Every year Nolan studies trends in banking to help our clients focus their improvement initiatives in the areas that should provide the greatest profit improvement potential.

Using the results of our 2003 study (data as of 12/31/02 and for the 2002 calendar year), we constructed four data models designed to identify any of the almost 1,100 study statistics that might tend to be more predictive of benchmark (top quartile) efficiency ratio performance.

With these models we were able to construct rules—sub-sets of line of business level performance statistics—that when grouped together are more statistically significant in predicting benchmark efficiency ratios than when analyzing all the study data taken collectively. Said another way, we wanted to simplify the results by identifying those lines of business that have the greatest impact on the overall efficiency ratio, thereby sharpening the focus on certain functions to achieve better overall results.

The first model looks at line of business category performance. Categories are groups of functions such as Administration, Commercial Banking, Retail Banking, Consumer Lending, Trust, etc. Our model determines whether top quartile performance in any of these categories correlates to overall bank efficiency ratio.

The second model studies each function within category or line of business. One example is the category of Retail Banking which is comprised of the lines of business of Retail Banking Administration, Branches and Deposit Operations. Another example is the category of Commercial Banking which is comprised of Corporate Lending, Commercial Real Estate Lending, Middle Market Lending and Small Business Lending along with Commercial Loan Operations and Cash Management. The functional or line of business level is the most granular level of detail reported in the study. The model relates top quartile performance in each line of business to overall bank efficiency ratio.

The third and fourth models are similar to the first two but instead of relating top quartile performance by category and line of business detail to top quartile overall bank efficiency, they use the broader top 50 percent overall bank efficiency ratio as the desired outcome.

Two of the most significant findings were those that designated three study categories—Commercial Banking, Retail Banking and Administration—as having greater impact on overall bank efficiency ratio than others like Consumer Lending, Direct Banking, Trust, Credit Card and Mortgage Lending.

The $1 billion plus asset-sized study participants that have top quartile efficiency ratios in Commercial Banking and Retail Banking are two and one-half times more likely to have a top 50 percent overall bank efficiency ratio. The models also reveal a strong indication that top quartile performance in the Commercial Banking and Administration categories indicates a better chance (43 percent odds) of a top quartile overall bank efficiency ratio. Said another way, if a bank does not perform in the top quartile of Commercial Banking or Administration, the chances of top quartile performance in overall bank efficiency ratio are about one third less.

The line of business level models identified efficiency ratio results for Information Systems Operations and Purchasing/Administrative Services as having the most influence on the Administration category efficiency ratio. Deposit Operations efficiency was the key driver of Retail Banking results and all Commercial Lending origination areas (Corporate, Commercial Real Estate, Middle Market and Small Business) along with Commercial Cash Management were the key areas that predict performance in the Commercial Banking category.

So what can be concluded from these findings? First, while benchmark performance in every line of business may be a goal to strive for, high performance in certain areas is a must. Second, by achieving top quartile performance in the areas identified by our models, the odds of attaining benchmark-level overall bank efficiency is increased dramatically.§