[Federal Register Volume 59, Number 199 (Monday, October 17, 1994)]
[Unknown Section]
[Page 0]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 94-25630]


[[Page Unknown]]

[Federal Register: October 17, 1994]


-----------------------------------------------------------------------

DEPARTMENT OF THE TREASURY
Office of Thrift Supervision
[No. 94-221]

 

Monthly Median Cost of Funds Index

AGENCY: Office of Thrift Supervision, Treasury.

ACTION: Notice; request for comment.

-----------------------------------------------------------------------

SUMMARY: Since 1982, the Office of Thrift Supervision (OTS) or its 
predecessor, the Federal Home Loan Bank Board (FHLBB), has collected 
data from savings associations for, and published, the Monthly Median 
Cost of Funds (MMCOF) index. This index is available for adjusting the 
interest rate on adjustable rate mortgages (ARMs).
    The MMCOF has become less important to the mortgage lending 
industry since development of the Eleventh District Cost of Funds Index 
and the One Year Treasury Bill Rate Index. At the same time, as the 
OTS-regulated portion of the thrift industry has contracted, the MMCOF 
has been based on data from a significantly smaller number of 
institutions, representing a smaller portion of mortgage loans.
    The OTS is currently evaluating the direct cost to the industry of 
supplying data for use in calculating the MMCOF and the indirect cost 
to the industry of OTS committing resources used to calculate the 
MMCOF. The OTS is considering whether it should: cease publishing the 
MMCOF; or modify the scope of the MMCOF to increase its usefulness or 
reduce the cost to savings associations.
    Potential modifications include expanding the number of reporting 
institutions to include all OTS-regulated savings associations or 
decreasing the reporting universe to a statistically valid sample of 
OTS-regulated institutions.
    If the OTS were to cease publishing the MMCOF, section 402(e)(3) of 
the Financial Institutions Reform, Recovery and Enforcement Act of 1989 
(FIRREA), requires the agency to designate acceptable substitute 
indices that could be used for adjusting ARMs currently based on the 
MMCOF. After reviewing available indices, OTS found that the Quarterly 
Average Cost of Funds Index (QCOFI) and the Eleventh District Cost of 
Funds Index appear to be acceptable substitute indices. Both of these 
indices are readily available for adjusting ARMs and have demonstrated 
a strong correlation over time to the MMCOF. If the agency determines 
to stop publishing the MMCOF, it anticipates providing a transition 
period of one or two quarters before releasing the last MMCOF.

DATES: Comments must be received on or before November 16, 1994.

ADDRESSES: Comments should be directed to Director, Information 
Services Division, Public Affairs, Office of Thrift Supervision, 1700 G 
Street NW., Washington, DC 20552, Attention Docket No. 94-221. These 
submissions may be hand delivered to 1700 G Street NW., from 9 a.m. to 
5 p.m. on business days; they may be sent by facsimile transmission to 
FAX number (202) 906-7755. Comments will be available for inspection 
from 1 p.m. until 4 p.m. on business days. Visitors will be escorted to 
and from the Public Reference Room at established intervals.

FOR FURTHER INFORMATION CONTACT: Stephen A. Whatley, Financial Analyst, 
(202) 906-7228, William Shively, Acting Deputy Assistant Director (202) 
906-5701, Supervisory Operations; Catherine A. Shepard, Senior 
Attorney, (202) 906-7275, Regulations and Legislation Division, Office 
of Chief Counsel, Office of Thrift Supervision, 1700 G Street NW., 
Washington, DC 20552.

SUPPLEMENTARY INFORMATION:

I. Background

A. Purpose of the MMCOF

    The MMCOF is a monthly index published by the OTS that can be used 
to adjust the interest rate on ARMs. The FHLBB first published the 
MMCOF in December, 1982. Before the MMCOF, the only national COF index 
available to savings associations to adjust interest rates on ARMs was 
the semiannual cost of funds index the FHLBB introduced in 1979, when 
ARMs first began to be used by the thrift industry to help manage 
interest-rate risks. Because this COF was only published semiannually, 
many mortgagors would wait until the rate was published to make their 
refinancing decisions. The result was a semiannual surge in the number 
of refinancings.
    The FHLBB and the savings and loan industry believed that creating 
a monthly cost of funds index, the MMCOF, would help disperse 
throughout the calendar year the number of refinancings associated with 
both fixed rate mortgages and ARMs that use COF indices. Since that 
time, other monthly indices have been developed and have become far 
more widely used than the MMCOF.

