects either “frequently enough to warrant notice” or “commonly or as standard procedure.”
• Examiners noted that almost 7 percent of the banks examined funded, or deferred, interest payments during the loan term “frequently enough to warrant notice;” an additional 5 percent did so “commonly or as standard procedure.”   
• According to examiners, just under 12 percent of the banks examined made loans without consideration of repayment sources other than the project being funded either “frequently enough to warrant notice” or “commonly or as standard procedure.” Additionally, for loans on which alternative repayment sources had been required, examiners reported that 10 percent of the banks failed to verify the quality of these sources either “frequently enough to warrant notice” or “commonly or as standard procedure.”

Commercial Real-Estate Loans. In commercial real-estate lending, the income generated from the property is the primary source of repayment.  However, because the income stream often is subject to uncertainty, sound underwriting practices generally require alternative sources of repayment.  Responses received during the six months ending March 31, 1997 from the banks examined and active in commercial real-estate lending showed:

• Less than 8 percent of the banks examined failed to consider alternative repayment sources other than the project being funded “frequently enough to warrant notice.”  Only two of the over 800 banks actively making commercial real-estate loans practiced this lending “commonly or as standard procedure.”
• About 11 percent of the institutions examined made short-term commercial real-estate loans with minimal amortization and large balloon payments “frequently enough to warrant notice.”  Another 3 percent were characterized as making these loans “commonly or as standard procedure.”

Agricultural Credits.  Last year, the Congress passed legislation to phase out farm subsidies on selected crops.  As a result, examiners were asked to determine the extent to which banks' agricultural loan portfolios are tied to major crops affected by this phase out.  Available answers included:  “none”, “enough to warrant notice”, and “substantially”.  For banks active in agricultural lending:

• Examiners noted that 28 percent of the banks examined have portfolios tied to crops affected by the phase out “enough to warrant notice”; an additional 7 percent of the banks examined have portfolios tied “substantially”.  
• Examiners also noted a “moderate” increase in the volume of loans carried over from the previous season in just over 18 percent of the banks examined; an additional 2 percent showed a “sharp” increase.  

Consumer Loans (Excluding Credit-card Lending).  Responses received during the six months ending March 31, 1997, from active consumer lenders showed the following:

• Banks made consumer loans without adequate collateral protection “frequently enough to warrant concern” in just under 13 percent of the banks examined; an additional 2 percent made loans lacking collateral protection “commonly or as standard procedure.”    
• Another practice cited by examiners was that the institution made loans to borrowers who lack demonstrable ability to repay.  Just over 14 percent of the banks examined were noted for making these loans “frequently enough to warrant notice;” and an additional 2 percent were cited for lending in this manner “commonly or as standard procedure.”  

Credit-Card Loans.  Just over 21 percent of the institutions examined during the six months ending March 31, 1997, were reported as active in credit-card lending.  Of these institutions, seven were credit-card specialty banks.  (Only one of these showed “above-average” risk in current underwriting practices for new credit-card loans, and the examiner explained that such loans are priced to account for credit risk).  The remaining institutions were not major players in credit-card lending, holding on average less than 1 percent of total assets in such credits.

• Eighty-nine percent of all institutions examined that were active in credit-card loans showed stable underwriting practices for new credit-card lending.  In comparison, 95 percent of the responses received during the previous six-month period showed stable practices.



• Eight percent of the responses from banks examined that were active in the credit-card market indicated that practices had tightened compared with only three percent showing looser practices.
• Ninety-five percent of responses indicated “average” or “below-average” risk in underwriting practices for new credit-card loans.

Home-Equity Loans.  Reports on the underwriting of home-equity loans continued to reveal few deficiencies among the 368 institutions actively making such loans.

Regional Results: General Underwriting Trends

According to the results from examiners received during the October 1996 - March 1997 reporting period, concerns regarding general underwriting practices were similar across the country. In each FDIC region, examiners cited the risk in loan administration as “above-average” more frequently than weaknesses in other underwriting practices.  Lenders making loans resulting in concentrations to one borrower or industry registered the second highest percentages of concern among examiners in four of the eight FDIC regions (Dallas, Memphis, New York, and San Francisco).  In both the Atlanta and Boston regions, examiners were more conerned that written lending policies differed from actual practices.  And, in the Kansas City and Chicago regions, the second greatest concern was the failure to adjust loan pricing on different quality loans.