Log in / Register
Home arrow Business & Finance arrow Reengineering community development for the 21st century
< Prev   CONTENTS   Next >

Application of SPMs to CDFI Banks

Financial Performance

It is possible to compare the financial performance of the CDFI Banks with the rough peer group of banks with less than $1 billion in assets using SDI data. This cutoff was made simply because of tractability; ideally, other criteria would define the small-bank category more narrowly as community banks, as de Young, Hunter, and Udell (2004) and de Young (2006) have done.

A few results are immediately apparent from these data. First, at the bottom- line level of ROA, CDFI banks have been less profitable on average than small banks in general throughout the period, although the differential is less pronounced when the median is calculated, suggesting some outliers. The gap has also closed somewhat over the period (see Figure 7.2). Certainly, CDFI banks have increased median and average ROA over the period.

The difference in performance between CDFIs and the larger peer group of small banks appears to be due to several factors, explored more fully in de Young (2006). For example, although CDFI banks show a consistently lower median charge-off rate than small banks and earn a consistently higher gross interest margin (4.6 percent versus 3.6 percent), these positive factors are not sufficient to outweigh the effects that draw down profitability. These include the following:

• The efficiency ratio of CDFIs is more than a third higher than that of small banks in general (76 percent versus 56 percent in 2004), reflecting in part

Figure 7.2 Return on Assets: CDFIs and Small Banks Compared

lower average size of the CDFIs compared to the banks in the under $1 billion asset category; and

• CDFIs are consistently less geared (tier 1 leverage ratio of 9.2 percent versus 8.7 percent) and less lent out (loans to deposit ratios of 77 percent versus 104 percent), reflecting in part the less commoditized, more specialized nature of CDFI lending.

Social Performance

Figure 7.3 shows a scatter plot of DDI and DLI measures we calculated for the sample of CDFI banks in 2004. As explained earlier, DLI could only be calculated using HMDA data, that is, for home lending. This is a significant factor, since only thirteen banks in the sample can be described as housing focused, with single- and multiple family lending constituting more than 50 percent of assets in 2004. Figure 7.3 shows a scatter across the DDI-DLI space, indicating housing-focused lenders, for which DLI (HMDA) is a more useful measure, with diamond shapes. There is a notable grouping along 100 percent on the DDI line. Indeed, the median DDI for CDFI banks is 100 percent; that is, the median CDFI bank has all its branches in qualifying areas. The bulk of CDFI banks sit in the northwest quadrant, that is, have DDI and DLI above 50 percent.

However, the significance of an arbitrary segmentation–above and below 50 percent alone – may be questionable. We require benchmarks for these measures

Figure 7.3 DDI (Branches) Versus DDI[b] Breakout by Housing Focus

DDI (Branches) Versus DDI[b] Breakout by Housing Focus

from other categories of entities to make meaningful distinctions as to what level is low or high. Table 7.5 gives initial equivalent calculations for other categories of banks.

It is clear from Table 7.5 that CDFI banks are different from large U.S. banks and indeed even from smaller banks in general. Using norms for small or community banks, we could define four quadrants, with the expectation that CDFIs would then more clearly be clustered in the top right-hand quadrant, although, given that the DLI numbers are HMDA only, not all CDFIs will be located there: some may not do much housing lending, and the lending they do may be deliberately done in middle- income areas as a strategy to diversify risk or raise income to pursue development impact by other means. However, we would be surprised to find CDFI banks in the bottom left-hand quadrant–below median or average on both counts.

Seen through these DDI and DLI lenses, have CDFI banks changed over time? Figure 7.4 summarizes the average of these two measures for the CDFI bank category over the sample period. The DDI measure shows little variation over the period, returning to its 1996 average by 2004. However, there is evidence of an upward trend in DLI, which has risen from 45 percent to 55.2 percent, although it has fallen from the peak shown in 2000.

Underlying these averages are of course the stories of forty-two individual CDFI banks, which have adjusted their business models over time, presumably in response to many factors, including competition and desire for impact. If we define a material change in terms of the 2004 number being more than 10 percent

Table 7.5. DDI and DLI Comparisons


Number in Category

DLI (HMDA) Average



Top 10 U.S. banks by assets




Small banks (< $1bn)




All minority-owned banks




CDFI banks




Figure 7.4 CDFI Banks: DDI and DLI over time

CDFI Banks: DDI and DLI over time

different from the 1996 number, then twenty-three banks saw no material change over this time, while twelve had a material rise in DLI and seven a material decline. The vast majority maintained or increased their development lending focus in housing; and for those that did not, it is not possible to say whether this is the result of, for example, increasing emphasis on other development lending such as commercial real estate.

How do we bring all this information together in an indicative performance rating? Earlier we argued that a long-run average ROA may be the best financial performance measure. In Table 7.6, we report on the CDFI banks that have performed best according to this measure in the period 2000 to 2004, but using the DLI and DDI lenses to create robust comparison groups.

Table 7.6. Performance Measurements for CDFI Banks Using Social Performance Lenses

Housing Focused

No. of Banks

Above CDFI bank median DLI HMDA

ShoreBank: 1.04%

Pacific Global Bank:


Seaway National Bank of Chicago: 0.97%


Below CDFI bank median DLI (HMDA)

Carver Federal Savings Bank: 0.70%

Mutual Community Savings Bank, FSB: -0.10%


Above CB median DDI

11 Banks–Top 3: ShoreBank: 1.04% Pacific Global: 1.02% Seaway National: 0.97%

26 Banks–Top 3:

Central Bank of Kansas City: 2.63% International Bank of Chicago: 2.12% University National Bank: 1.75%

Below CB median DDI HMDA


6 Banks–Top 3:

Community Commerce Bank: 1.88% Park Midway Bank: 1.77%

Inter National Bank: 1.54%

First, Table 7.6 reports a segmentation using above and below CDFI bank median DLI, which, as discussed earlier, is relevant only for housing-focused lenders, a relatively small group, and then, the median DDI for community banks as a whole is used, and banks are further categorized according to their lending focus – whether or not a majority of loans in 2004 was for single- and multifamily housing. Of course, "not” covers several major categories we cannot distinguish from these available data alone. The table reports only the top three institutions in each category, ranked by the average ROA for the last live years.

This categorization could inform and assist a double bottom line investor who seeks the best financial return possible from investing in banks that, for example, focus on housing and are more “present” in poorer communities. These criteria would be a good starting point for further analysis by the investor. Providing an easy starting point for investment decisions is all mutual fund rating services seek to do.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science