B. Current Usage of the MMCOF

    The Federal Housing Finance Board's (FHFB) Conventional Mortgage 
Interest Rate Survey (MIRS), indicates that the MMCOF is currently of 
minor importance to the mortgage lending industry. This survey showed 
that instruments used to adjust the interest rate on ARMs included:
1. Treasury bills with maturities of less than one year;
2. One-year Treasury bills;
3. Treasury notes/bonds with maturities greater than one year;
4. Eleventh District Cost of Funds;
5. Federal Housing Finance Board contract rate series on previously 
occupied homes; and
6. Other cost of funds indices, including the MMCOF.

    According to the March 1994 MIRS, of ARMs closed in March 1994, 
57.7 percent were adjusted by the one-year Treasury constant maturity 
index, while 21.3 percent of the ARMs were tied to the Eleventh 
District COFI. Combined, these two indices controlled 79.0% of the new 
ARM business. Only 1.8% of ARMs closed in March 1994 were adjusted with 
indices included in the ``Other Cost of Funds Indexes,'' which includes 
mortgages adjusted with the MMCOF. When both fixed rate mortgages and 
ARMs were considered, the ``Other Cost of Funds'' mortgages were less 
than one percent of all loans closed.
    While the OTS has no data that explain why the MMCOF has been used 
by such a small percentage of the ARM market, it notes that the 
universe of reporting institutions providing the data from which the 
MMCOF is calculated has declined dramatically since its inception. In 
August 1989, the MMCOF was calculated from data reported by 
approximately 3,350 FHLBB-regulated, Federal Savings and Loan Insurance 
Corporation-insured institutions. The then relatively small universe of 
FHLBB-regulated, Federal Deposit Insurance Corporation (FDIC)-insured 
federal savings banks was not included in calculating the MMCOF. 
Although these institutions were few in number, they then represented, 
and continue to represent, approximately 9-10% of the industry's 
assets.
    The number and aggregate assets of OTS-regulated, Savings 
Association Insurance Fund (SAIF)-insured institutions declined rapidly 
after FIRREA, with the result that the MMCOF was being prepared from an 
increasingly smaller universe. Only 1,614 institutions were reporting 
as of March 1994. This reporting universe includes only OTS-regulated, 
SAIF members. It no longer includes approximately 221 FDIC-regulated, 
private sector SAIF-insured institutions.

C. Current Costs of the MMCOF

    The OTS estimates that it costs the industry approximately $400,000 
per year to complete and file the MMCOF index data 
electronically.1 The total annual cost to the agency to prepare 
and distribute the MMCOF, which OTS does not use for any supervisory or 
regulatory purpose, is between $50,000 and $100,000. Accordingly, the 
OTS believes that the significant resource costs it incurs in 
calculating and publishing the MMCOF and the relatively high data 
preparation cost to the industry may outweigh the continued usefulness 
of the publication, at least in its current form. Therefore, OTS is 
reviewing: (1) whether the usefulness of the MMCOF can be increased or 
the burden on the industry and OTS can be decreased through modifying 
the scope of the MMCOF; or (2) whether OTS should stop publishing the 
MMCOF.
---------------------------------------------------------------------------

    \1\OTS estimates that filing data for the MMCOF cost 
approximately $20 per report, per month, per savings association. 
Before January, 1993, the MMCOF was calculated based on information 
extracted from the monthly Thrift Financial Report (``TFR''). As of 
January 1, 1993, OTS abolished the monthly TFR to reduce the 
regulatory reporting burden on the industry. Since then, OTS-
regulated SAIF-insured savings associations have been required to 
submit information directly to OTS for calculation and publication 
of the MMCOF.
---------------------------------------------------------------------------

