Federal Agricultural Mortgage Corporation Funding and Fiscal Affairs; Federal Agricultural Mortgage Corporation Disclosure and Reporting Requirements; Risk-Based Capital Requirements, 77247-77262 [E6-21831]

Agencies

• Farm Credit Administration

[Federal Register Volume 71, Number 247 (Tuesday, December 26, 2006)]
[Rules and Regulations]
[Pages 77247-77262]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: E6-21831]


=======================================================================
-----------------------------------------------------------------------

FARM CREDIT ADMINISTRATION

12 CFR Parts 652 and 655

RIN 3052-AC17


Federal Agricultural Mortgage Corporation Funding and Fiscal 
Affairs; Federal Agricultural Mortgage Corporation Disclosure and 
Reporting Requirements; Risk-Based Capital Requirements

AGENCY: Farm Credit Administration.

ACTION: Final rule.

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

SUMMARY: The Farm Credit Administration (FCA, Agency, we) is amending 
regulations governing the Federal Agricultural Mortgage Corporation 
(Farmer Mac or the Corporation) risk-based capital stress test (RBCST 
or model). We are making these amendments in response to changing 
financial markets, new business practices and the evolution of the loan 
portfolio at Farmer Mac, as well as continued development of industry 
best practices among leading financial institutions. The rule modifies 
regulations in 12 CFR part 652, subpart B. The rule is intended to more 
accurately reflect risk in the model in order to improve the model's 
output--Farmer Mac's regulatory minimum risk-based capital level. The 
rule also clarifies Farmer Mac's reporting requirements in Sec.  
655.50(c).

DATES: Effective Date: This regulation will be effective the later of 
30 days after publication in the Federal Register during which time 
either or both Houses of Congress are in session, or March 31, 2007. We 
will publish a notice of the effective date in the Federal Register.

FOR FURTHER INFORMATION CONTACT: 

Joseph T. Connor, Associate Director for Policy and Analysis, Office of 
Secondary Market Oversight, Farm Credit Administration, McLean, VA 
22102-5090, (703) 883-4280, TTY (703) 883-4434;

 or

Rebecca S. Orlich, Senior Counsel, Office of the General Counsel, Farm 
Credit Administration, McLean, VA 22102-5090, (703) 883-4020, TTY (703) 
883-4020.

SUPPLEMENTARY INFORMATION: 

I. Purpose

    The purpose of this rule is to revise the risk-based capital (RBC) 
regulations that apply to Farmer Mac. Our proposed rule was published 
in the Federal Register on November 17, 2005.\1\ The final rule makes 
the following changes to the RBCST:
---------------------------------------------------------------------------

    \1\ 79 FR 69692. See the preamble to our proposed rule for a 
full discussion of our proposed changes.
---------------------------------------------------------------------------

    1. Establishes specific proxy values for loans with missing or 
anomalous or ambiguous data. In the final rule, the Debt-to-Assets 
ratio (DA) proxy value is 0.50, the Loan-to-Value ratio (LTV) remains 
at 0.70, and the Debt Service Coverage ratio (DSC) is 1.25.
    2. Requires the application of known data on Long-term Standby 
Purchase Commitment (Standby) loans in the model.
    3. Revises the estimate of future years' miscellaneous income to 
the annualized 3-year weighted average of the most recent quarterly 
miscellaneous income rate as a fraction of the current quarter's sum of 
cash, investments, guaranteed securities, and loans held for 
investment.
    4. Revises the treatment of gain on sale of agricultural mortgage-
backed securities (AMBS) by applying the 3-year gain rate factor to the 
most recent 4 quarters of AMBS sales.
    5. Revises the method used to estimate operating expenses to a 
moving-average of operating expenses as a percent of non-program assets 
and on- and off-balance sheet program investments.
    The proposed rule also included provisions related to improved 
estimates of the carrying costs of troubled loans by revising 
assumptions regarding Loan Loss Resolution Timing (LLRT), and related 
to adding a component to reflect counterparty risk. These two items are 
not included in the final rule. The Agency plans to address these 
issues in a future rulemaking.
    In developing this rule, we considered the comments and 
recommendations pertaining to the RBCST in the Government 
Accountability Office (GAO) report entitled, ``Farmer Mac: Some 
Progress Made, but Greater Attention to Risk Management, Mission, and 
Corporate Governance is Needed.'' \2\ We also met with Farmer Mac 
representatives on several occasions prior to the development of the 
proposed rule and discussed possible Agency revisions to the RBCST.
---------------------------------------------------------------------------

