
|
FDIC Outlook
In Focus This Quarter:
Rate/Volume Analysis: An Off-Site Approach to Measuring Interest Rate Risk
Interest rate risk (IRR) is the exposure of a bank's current or future earnings
and capital to interest rate changes.1 One
method of assessing the level of IRR, or the sensitivity of an institution's
earnings to changing interest rates, is by using rate/volume analysis
(RVA). RVA is an effective, after-the-fact, off-site monitoring technique
that measures IRR by analyzing the separate components of net interest
income (NII) over a specific time period. (See the inset box for more
on the basics of RVA.) In addition, RVA helps identify outlier institutions
where changes in interest rates significantly affect NII. Using the RVA
method, this article assesses the level of IRR among institutions insured
by the Federal Deposit Insurance Corporation (FDIC) across the country
during the 12-month period ending September 30, 2004.
Background
After two years of the lowest federal funds rates since the late 1950s,
short-term rates began to rise in June 2004 when the Federal Reserve's
Federal Open Market Committee (FOMC) raised the federal funds target interest
rate a quarter point to 1.25 percent. During the remainder of 2004 the
federal funds target rate was raised four more times, closing the year
at 2.25 percent. In contrast, 10-year Treasury notes fluctuated within
a band of between 4.73 percent and 4.10 percent during the same six-month
period, ending the year at 4.23 percent, which was slightly below year-end
2003 levels.2 Taken in combination, the changes
in short- and long-term interest rates during the year resulted in a modestly
flatter yield curve by the end of 2004 (see Chart 1). The yield curve
spread, defined as the difference between the yields on the 10-year Treasury
bond and the 3-month Treasury bill, declined from a high of 368 basis
points at May 2004 to 201 basis points by year-end 2004.3
Typically, a flattening yield curve tends to compress net interest margins
(NIMs) for financial intermediaries. However, despite a moderately flatter
yield curve, NIMs actually improved at the majority of the nation's community
banks.4 During the first nine months of 2004,
52 percent of community banks reported margin improvement, up from 27
percent a year earlier.
The improvement in NIMs during the recent period of a flattening yield
curve may be the result of a number of factors. First, this analysis is
through third quarter 2004 and represents only one-quarter of increases
in the federal funds rate; further, only one increase was effective for
the entire period. (Generally, there is a lag between rising market rates
and increases in funding costs for some institutions, so the effects of
the rate hike may not yet be fully reflected in funding costs.) In addition,
the rise in the federal funds rate has taken place gradually at what FOMC
statements have termed "a measured pace," which means that the increases
were relatively modest and widely anticipated by financial market participants.
Further, despite recent flattening, the yield curve spread of 201 basis
points at year-end 2004 remained relatively steep and above its 20-year
average of 180 basis points. The yield curve may need to flatten more
significantly before there is a pronounced negative effect on community
bank NIMs.
Rate/Volume Analysis: The Basics
In an effort to assess the impact of interest rate changes
on NII, RVA analyzes an institution's change in NII by separating
it into three components: changes in yields and costs (rate
variance), fluctuations in volume of earning assets and
interest-bearing liabilities (volume variance), and
residual interest income and expense arising from the combination
of rate and volume changes (mix variance). Conceptually,
RVA follows a performance attribution methodology by breaking
down NII into its components and then measuring the contribution
of each component during a given period. RVA can be applied
to any two consecutive periods (see Table 1).
Table 1
| Rate/Volume Analysis Methodology |
| Variance Measures |
Formula Components |
| Income rate variance |
(3Q2004 earning asset yield less 3Q2003 earning asset yield ) times 3Q2003 average earning assets |
| Expense rate variance |
(3Q2004 cost of funds less 3Q2003 cost of funds) times
3Q2003 average interest-bearing liabilities |
| Income volume variance |
(3Q2004 average earning assets less 3Q2003 average earning
assets) times 3Q2003 earning asset yield |
| Expense volume variance |
(3Q2004 average interest-bearing liabilities less 3Q2003
average interest-bearing liabilities) times 3Q2003 cost
of funds |
| Income mix variance |
(3Q2004 average earning assets less 3Q2003 average earning
assets) times (3Q2004 yield less 3Q2003 yield) |
| Expense mix variance |
(3Q2004 average interest-bearing liabilities less 3Q2003
average interest-bearing liabilities) times (3Q2004 cost
less 3Q2003 cost) |
Note: The "net" position for each of
the variance measures is the difference between the
income and expense variances. For example, the net rate
variance is equal to the income rate variance less the
expense rate variance. The sum of the three net variance
measures (rate, volume, and mix) should equal the total
change in net interest income during the period.
