Comment: Science Pays Off in Consumer Lending

The consumer lending business at many of the nation's commercial banks is faltering. Although margins for many products remain strong, weak demand coupled with the competitive challenge of nonbanks, product substitutes, and securitized offerings are slowing the growth of bank loan footings. From 1991 through 1993, the compound annual growth rate of bank revolving consumer loans came to just under 4%, compared to a 10-year figure of 11.1% (Rates for 1994 will shortly become available and are not expected to change the basic trend.)

In 1984, banks held 59% of outstanding consumer revolving loans; by 1993, that share had fallen to 49%. The bank share of auto loans dropped from 48% to 44% over the same period. In mortgage originations, banks have been taking share away from thrifts over the course of the last decade, although thrifts have recently recouped a bit of this loss. Both intermediaries, however, have been losing to the mortgage companies. Thus, in 1993, banks accounted for 26% of originations versus a 41% share in 1987.

In calculating true loan growth trends for individual banks, it is essential to adjust the totals for the impact of mergers and acquisitions as well as for asset securitization. FMCG's compilation of merger-adjusted and securitization-adjusted growth figures for a sample of 30 of the top money-center and superregional banks shows that the average institution managed only a 2.3% compound annual increase in consumer loans outstanding from 1991 through 1993.

As one might expect in most analyses of retail banking, however, the figures are quite skewed. Although average growth is disappointing, there are a substantial number of banks that greatly exceeded the average during this period. In fact, 12 of the 30 had acquisition-adjusted annual rates of growth that topped 12%. Conversely, there are a large number whose consumer lending totals actually dropped.

Growth leaders for the 1991-1993 period (figures are compound annual rates) include Signet (26%), Banc One (18%), Fifth Third (17%), First Chicago (17%) and Bank of New York (15%). Apart from their disdain for merely average performance, what factors predispose banks like these to excel while so many others languish? The short answer (or at least one short answer) is that they specialize. The top dozen high-growth consumer lending banks can be subdivided into three groups - card banks, other specialty banks, and only a few across-the-board consumer lenders.

About half of the growth leaders are card specialists that often spurn other areas. Thus, from 1991 through 1993, BONY grew its card portfolio by 32% per annum, but its home-equity loan and line business by only 4%. Other specialist types eschew cards just as emphatically. Fifth Third grew its installment loan portfolio by 21%, but its card portfolio by only 2%.

The top card banks have evolved a formula that probably will be emulated by most successful consumer-lending banks for the rest of the decade. These banks understand that the profitability of their customer populations is just as skewed as are bank growth trends. Where these banks excel is in their capacity to identify which consumers are likely to be profitable and which will probably prove unprofitable. Having done so, they proceed to alter their products to attract and retain a disproportionate share of the former.

The strategy is simply called customer data base management. The best credit card institutions begin by employing regression models that predict the potential for profitable purchase - i.e., the likelihood of being a "revolver" - in various market segments. Then they hypothesize changes in product features or distribution mechanisms that will appeal to those customers most apt to revolve without defaulting or going delinquent. In credit cards, such hypothesized changes run the gamut from lower rates to gifts or bonus points.

Next the banks select appropriate segments for testing and run a pilot test on a subsegment or "cell." Finally, they evaluate the results and roll out the successful tests to ever larger portions of the customer base. The net result is the ability to predict actuarially the product characteristics, service features, and prices that will stimulate enhanced profitable usage and lower customer attrition rates.

This type of approach - it is really just the application of the old- fashioned scientific method - is also beginning to be used by leaders in the mortgage business and by some in home-equity lines of credit and other products. And when applied to the deposit side of retail, the approach could reverse a decade-long slide in profitability.

Additionally, creative data-base marketing and customer management may spawn an entirely new business opportunity. Skilled customer data base managers are beginning to offer themselves as potential alliance partners to banks that have large customer bases, but precious little knowledge about them. Thus, the leaders in scientific customer management will be twice blessed - blessed in what they can do and also in what they can teach. So endowed, at least a handful of top consumer lenders will never have to endure the disappointment of merely average performance.

Mr. Zizka is a managing vice president for First Manhattan Consulting Group, New York. Mr. Rose, formerly senior columnist for this newspaper, is also associated with First Manhattan.

For reprint and licensing requests for this article, click here.
MORE FROM AMERICAN BANKER