    \2\ United States General Accounting Office, Farmer Mac: Some 
Progress Made, but Greater Attention to Risk Management, Mission, 
and Corporate Governance is Needed, GAO-04-116 (2003). At the time 
of the report's publication, the GAO was known as the General 
Accounting Office.
---------------------------------------------------------------------------

II. Background

    Our analysis of the RBCST has identified a need to update the model 
in response to changing financial markets, new business practices and 
the evolution of the loan portfolio at Farmer Mac, as well as continued 
development of industry best practices among leading financial 
institutions. Our goal is to ensure that the RBCST reflects changes in 
the Corporation's business structure and loan portfolio that have 
occurred since the model was originally developed by FCA, while 
complying with the statutory requirements and constraints on the 
model's design.
    Our proposed rule was published in the Federal Register on November 
17, 2005, and provided for a 90-day comment period to end on February 
15, 2006. We later extended and reopened the comment period, which 
ended on May 17, 2006.\3\
---------------------------------------------------------------------------

    \3\ In response to requests from commenters, we extended the 
original comment period to April 17, 2006 (71 FR 7446, Feb. 13, 
2006), and subsequently reopened the comment period until May 17, 
2006 (71 FR 24613, Apr. 26, 2006).
---------------------------------------------------------------------------

III. Comments

    We received seven comment letters on the proposed rule from the 
following: Farmer Mac, the Farm Credit Bank of Texas (FCBT), AgFirst 
Farm Credit Bank (AgFirst), U.S. AgBank FCB (U.S. AgBank), Sacramento 
Valley Farm Credit (Sac Valley), First Dakota National Bank (Dakota 
Mac), and AgStar

[[Page 77248]]

Financial Services, ACA (AgStar).\4\ In general, the commenters agreed 
with FCA's objective to revise the RBCST to reflect Farmer Mac's actual 
business risks more accurately but asserted that our proposal would not 
achieve that objective. The commenters contended that the proposed 
changes would result in a risk-based capital requirement that is higher 
than it should be and would drive up the cost of doing business with 
Farmer Mac. Specific comments were primarily focused on two changes: 
(1) The proposed data proxy values for loans with missing data; and (2) 
the method of implementing the carrying cost of nonperforming loans. 
The latter provision is not included in this final rule.
---------------------------------------------------------------------------

    \4\ All of the commenters except Dakota Mac are Farm Credit 
institutions.
---------------------------------------------------------------------------

IV. Summary of the Provisions of the Final Rule and FCA's Responses to 
Comments

    We begin by summarizing and responding to general comments on the 
proposed rule and then provide a summary of specific comments on the 
proposed rule and FCA's responses to the comments.

A. General Comments

    FCBT stated that its chief concern was that certain proposed 
changes appear to have been selected primarily for the purpose of 
increasing the risk-based capital requirement. FCBT and each of the 
other commenters criticized the proposed rule as not being based on 
Farmer Mac's actual underwriting practices and loss experience.
    U.S. AgBank called the proposed regulation overly prescriptive and 
stated that it would be better for FCA to direct Farmer Mac to create 
an RBCST calculation process that complies with the statute, than to 
continue the FCA-designed risk-based capital model. U.S. AgBank also 
stressed the importance of a model that is statistically valid and not 
biased toward overly conservative assumptions, thereby avoiding 
artificial results that could result in unintended consequences. It 
asserted that such consequences could include the compromising of sound 
governance practices at Farmer Mac and of management's accountability 
to its shareholders. Finally, U.S. AgBank said that the model is too 
inflexible given the dynamic nature of agricultural finance and Farmer 
Mac's lines of business that include unique risk factors such as part-
time farm loans.
    Sac Valley, FCBT, and AgFirst also provided their general support 
for the comments submitted by Farmer Mac. Sac Valley stated its 
concurrence with FCA's objective of estimating risk-based capital in a 
way that reflects the risks of Farmer Mac's business and incorporates 
as much as possible best business practices.

