Regional Bank Health: Trends in Net Charge-Offs
According to a survey by the American Bankers Association, regional banking organizations (RBOs) operate in all 50 states, and in 2012 they were responsible for more than $1.7 trillion in lending to the communities in which they operated. RBOs not only generate a large amount of loans, lending also constitutes a significant segment of the RBOs’ balance sheets—net loans and leases constitute over half of their total assets, according to a recent article in the Quarterly Journal of the Clearinghouse.org. Even though no one regional bank is likely to be so large as to be systemically important—RBOs are generally considered to be bank holding companies with between $10 billion and $50 billion in assets—their collective impact on the US economy could be substantial.
For this reason, we want to assess the health of RBOs’ loan portfolios by analyzing their net charge-off behavior over the past two years. Net charge-offs are the difference between loans that have been deemed uncollectable and written off the bank’s balance sheet—charge-offs—and any subsequent recovery of those loans. Net charge-offs are often used by researchers as a proxy for bank risk because they tend to increase with riskier lending activities. We use quarterly data from SNL Financial to analyze regional banks’ net charge-offs over the past two years.
Net Charge-Offs as a Percentage of Total Loans and Leases
As a percent of total loans and leases, total net charge-offs have fallen from about 0.20 percent in the first quarter of 2012 to about 0.10 percent in the first quarter of 2014. In other words, the banks in our sample have been writing off increasingly smaller fractions of outstanding loans over the past couple of years. This is good news. This trend could result from either declining charge-offs or from increased recovery rates on previous charge-offs. A closer look at charge-offs and recoveries suggests a decline in charge-offs is more likely.
Charge-offs fell from $1.4 billion to about $800 million over the couple of years before 2014, which equals a decline from about 0.25 percent of total loans to about 0.10 percent. During the same period, recoveries hovered between $200 million and $400 million, or, as a percent of total loans, between 0.05 percent and 0.03 percent.
Decomposing Net Charge-Offs across Business Lines
Now that we’ve drawn the big picture, let’s take a closer look at the composition of net charge-offs across different business lines to see which, if any, loan type is driving the overall trend. We examine five broad business lines: real estate loans, agricultural production loans, commercial and industrial (C&I) loans, consumer loans, and all other loans.
In the context of this discussion, we should mention that banks follow different criteria when writing off different types of delinquent loans. In the case of consumer loans, banks generally follow a uniform charge-off policy set by the banks’ regulators: open-end credit (such as a home equity line of credit) is written off at 180 days delinquency, and closed-end credit (such as an auto loan) is written off at 120 days delinquency. The criteria for other loans, such as C&I loans, are less stringent and more subject to standards that the bank’s management sets.
The two loan categories that comprise the largest shares of total net charge-offs are consumer loans and real estate loans. In the most recent quarter for which we have data (2014:Q1), each of these categories comprised about 39 percent of net charge-offs, while C&I loans were at a distant third, with 19 percent. Net charge-offs of consumer loans have gained a greater share of total net charge-offs since 2012, while real estate’s share has fallen by almost half. This is consistent with the expectation that charge-offs tend to follow growth in the bank’s loan portfolios. That is, changes in the dollar value of charge-offs is typically proportional to the growth in the loan portfolios. In recent years, RBOs have experienced higher growth rates in their consumer lending than in their real estate lending.
Since the composition of net charge-offs tends to vary over time, we construct a quarterly charge-off concentration index akin to those used to measure market concentration. The index can be used to quickly assess concentrations of charge-offs across loan categories. We compute this concentration index as the sum of the squared share of each loan category's net charge-offs in each quarter. Concentration indices are typically bounded between 0 and 1, and higher values would indicate, in the present case, that charge-offs are being driven by a particular loan type. To illustrate the interpretation, two extreme examples of the calculation we made are provided below.
|Loan category||Share of net charge-offs||Share of net charge-offs (squared)|
|Real estate loans||1||1|
|Agricultural production loans||0||0|
|Loan category||Share of net charge-offs||Share of net charge-offs (squared)|
|Real estate loans||0.20||0.04|
|Agricultural Production Loans||0.20||0.04|
Our concentration index has been declining since 2012. From this, we can conclude that net charge-offs have become more dispersed across loan categories and that no one loan business line is driving the overall trend.
Regional Differences in Net Charge-Offs
Next we compare the composition of net charge-offs and loan portfolios across a few Federal Reserve Districts: New York, Richmond, Kansas City, and Cleveland. These four districts (along with Minneapolis, which we don’t include because it has only one regional bank) had the highest average ratios of net charge-offs to total loans of the 12 districts, as of 2014:Q1. New York had the largest number of regional banks in its jurisdiction (10 RBOs) at the time, Richmond had three, Kansas City five, and Cleveland three.
The comparison of charge-offs and loan composition data is consistent with the expectation that charge-offs and loan growth tend to move in tandem (though we acknowledge that we are looking at only a single point in time). For example, in the case of RBOs in the Cleveland and Richmond Districts, real estate loans comprise a majority of the lending portfolio, 74 percent and 86 percent, respectively, and consistent with this, the charge-offs on real estate loans are indeed larger compared to other loan categories in these districts, 60 percent and 86 percent, respectively. The pattern of charge-offs at RBOs in the New York and Kansas City Districts is somewhat consistent with this expectation. While real estate loans do comprise a larger share of the lending portfolio at these banks, it appears that the charge-offs on consumer loans are greater than that of real estate loans. In general, these patterns in charge-offs appear to be similar to those of recent quarters, according to transcripts of the earnings conference calls of some of the RBOs (the two publicly traded RBOs in the Cleveland District and two of the largest publicly traded RBOs in the New York District).
The table below shows the aggregated ratio of net charge-offs to loan balances by loan category in each of the four districts (excluding agricultural production loans, which comprise a very small proportion of total loans). The ratios in the four districts are in line with the aggregate trends discussed earlier. At the RBOs in these districts, charge-offs are higher for consumer loans than the other types of loans. In addition, the banks in the New York District are charging off proportionally greater amounts in their C&I portfolio compared to the other districts.
|District||Real estate||C&I||Consumer||All other|
Source: SNL Financial.
Our analysis has shown that regional banks have been writing off increasingly smaller amounts of loans over the past couple of years, and that these net charge-offs have become less concentrated in particular loan categories. By this one measure, at least, the evidence suggests that regional bank loan portfolios may have become less risky.