Source: Federal Deposit Insurance Corporation. |
RVA is just one analytical tool to evaluate IRR in the banking system. Although RVA is an excellent tool for assessing how interest rate changes affect the components of NII, it is limited in its ability to forecast future results. These limitations arise from a reliance on historical data as well as the potential for change in a bank's asset and liability mix over time. Regardless of the results of RVA for a past period, either balance-sheet repositioning or changes in the level or direction of interest rates could alter the degree and direction of IRR risk in future periods. In addition, RVA is limited in its ability to handle off-balance-sheet items.
Results of Rate/Volume Analysis
As of September 30, 2004, a trailing four-quarter RVA on the nation's
community banks indicates that the interest rate environment had a nearly
equivalent influence on asset yields and liability costs. These data suggest
that the IRR exposure of the average community bank during the past year
was relatively limited. The average asset yield of community banks fell
44 basis points to 5.62 percent, while the interest-bearing liability
cost fell by an almost equivalent amount, 45 basis points, to 1.84 percent.
As a result, the net interest spread for community banks remained relatively
stable at 3.79 percent during the period.5
Robust growth in earning assets, particularly in commercial, construction,
and residential real estate loans, contributed to the increase in interest
income (see Table 2).
|
Table 2
| An Interest Rate/Volume Analysis Indicates Little Interest Rate Risk Was Prevalent at Community Banks during 2004
|
| |
4 Quarters Ending 9/30/2003 |
4 Quarters Ending 9/30/2004 |
Rate/Volume Analysis |
| |
Average Balance ($) |
Income/ Cost ($) |
Rate (%) |
Average Balance ($) |
Income/ Cost ($) |
Rate (%) |
Volume ($) |
Rate ($) |
Volume/ Rate ($) |
Total($) |
Assets
Interest-Earning Assets
|
| Short-Term Investments: |
| Interest-bearing deposits |
12,465,622 |
235,472 |
1.89 |
11,133,288 |
194,035 |
1.74 |
-25,167 |
-18,217 |
1,947 |
-41,437 |
| Securities (includes United States, mortgage- backed securities,
subdivision, equities) |
230,463,431 |
9,339,732 |
4.05 |
255,868,829 |
9,544,735 |
3.73 |
1,029,576 |
-742,701 |
-81,872 |
205,003 |
| Federal funds sold/repurchased |
38,381,810 |
450,144 |
1.17 |
30,335,311 |
317,325 |
1.05 |
-94,370 |
-48,648 |
10,199 |
-132,819 |
| Trading assets |
95,974 |
3,917 |
4.08 |
64,710 |
2,264 |
3.50 |
-1,276 |
-559 |
182 |
-1,653 |
| Total short-term investments |
281,406,836 |
10,029,265 |
3.56 |
297,402,138 |
10,058,359 |
3.38 |
908,763 |
-810,124 |
-69,545 |
29,094 |
| Loans: |
| Real estate |
459,968,718 |
31,789,935 |
6.91 |
516,309,348 |
32,638,956 |
6.32 |
3,893,884 |
-2,712,602 |
-332,261 |
849,021 |
| Agriculture |
26,815,372 |
1,837,382 |
6.85 |
26,675,989 |
1,725,667 |
6.47 |
-9,550 |
-102,698 |
534 |
-111,715 |
| Commercial & industrial |
107,605,511 |
7,504,588 |
6.97 |
114,569,897 |
7,578,985 |
6.62 |
485,708 |
-386,308 |
-25,002 |
74,397 |
| Consumer |
58,952,463 |
5,255,430 |
8.91 |
57,263,410 |
4,731,651 |
8.26 |
-150,574 |
-384,213 |
11,008 |
-523,779 |
| Total loans |
664,700,220 |
46,957,306 |
7.06 |
727,200,873 |
47,234,409 |
6.50 |
4,415,317 |
-3,782,548 |
-355,667 |
277,103 |
| Lease Financing Receivables |
3,170,310 |
251,655 |
7.94 |
3,208,173 |
240,122 |
7.48 |
3,005 |
-14,367 |
-172 |
-11,533 |
| Total Interest-Earning Assets |
949,277,366 |
57,503,300 |
6.06 |
1,027,811,184 |
57,799,612 |
5.62 |
4,757,254 |
-4,120,087 |
-340,855 |
296,312 |
Liabilities
Interest-Bearing Liabilities |
| Interest-Bearing Deposits: |
| Transaction accounts |
93,180,729 |
785,716 |
0.84 |
100,599,189 |
665,802 |
0.66 |
62,554 |
-169,012 |
-13,456 |
-119,914 |
| Nontransaction accounts |
| Savings deposits (including money market deposit accounts) |
250,830,173 |
2,900,633 |
1.16 |
285,435,790 |
2,575,703 |
0.90 |
400,184 |
-637,203 |
-87,911 |
-324,930 |
| Time deposits > $100,000 |
128,850,970 |
3,826,251 |
2.97 |
138,552,362 |
3,408,348 |
2.46 |
288,084 |
-656,554 |
-49,433 |
-417,903 |
| Time deposits, all others |
255,584,260 |
8,076,314 |
3.16 |
251,126,435 |
6,472,930 |
2.58 |
-140,865 |
-1,488,481 |
25,962 |
-1,603,384 |
| Federal Funds |
17,750,355 |
244,710 |
1.38 |
21,955,977 |
270,602 |
1.23 |
57,980 |
-25,941 |
-6,146 |
25,892 |
| Other Borrowed Money |
45,278,892 |
2,232,537 |
4.93 |
54,732,243 |
2,247,923 |
4.11 |
466,110 |
-372,875 |
-77,849 |
15,386 |
| Total Interest-Earning Liabilities |
791,988,786 |
18,098,165 |
2.29 |
853,133,579 |
15,671,644 |
1.84 |
1,397,253 |
-3,549,721 |
-274,053 |
-2,426,521 |
| Changing Net Interest Income |
157,288,580 |
39,405,135 |
3.77 |
174,677,605 |
42,127,968 |
3.79 |
3,360,001 |
-570,366 |
-66,802 |
-2,722,833 |
Notes: Community banks are commercial banks with
assets of $1 billion or less. For this analysis, the sample was
limited to community banks open since September 30, 2001, and excludes
specialty banks. Subcategories will not always sum to the total
since detail on minor categories may not be listed.