B. Proxy Data Values for Loans With Missing or Anomalous Loan 
Origination Data and for Standby Loans--Appendix A, Section 4.1 d.

1. FCA's Proposal
    As noted in the preamble to the proposed rule, the RBCST model was 
designed to use loan origination data--specifically the loan amount, 
DA, LTV, and DSC to estimate the lifetime probability of default on the 
loans, which is then seasoned to reflect the current age of the loan. 
At the time the model was designed, Farmer Mac had complete origination 
data for most loans in its portfolio. In 1998, it had complete 
origination data on approximately 88 percent of Cash Window loans, 
excluding pre-1996 loans. For the remaining loans, state-level average 
loss rates estimated from the loans with complete data were applied to 
loans where data were missing.
    Today, a significant proportion of Farmer Mac's current portfolio 
has incomplete or anomalous loan origination data, or has data that are 
not used in the model. Some data are missing because Farmer Mac has 
several programs whose underwriting standards do not require the 
collection of such data. These programs include part-time farm, 
seasoned and fast-track loans. In addition, the model treats unseasoned 
Standby loans for which loan origination data are available as if the 
loan data were missing. This means that, as of June 30, 2006, complete 
loan origination data were available, and used in the RBCST, on well 
under half of Farmer Mac's loan portfolio, excluding pre-1996 loans. We 
proposed to revise this part of the model to replace the application of 
state-level loss estimates with the application of specified proxy 
values to all loans with missing or anomalous data, and to use known 
data for unseasoned Standby loans when such data are known. The proxy 
values we proposed were a DA ratio of 0.60, an LTV ratio of 0.70, and a 
DSC ratio of 1.20. As we explained in the preamble to the proposed 
rule, we chose conservative proxy values directly related to Farmer 
Mac's underwriting standards on the ground that using conservative 
proxy data best preserves the theoretical and structural integrity of 
the RBCST.
2. Comments
    Farmer Mac agreed that the use of proxy values could be appropriate 
in these circumstances. It asserted, however, that the proposed proxy 
values are flawed because they are ``inconsistent with Farmer Mac's 
underwriting standards for the vast majority of full-time farm loans, 
as well as with Farmer Mac's own risk exposure in actual practice''; 
that they are ``arbitrary, unsupported by any reasoned methodology, and 
based on'' an incorrect interpretation of the Act; that they are 
``unacceptable because they do not correlate strongly, or even 
adequately, to Farmer Mac's actual core business and underwriting 
standards''; and that it knows of no requirement in the Act that the 
loans ``should, unto themselves, represent a worst-case scenario for 
the abuse of Farmer Mac underwriting discretion.'' Farmer Mac asserted 
that, ``if there is available information that would more closely 
approximate Farmer Mac's actual book of business, it should be 
utilized, as opposed to unrelated conservative proxy values.'' Farmer 
Mac raised a concern that the proposed proxy values ``likely will'' 
distort or misrepresent the risks of its business and ``create 
unintended incentives for or against particular classes of loans.''
    Farmer Mac recommended that the proxy values be based instead on 
its historical loan data, using a statistical process for the 
imputation of missing data, or alternatively selecting cutoff 
percentiles. Farmer Mac described possible methodologies and stated its 
view that there was no reason to depart from the model's current 
method, which it characterized as most similar to treatment of data 
that are ``Missing Completely At Random,'' absent ``evidence that [the 
current method] is untenable.'' Farmer Mac also contended that two of 
the proxy values, DA and DSC, are not relevant to its underwriting 
standards for part-time farm loans and offered to work with FCA to 
develop an appropriate RBCST submodel for those loans.
    The other commenters submitted comments that were very much in line 
with Farmer Mac's. They asserted that:
     The proxy values appear arbitrary and not supported in the 
preamble to the proposed rule by any defined methodology or evidence;
     The proxy values are not representative of the commenters' 
loan experience with Farmer Mac or with Farmer Mac's portfolio (as 
understood by the commenters) and are not representative of Farmer 
Mac's underwriting standards; and