D. Statutory Requirements

    Section 402(e)(3) of FIRREA requires the OTS to take such action as 
is necessary to assure that the indices prepared by the FHLBB, 
including the MMCOF, immediately prior to the enactment of FIRREA and 
used to calculate the interest rate on ARMs continue to be published. 
Under section 402(e)(4), the OTS may cease publication of an index if 
it substitutes an index that is substantially similar to the index that 
is being deleted and if the Director of OTS determines, after notice 
and opportunity for comment, that:

1. the new index is based on data substantially similar to that of the 
original index; and
2. the substitution of the new index will result in an interest rate 
substantially similar to the rate in effect at the time the original 
index became unavailable.

    This statutory protection is designed to mitigate the potential 
impact of substantially modifying or stopping publication of an index 
on outstanding mortgages. Currently, ARMs purchased by both the Federal 
National Mortgage Association and the Federal Home Loan Mortgage 
Corporation contain ``adjustable rate note'' language discussing the 
index to be used and what must occur if an index is no longer 
available. The standard language is: ``If the Index is no longer 
available, the Note Holder will choose a new index which is based upon 
comparable information. The Note Holder will give me notice of this 
choice.'' This means that the mortgage holder does not have to 
renegotiate with a borrower so long as the new index selected is based 
upon comparable information and the borrower is notified of the change.
    In determining whether to modify the MMCOF, OTS must therefore 
consider whether such a modified index would be based on comparable 
data and provide comparable results. Similarly, in considering stopping 
publication of the MMCOF, OTS must determine whether acceptable 
substitute indices are available.

II. Potential Modifications to the MMCOF

    OTS has considered two potential modifications to the MMCOF and 
believes that neither would substantially affect the information 
provided in the MMCOF or the results obtained by using it in adjusting 
ARMs.

A. Use a Randomly Selected Sample of the Industry

    The OTS has considered computing the MMCOF using a randomly 
selected sample of OTS-regulated institutions. This might include the 
OTS-regulated, BIF-insured institutions discussed above in section 
II.A. The OTS estimates that a random sample of 476 institutions would 
provide a confidence level of 95% that the MMCOF will be within +(-) 
.05% (5 basis points) of the sample value. By rotating the institutions 
comprising the statistical sample every twelve to eighteen months, the 
regulatory reporting burden on the industry as a whole would be 
reduced. The agency resources involved in calculating the MMCOF would 
also likely be reduced.

B. Expand the Number of Reporting Institutions 
    One of the past benefits of the MMCOF as an index for adjusting 
ARMs was that it represented a broad, national cross-section of the 
thrift industry. If the MMCOF is no longer widely used because it is 
calculated from a decreasing number of OTS-regulated SAIF-insured 
thrifts, one alternative that might increase its utility is to expand 
the number of reporting institutions. While OTS does not have the 
authority to require depository institutions that it does not regulate, 
such as commercial banks to provide information on their cost of funds 
to the OTS, it could expand the MMCOF to include OTS-regulated, Bank 
Insurance Fund (BIF)-insured institutions. This would add 16 
institutions currently not included in the reporting universe. These 
institutions control between 9 and 10% of the industry assets.
    Expanding the number of reporting institutions to include the OTS-
regulated BIF-insured institutions would only slightly affect the MMCOF 
index, however, because it is a median index2 and not a weighted 
average.3 The additional burden such an expansion would cause OTS 
would also be slight.
---------------------------------------------------------------------------

