FR2644_20250329_omb_B

Supporting Statement Part B for the
Weekly Report of Selected Assets and Liabilities of Domestically Chartered
Commercial Banks and U.S. Branches and Agencies of Foreign Banks
(FR 2644; OMB No. 7100-0075)
Summary
For all information collections that involve surveys or require a statistical methodology,
the Board of Governors of the Federal Reserve System (Board) is required to provide a complete
justification and explanation of the use of such a methodology. For collections that employ
surveys without such a methodology, the Board should be prepared to justify its decision not to
use statistical methods in any case where such methods might reduce burden or improve
accuracy of results.
Background
The FR 2644 is a balance sheet report that is collected as of each Wednesday from an
authorized stratified sample of 850 domestically chartered commercial banks and U.S. branches
and agencies of foreign banks. The FR 2644 is the only source of high-frequency data used in the
analysis of current banking developments. The FR 2644 collects sample data that are used to
estimate universe levels for the entire commercial banking sector in conjunction with data from
the quarterly commercial bank Consolidated Reports of Condition and Income (FFIEC 031,
FFIEC 041, and FFIEC 051; OMB No. 7100-0036) and the Report of Assets and Liabilities of
U.S. Branches and Agencies of Foreign Banks (FFIEC 002; OMB No. 7100-0032) (Call
Reports). Data from the FR 2644 and the Call Reports are utilized in construction of weekly
estimates of U.S. bank credit, balance sheet data for the U.S. commercial banking sector, and
sources and uses of banks’ funds, and to analyze current banking developments, including the
monitoring of broad credit and funding conditions. The Board publishes the data in aggregate
form in the weekly H.8 statistical release, Assets and Liabilities of Commercial Banks in the
United States, which is followed closely by other government agencies, the banking industry,
financial press, and other users.1 The H.8 release provides a balance sheet for the commercial
banking industry as a whole as well as disaggregated data for three bank groups: large
domestically chartered banks, small domestically chartered banks, and U.S. branches and
agencies of foreign banks. The data are also used in constructing the commercial bank
component of the Federal Reserve’s G.19 release, Consumer Credit,2 and the Federal Reserve
Bank of New York’s Reserve Demand Elasticity (RDE).3

The H.8 release is available on the Board’s website, http://www.federalreserve.gov/releases/h8/current/default.htm.
The G.19 release is available on the Board’s website,
https://www.federalreserve.gov/releases/g19/current/default.htm.
3
The Reserve Bank of New York’s Reserve Demand Elasticity (RDE) is available on the Federal Reserve Bank of
New York’s website,
https://www.newyorkfed.org/research/reserve-demand-elasticity/#interactive.
1
2

Universe and Respondent Selection
The FR 2644 panel has an authorized maximum size of 850 domestically chartered
commercial banks and U.S. branches and agencies of foreign banks. Currently, the panel consists
of 761 total reporters4—696 domestically chartered banks and 65 foreign-related institutions—
covering all 12 Federal Reserve Districts. The panel accounts for about 89 percent of the total
assets of U.S. commercial banks, as well as a high level of coverage for most reported items. The
number of respondents is less than the authorized size due to mergers among reporters and loss
of respondents due to the voluntary nature of the collection. The proposed reduction in reporting
frequency for smaller institutions is intended to both address burden minimization for these
banks as well as provide a mechanism to increase panel participation.
Table 1 presents the number of reporters disaggregated by district and by bank group for
the current panel.
Table 1
Number of Reporters on the Current FR 2644 Panel5
District
1
2
3
4
5
6
7
8
9
10
11
12
Total