[[Page 77249]]

     The proposal, by requiring Farmer Mac to hold capital in 
excess of actual risk, could cause Farmer Mac to increase fees and 
could harm the secondary agricultural loan market and the commenters' 
business with Farmer Mac.
    The commenters recommended basing the proxy values on Farmer Mac 
historical loan data and using a ``well defined methodology'' to 
determine the values. In the section below, we address specific 
comments of Farmer Mac and other commenters.
3. Final Rule
    In the final rule, we establish proxy values for the DA, LTV, and 
DSC ratios that are related to Farmer Mac's underwriting standards, but 
we have moderated them somewhat from the proposed rule. In the final 
rule, the DA ratio proxy value is 0.50, down from 0.60 in the proposed 
rule; the LTV ratio remains at 0.70, the same value as in the proposed 
rule; and the DSC ratio is 1.25, up from 1.20 in the proposed rule. 
Upon further review and consideration of the ranges of Farmer Mac's 
underwriting standards and the relative proportions of the various loan 
types in the portfolio, we have decided that these values are more 
appropriate to the underwriting standards for the loan types that make 
up the preponderance of Farmer Mac's portfolio. In our judgment, these 
proxy values are appropriate for application to loan programs that have 
different underwriting standards but account for a smaller proportion 
of the portfolio. We believe these values are still sufficiently 
conservative to maintain the theoretical integrity of the model while 
avoiding unintended consequences related to inappropriate incentives to 
underwrite more aggressively in reduced-documentation loan programs. We 
note that, if the relative proportions of various loan types with 
differing underwriting standards change over time, the Agency may 
consider further adjustment to the proxy values.
    We disagree with many of the comments we received. To begin with, 
we do not believe our proposal is based on an incorrect interpretation 
of the Act or that it imposes ``worst-case'' proxy values. The Act 
provides FCA with significant discretion in establishing the RBCST. 
Section 8.32 of the Act states that the FCA (through the Director of 
the Office of Secondary Market Oversight (OSMO)) must, among other 
things, ``take into account appropriate distinctions based on various 
types of agricultural mortgage products, varying terms of Treasury 
obligations, and any other factors the Director considers appropriate * 
* *.'' \5\ The model uses, and will continue to use, Farmer Mac's 
``actual book of business'' as represented by actual data. The 
incompleteness or non-use of loan origination data for what is now a 
significant portion of Farmer Mac's portfolio is an important factor in 
evaluating the reliability of the RBCST output. The Agency must decide 
how best to treat the loans whose data are not in the model. We believe 
the model's current treatment is no longer adequate to represent loan 
risk for such a large portion of the portfolio. Our choice of 
conservative proxy values takes into consideration not only the role of 
the FCA to provide for the general supervision of the safe and sound 
performance of Farmer Mac under section 8.11(a) of the Act, but also 
Farmer Mac's actual loan data and practices. We do not believe the 
proxy values represent a ``worst-case scenario.'' In setting the proxy 
values, we considered Farmer Mac's actual practice of accepting loans 
with ratios that are riskier than those permitted under its 
underwriting standards when the loan has compensating strengths in 
other ratios or risk indicators. A ``worst-case'' approach would have 
yielded proxy values much higher than the proposed DA and LTV and much 
lower than the DSC.
---------------------------------------------------------------------------

