Data from call reports paint a bleak picture for banks in large Eighth District cities. In aggregate, banks in the larger metropolitan areas are performing worse than banks in smaller markets.
Why are smaller banks doing better? The reasons are not yet clear. We used various metrics to help discover why by reviewing three groups of District banks: banks in the four major metropolitan statistical areas, or MSAs (St. Louis, Little Rock, Louisville and Memphis), banks in 18 mid-tiered metro areas (population between 50,000 and 500,000) and banks in 60 micropolitan statistical areas (population between 10,000 and 50,000). To gauge the performance of banks in these three groups, we looked at aggregate numbers for health of the overall industry and then looked at the median numbers for the health of the average banks. The total number of banks in each of these areas is similar: 145 in the major areas, 139 in the mid-tier metro areas and 175 in the micro areas.
The figure shows the aggregate return on assets (ROA) for the three bank groups. Until the end of 2006, all the groups were performing at a similar level of profitability, with an ROA of about 1 percent or higher. Since then, there has been a striking difference in performance. Banks in the major MSAs have been hardest hit, with a weighted ROA of -0.6 percent as of June 30, 2009. Banks in the micro and mid-tier metro areas have also experienced challenges, though not as severe. Micro-area banks have performed consistently better than the other two groups through this downturn, with a weighted ROA of 0.6 percent, while profitability of mid-tier metro area banks falls between the two groups, with an ROA of 0.2 percent.
On other aggregated metrics, such as nonperforming loans, loan losses and net interest margin, the conclusion is similar. Banks in micro areas are performing better than banks in mid-tier metro areas, which, in turn, are performing better than banks in major metro areas.
What affects the performance of the different groups of banks? When we aggregate ratios such as ROA, larger banks weigh more heavily than smaller banks within a geographic group. To overcome this size issue, we looked at the median bank performance metrics for the three groups of banks. The median banks in the three groups are similar in ROA size: $187 million in the major metro areas, $171 million in the mid-tier metro areas and $143 million in the micro areas. Although some differences in size still exist, for practical purposes these median banks are all small-sized community banks.
The table shows the median statistics and bank performance measures for the three groups of banks. Even at the median level, banks in micro areas performed best, followed by mid-tiered metro area banks and then the major metro banks. However, the performance range narrows when switching from aggregate to median. The median (50th percentile) ROAs now range from 0.54 percent in major MSAs to 0.91 percent for the micro market banks. This narrowing of the performance band shows that the performance of larger banks in the major and mid-tier metro markets pulls down the averages. Sizable differences still exist in the average bank profitability for different size markets after accounting for outliers. This difference in performance holds for most of the metrics we analyzed, including the ratio of nonperforming loans, net interest margin and coverage ratios.
All the factors influencing this performance difference may not be fully understood today. However, our experience with commercial bank examinations raises some possible explanations. One distinguishable factor is the level of commercial real estate (CRE) lending. The proportion of CRE loans to total loans at a micro area bank (18.3 percent) is 3.9 percentage points less than at a mid-tier metro area bank (22.2 percent) and 10.5 percentage points below a major metro bank (28.8 percent). CRE, which is under severe stress in the current environment, could be a possible reason for the difference in performance.
Another possible explanation is that smaller market banks may lend more on a relationship basis compared with transaction lending in larger banks. Economic factors, such as higher unemployment rates in the larger markets, could also be at play. It is also possible that the smaller market banks could be somewhat slower to recognize troubled assets. If this is true, then the performance gap should close in time.
Smaller community banks have outperformed their larger market brethren during this economic downturn both at the aggregate and individual level. In the ensuing quarters, the reasons will become clear. We may conclude that banks in smaller markets have indeed been more conservative and managed risk better. Or, we may find it takes varying amounts of time for risk recognition to work through all sizes of banking markets. As Paul Harvey famously stated, “Stay tuned for the rest of the story.”
|Major Metro Areas||Mid-Tier Metro Areas||Micro Areas|
|Number of Banks||145||139||175|
|Median Assets||$186.7 million||$170.7 million||$143 million|
|Return on Assets||0.54%||0.67%||0.91%|
|Net Interest Margin||3.53||3.82||3.86|
|Nonperforming Loans / Total Loans||1.88||1.46||1.08|
|Loan Loss Reserves / Nonperforming Loans||77.29||97.87||115.63|
|Loan Losses / Total Loans||0.31||0.36||0.23|
|Tier 1 Leverage Ratio||9.28||9.08||9.1|
|CRE to Total Loans||28.84||22.21||18.32|
Previous MSA banking condition reports:
Keep up with what’s new and noteworthy at the St. Louis Fed. Sign up now to have this free monthly e-newsletter emailed to you.
Fed in Print: An index of the economic research conducted by the Fed.