Domestically
Chartered
23
48
38
43
35
61
143
73
29
68
69
66
696

Foreign-Related
0
57
0
0
1
0
3
0
0
0
2
2
65

Pending
Replacements6
5
1
5
3
5
7
2
9
7
1
35
9
89

The current FR 2644 sample’s coverage for each data item is included in Attachment 1.
The accuracy of the sample can be assessed by the size of the revisions when the data are
benchmarked to the quarterly Call Report data, which cover the full universe of commercial
banks. The accuracy experienced with the current panel is presented in Table 2.a for small banks
and Table 2.b for foreign-related institutions, summarizing the benchmark effects since the last
renewal in 2022. While the average revisions are not overly large, they are significant. Therefore,

4

As of April 24, 2024.
As of April 24, 2024, 761 respondents filed the FR 2644. There have been continuous efforts underway to restore
the panel size to 850 reporters.
6
These are the number of banks to which each district has sent out invitations to join the panel that have not yet
been accepted or declined.
5

2

the Board recommends retaining the current authorized sample size of 850 respondents to avoid
a deterioration in accuracy.
Table 2.a
Recent Benchmark Revisions to Estimates for Small Banks7
($ millions, n.s.a.8)

Item
Bank Credit
U.S. Treasury and Agency
Securities, MBS
U.S. Treasury and Agency
Securities, non-MBS
Other Securities, MBS
Other Securities, non-MBS
CLD Loans
Farm Loans
Multifamily (Over 5) Loans
Nonfarm Nonresidential Loans
Closed-end 1-4 Family Loans
Revolving Home Equity Loans
Commercial and Industrial Loans
Credit Card and Other Revolving
Auto Loans
All Other Consumer Loans
Loans to Nondepository Fin Inst.
All Other Loans
Cash
Total Assets
Other Deposits
Large Time Deposits
Total Borrowings
Total Liabilities

Root Mean
Root Mean Maximum Average Level
Square
Square
Absolute of Asset Item
Percentage
Revision9
Revision
($ billions)10
Revision11
14,791
27,017
5,322
0.28
3,025

7,325

626

0.48

4,704
740
1,508
888
463
1,283
2,159
2,394
388
4,136
495
249
1,790
4,060
836
8,706
16,901
14,012
2,437
6,105
14,238

11,961
1,581
2,358
1,434
1,108
1,910
3,359
4,306
808
8,683
1,131
500
4,024
13,692
1,589
19,617
29,845
22,298
6,201
12,953
24,715

267
38
374
285
99
280
1,073
752
93
713
162
73
167
108
212
568
6,269
4,837
437
334
5,752

1.76
1.94
0.40
0.31
0.47
0.46
0.20
0.32
0.42
0.58
0.30
0.34
1.07
3.77
0.39
1.53
0.27
0.29
0.56
1.83
0.25

The root mean square revision as a result of 12 quarterly benchmarks between March
2021 and December 2023 was $14.8 billion or 0.28 percent of the average level of bank credit.
This measure implies that benchmarking revises quarterly bank credit growth estimates for small
7

Summary statistics are calculated for 12 quarterly benchmarks from March 2021 to December 2023.
n.s.a. = not seasonally adjusted.
9
The root mean square revision is the square root of the averaged sum of squared revisions. This term may also be
referred to as the standard deviation of the revisions around zero.
10
Average levels are averages of weekly data over the period of January 2021 to December 2023.
11
Root mean square revision divided by average level of asset item, multiplied by 100.
8

3

banks by an average of 1.1 percent at an annual rate. The maximum revision was 1.8 times
greater than that amount.
Table 2.b
Recent Benchmark Revisions to Estimates for Foreign-Related Institutions12
($ millions, n.s.a.13)

Item
Bank Credit
U.S. Treasury and Agency
Securities, MBS
U.S. Treasury and Agency
Securities, non-MBS
Other Securities, MBS
Other Securities, non-MBS
CLD Loans
Farm Loans
Multifamily (Over 5) Loans
Nonfarm Nonresidential Loans
Closed-end 1-4 Family loans
Commercial and Industrial Loans
Loans to Nondepository Fin. Inst.
All Other Loans
Cash
Total Assets
Other Deposits
Large Time Deposits
Total Borrowings
Total Liabilities