    \5\ 12 U.S.C. 2279bb-1(b)(1)(A).
---------------------------------------------------------------------------

    We considered Farmer Mac's suggestion to substitute values based on 
Farmer Mac's historical loan data for the missing data.\6\ In our 
judgment, the historical data are not necessarily representative of the 
portion of the portfolio that is missing data, and they are not 
necessarily representative of Farmer Mac's future underwriting 
practices on loans for which they will not collect complete data. We do 
not agree with the underlying assumption of the comments that the 
historical loan data on full-time farm loans would correlate strongly 
with the loans missing data. In a circumstance where we cannot know the 
predictive value of the historical loan data, we do not agree that a 
valid statistical methodology is available to set the proxy levels. 
Moreover, as we have noted, the loans for which loan origination data 
are complete now represent a much smaller proportion of Farmer Mac's 
loan portfolio. Therefore, we concluded that using the historical data 
is not the best means to determine appropriate proxy values.
---------------------------------------------------------------------------

    \6\ We note that the current version of the RBC model, through 
its application of average loss rates by state to loans with missing 
data, is similar to the approach recommended in the comment. The 
insufficiency of this approach and the significant proportion of 
loans that have incomplete data are, in fact, the conditions that 
prompted the development of this revision to the RBCST.
---------------------------------------------------------------------------

    A statistical approach suggested by Farmer Mac is the SAS Proc MI 
multiple imputation (MI) procedure for developing consistent estimates 
of confidence limits around the mean of each individual underwriting 
variable for loans for which data existed. The implication is that 
these estimates would be appropriate for use as proxies in cases where 
the underwriting data were absent in individual loans. The specific 
process demonstrated assumes that the underwriting data are 
multivariate normal and that the missing elements may depend on the 
remaining observed variables, but not on their own values. The 
resampling method generates consistent estimates of the variances from 
which confidence limits around the statistics can be constructed.
    The MI methods cited in the comment are most often used in efforts 
to avoid deletion of observations in data sets with partially missing 
data, but for which portions of the covariate data sets exist.\7\ The 
use of no additional independent variables (e.g., age, loan size) which 
are observable across loans both with and without underwriting data 
implies: (i) No conditioning on additional variables was considered, 
and (ii) an assumption of equivalent distributions between those 
missing data and those not missing data.
---------------------------------------------------------------------------

    \7\ For example, if multiple health factors/indicators 
individually contribute to incidence rates of a serious health 
problem, but not all variables are observed or collected on all 
individuals, MI procedures allow the use of the data with incomplete 
measures across the independent variables rather than excluding 
entire observations that are missing only portions of their 
independent variables.
---------------------------------------------------------------------------

    We do not agree that the Farmer Mac-suggested method provides a 
reasonable method for identifying candidate proxy values for numerous 
reasons. First, we did not intend that the proxies represent mean 
values of the underwriting data in cases where the data exist, or as 
conditioned by the pattern of missing data from cases missing only a 
portion of the underwriting variables. We do not believe that this 
approach is reasonable given the stark differences in other 
characteristics of the subset of loans that do and do not have complete 
underwriting data. To emphasize this point, Table 1 is provided which 
summarizes key attributes of the loans grouped by whether the loan 
observation received at least one proxy value due to missing or 
anomalous data.

[[Page 77250]]



                                 Table 1
------------------------------------------------------------------------
                                     No proxy    At least 1
               Data                    data        proxy       Combined
------------------------------------------------------------------------
Average Age (in years)...........         5.83        11.83         9.16
Average Current Balance..........  $433,568     $164,542     $284,199
Average Original Balance.........  $570,119     $267,039     $401,842
Number of Loans..................     7,269        9,074       16,343
------------------------------------------------------------------------