    \2\ The median cost of funds is defined as the middle value (the 
point that has an equal number of observations above and below it) 
of a set of institutions' individual cost of funds ratios.
    \3\ The weighted mean cost of funds ratio is defined as the sum 
of quarterly interest expense paid or accrued on deposits and 
borrowings, excluding escrow accounts, divided by the average 
(current and previous cycle) sum of balances in deposits and 
borrowings (both Federal Home Loan Bank advances and other 
borrowings, excluding escrow accounts). 
III. Potential Substitute Indices if OTS Stopped Publishing the 
---------------------------------------------------------------------------
MMCOF 
    OTS reviewed three potential substitute indices to determine if any 
would be acceptable substitutes under the requirements of section 
402(e)(3) of FIRREA. Two of these indices either use a calculation 
formula substantially similar to the MMCOF or yield results 
substantially similar to the MMCOF: the Eleventh District COFI and the 
QCOFI. These indices are based on data either supplied to the OTS or 
that the OTS requires thrifts to supply to their Federal Home Loan 
Banks on request.
    The index most frequently used to adjust ARMs, One Year Treasury 
Bills (Constant Maturity), was also reviewed as a possible acceptable 
substitute. OTS believes, however, that this index does not satisfy the 
statutory requirements for substitute indices because it is neither 
technically similar to the MMCOF index, nor is its movement 
substantially similar. Exhibit 1 shows the movement of the MMCOF 
compared to the One Year Treasury Bill index. The two indices' only 
similarity is that both are released monthly.

A. Eleventh District Cost of Funds Index

    The Federal Home Loan Bank of San Francisco's ``Eleventh District 
Cost of Funds'' index is calculated from data submitted from all OTS-
regulated SAIF-insured savings institutions located in Arizona, 
California, and Nevada.4 In March, 1994, these institutions 
represented 6.4 percent of the total number and approximately 35 
percent of the total assets of OTS-regulated institutions. The Eleventh 
District COFI and the MMCOF are substantially similar in that the 
formula used in the Eleventh District COFI is based on the same Thrift 
Financial Report lines as are used in the MMCOF. While the Eleventh 
District COFI is a weighted mean and the MMCOF is a median, the 
Eleventh District COFI closely tracks the MMCOF. For example, from mid-
1991 through March 1994, the monthly results were substantially 
similar, particularly over the last two years. Exhibit 2 shows the 
movement of the Eleventh District COFI as compared to the MMCOF from 
1980 through March 1994.
---------------------------------------------------------------------------

    \4\ 12 CFR 563.180(e) requires savings associations to supply 
this data to their respective Federal Home Loan Banks upon request.
---------------------------------------------------------------------------

B. OTS Quarterly Average Cost of Funds Index

    The QCOFI, published by the OTS, closely tracks the MMCOF. The 
QCOFI is published less frequently than the MMCOF. Thus, it would only 
be available to savings associations on a quarterly basis. The only 
distinction between the two indices is that the MMCOF is a median while 
the QCOFI is a mean (weighted average). However, as shown in Exhibit 3, 
the trend lines between the MMCOF and the QCOFI are quite close.
    Because the QCOFI is a weighted average, the addition of the OTS-
regulated, BIF members to the reporting universe would have a larger 
impact on the QCOFI than on a median index such as the MMCOF. The QCOFI 
would have declined if BIF-insured institutions had been included in 
the calculation (see Exhibit 4). Such a decrease would lower the yield 
on mortgages tied to the national index. In addition, indices 
calculated for states and/or regions that contain BIF institutions 
could be affected. One possible alternative would be for OTS to 
calculate and publish a Quarterly Median Cost of Funds index.

IV. Request for Comment

    OTS invites comment on all aspects of the proposed modifications 
to, or elimination of publication of, the MMCOF. Comments should focus 
on the impact of the proposal on lending institutions in adjusting the 
interest rate on their ARMs. Specific comments should address any 
difficulties savings associations encounter in preparing the monthly 
submission of data to the OTS; any potential benefits of continuing 
publication of the MMCOF index with modifications; and the suitability 
of the two substitute indices proposed by the OTS. Comments from 
savings associations might also address the potential savings resulting 
from the proposed reduction in the reporting burden.

    Dated: October 11, 1994.

    By the Office of Thrift Supervision.
Jonathan L. Fiechter,
Acting Director.
BILLING CODE 6720-01-P

TN17OC94.000


TN17OC94.001


TN17OC94.002


TN17OC94.003


[FR Doc. 94-25630 Filed 10-14-94; 8:45 am]
BILLING CODE 6720-01-C