Root Mean
Root Mean Maximum Average Level
Square
Square
Absolute of Asset Item
Percentage
Revision14 Revision
($ billions)15
Revision16
5,402
9,177
1,164
0.46
393

866

38

1.04

517
13
574
734
3
246
998
69
3,316
954
754
15,394
12,772
3,686
6,285
6,298
11,737

855
30
1,484
1,416
6
679
2,348
199
6,516
1,814
1,876
32,079
22,496
8,492
17,739
12,243
19,876

107
2
91
12
0
16
66
1
452
165
212
1,224
3,118
493
782
967
3,115

0.49
0.69
0.63
6.11
2.13
1.56
1.51
8.88
0.73
0.58
0.36
1.26
0.41
0.75
0.80
0.65
0.38

As shown in the last column of Table 2.b, the percentage root mean square revisions over
the past 12 benchmarks for foreign-related institutions greatly exceed those for the small banks
shown in Table 2.a. The root mean square revision of $5.4 billion, or 0.46 percent of the average
level of bank credit, implies an average 1.8 percent benchmark revision at an annual rate. Some
components of bank credit and total assets are significantly worse in accuracy; components of
commercial real estate loans, for example, have average benchmark revisions ranging from over
6 percent to over 24 percent at an annual rate and cash has an average benchmark revision of
12

Summary statistics are calculated for 12 quarterly benchmarks from March 2021 to December 2023.
n.s.a. = not seasonally adjusted.
14
The root mean square revision is the square root of the averaged sum of squared revisions. This term may also be
referred to as the standard deviation of the revisions around zero.
15
Average levels are averages of weekly data over the period of January 2021 to December 2023.
16
Root mean square revision divided by average level of asset item, multiplied by 100.
13

4

over 5 percent.
Procedures for Collecting Information
The FR 2644 currently collects 29 balance-sheet items as of each Wednesday from an
authorized maximum stratified sample of 850 domestically chartered commercial banks and U.S.
branches and agencies of foreign banks.
The FR 2644 report would continue to be submitted weekly, as of the close of business
each Wednesday, for banks with $5 billion or more in total assets. The Board proposes that the
smallest banks (below $5 billion in total assets as of the prior June 30 Call Report), which have
much smaller intramonthly movements than the large banks, be given the option to file data for
the first Wednesday of every month only, which will be utilized to derive weekly interpolated
data. Weekly data are needed for accurate and timely construction of universe estimates used in
the analysis of current banking developments. The balance sheet series are calculated and
published weekly. The various data series are routinely monitored by staff and regularly included
in materials prepared for the Board of Governors, the Federal Open Market Committee, and in
public reports (including, for example, the semiannual Monetary Policy Report to Congress).
Monitoring of these weekly data has also proven important during periods of rapid change in
banking conditions or of volatility in financial markets. If these data were available on a less
timely basis, staff’s ability to monitor banking conditions, particularly during periods of rapid
change or volatility, would be hindered. Currently, respondents file the FR 2644 weekly, within a
week of the as of Wednesday report date. The data are transmitted to the Reserve Banks via
Reporting Central.
Methods to Maximize Response
The process to recruit respondents for the FR 2644 panel uses an automated program,
Centralized Panel Maintenance System (CPMS), which, after verifying that there is still room on
the panel based on outstanding panel requests, selects one replacement bank randomly each day
based on panel needs. The largest banks are asked first; however, a bank cannot be asked to join
the panel more than once every two years. When the program exhausts the largest banks,
recruitment moves to strata needs. The estimation process for the H.8 release works most
efficiently with at least the minimum number of banks in each stratum. In addition, banks can be
asked to join the panel to satisfy needs for the estimation of particular items; for instance, smaller
banks with large amounts of commercial real estate loans are targeted. Responses to the panel
requests may take several months, including preparation time.
Testing of Procedures
There is no testing of procedures.

5