Source: Federal Deposit Insurance Corporation Call Reports (merger-adjusted).
|
|
|
Only a small subsetabout 3.6 percent of the 6,904 institutions
analyzeddisplayed a high level of sensitivity to rate movements
during the period (see Table 3).6 Of these
institutions, approximately one-third reported a negative earnings reaction,
whereas the earnings of the remaining "highly sensitive" institutionsreacted
favorably to recent interest rate changes. Geographically, community
banks in the FDIC's San Francisco and Dallas Regions reported the highest
percentage of highly sensitive banks, and the Boston area reported the
lowest percentage. Despite these differences, the percentage of highly
sensitive institutions during this period was lowbelow 5 percentin
every FDIC region.7
|
Table 3
| Community Banks Displaying High Interest Rate Sensitivity (during the four quarters ending 9/30/04) |
| |
Community Banks Reporting a Large Negative
Reaction (%) |
Community Banks Reporting a Large Positive
Reaction (%) |
Total of All Community Banks Reporting
a High Sensitivity (%) |
| San Francisco |
1.7 |
3.3 |
5.0 |
| Dallas |
1.4 |
3.4 |
4.8 |
| New York |
1.5 |
2.5 |
4.0 |
| Chicago |
1.1 |
2.5 |
3.6 |
| Kansas City |
1.4 |
1.8 |
3.2 |
| Atlanta |
0.9 |
2.2 |
3.1 |
| Memphis |
1.1 |
1.3 |
2.5 |
| Boston |
0.8 |
1.2 |
2.0 |
| Nationwide |
1.2 |
2.4 |
3.6 |
Note: Sensitivity is defined as net rate variance
exceeding 20 percent of the prior four quarters' net interest income.
This variance reflects reaction to the specific interest rate environment
for a particular period. Different banks will likely display high
sensitivity during differing rate movements.
Source: Federal Deposit Insurance Corporation. |
Conclusion
Overall, this RVA analysis indicates that little IRR was prevalent among
community banks, at least over the past year. For the most part, asset
yields and liability costs of community institutions generally have moved
in tandem with the changing rate environment, with increases in earning
asset volume helping to raise NII. Even among the small group of institutions
highly sensitive to rate movements, only one-third reported an adverse
reaction to this rate environment, with the remainder reporting an improvement
in spread income. Notwithstanding, this analysis represents an aggregate
analysis of the nation's community banks, and vulnerability to changes
in interest rates will vary among individual institutions.
Ronald Sims II, CFA, Senior Financial Analyst
1 IRR includes repricing risk, basis risk,
yield curve risk, option risk, and price risk. For more information, see
the Federal Deposit Insurance Corporation Manual of Examination Policies,
http://www.fdic.gov/regulations/safety/manual/index.html.
2 Based on monthly average data for the 10-year
U.S. Treasury note.
3 Treasury yields represent monthly averages.
4 Community banks consist of all FDIC-insured
institutions with assets of less than $1 billion and exclude credit card
and other specialty institutions as well as de novo institutions.
5 The net interest spread is calculated slightly
differently from NIM. The net interest spread equals [interest income
divided by average earning assets] less [interest expense divided by average
interest-bearing liabilities]. The NIM equals [interest income divided
by average earning assets] less [interest expense divided by average earning
assets].
6 For this analysis, interest-rate-sensitive
banks are defined as those banks where the net rate variance exceeded
20 percent or more of prior period NII. See Table 1 for a definition of
net rate variance.
7 Geographical variations reflect differences
among individual institutions over this specific time period. Results
will change during different interest rate periods.
|
|