    As shown in the table, the loans missing data are considerably 
older (rendering the cases where a portion of the underwriting data 
does exist to be less likely to be reliable), have much smaller 
original balances, and have correspondingly lower current balances. 
Standard tests of the equivalence of means strongly reject the 
hypothesis of equivalence of the means between the two groups of loans 
by age, original size, or current balance (p-value = 0.0000, all 
cases).
    As an alternative to using an imputation methodology, Farmer Mac 
also suggested that percentile cutoffs of actual ratios in its 
portfolio of unseasoned standard full-time farm loans should be 
considered as an acceptable method to derive proxies, though less 
appropriate (in their view) than imputation of mean values. Farmer Mac 
asserted that the proposed proxy levels are statistical outliers.\8\ In 
general, we have the same concern here as with the multiple imputation 
approach regarding basing proxy values on historical measurements of a 
potentially uncorrelated portfolio. The appropriateness of using a 
cutoff percentile depends on the congruence in the data between the set 
missing underwriting data and those with data. Moreover, the 
distribution around a given ``consistent'' percentile choice is not 
necessarily comparable across the three underwriting variables (i.e., 
there may be only a small ``distance'' between the 95th percentile and 
the maximum D/A in the available data, while there is a large 
``distance'' from the 95th percentile of the order-adjusted DSC to the 
most undesirable one in the data set).\9\
---------------------------------------------------------------------------

    \8\ The proposed DA proxy equated to the 95th percentile of 
Farmer Mac's portfolio of unseasoned full-time farm loans, the 
proposed LTV proxy equated to some percentile in excess of the 90th 
percentile, and the proposed DSC proxy equated to some percentile in 
excess of the 5th percentile (or, for greater ease of comparison, 
its inverse--the 95th percentile). In the final rule, the proxy 
values would equate to the 91st, 90th, and 9th percentiles 
respectively, as of June 30, 2006. Although we did not base our 
proxy values on the percentile cutoffs, we believe the relationships 
of those values to the percentile cutoffs is appropriate.
    \9\ Farmer Mac contends that the proposed proxies represent 
outliers in the data. However, we note that its comment includes a 
table showing that the proxy values indicated by the 95th 
percentiles in the set of unseasoned Full-time Farm loans all exceed 
Farmer Mac's underwriting limits for such loans. Thus, the proxy 
values would not appear to be unreasonably conservative.
---------------------------------------------------------------------------

    We would also note that average loss rates generated by the RBCST's 
Credit Loss Module (CLM) are not especially sensitive to the level of 
the proxy values. To illustrate this point, we provide the following 
data tables. Table 2A sets forth the average loss rates generated by 
the CLM as of June 30, 2006, under various LTV and DA proxy value 
combinations, keeping DSC constant at 1.25. The table indicates that 
the average loss rate across all combinations presented varies within a 
range of 27 basis points. Under the final rule's proxy values (0.50, 
0.70, and 1.25, for DA, LTV, and DSC proxies, respectively) the table 
shows that at June 30, 2006 the average loss rate would have been 3.782 
percent.

                                                    Table 2A
----------------------------------------------------------------------------------------------------------------
 
------------------------------------------------------------------------------------------------------
       DSC proxy = 1.25                                       LTV proxies
----------------------------------------------------------------------------------------------------------------
DA Proxies....................  ..........     0.60        0.65        0.70        0.75        0.80
                                   0.45        3.694%      3.706%      3.724%      3.748%      3.783%
                                   0.50        3.751%      3.764%      3.782%      3.807%      3.842%
                                   0.55        3.811%      3.824%      3.843%      3.869%      3.905%
                                   0.60        3.874%      3.888%      3.907%      3.934%      3.966%
----------------------------------------------------------------------------------------------------------------

    Table 2B presents the calculated average loss rate across 
combinations of DCS and DA ratios, holding LTV constant. Under these 
combinations, the average loss rate varies within a range of 16 basis 
points.

                                                    Table 2B
----------------------------------------------------------------------------------------------------------------
 
----------------------------------------------------------------------------------------------------------------
            LTV proxy = 0.70                                            DSC proxies
----------------------------------------------------------------------------------------------------------------
DA Proxies..............................  ..........     1.15        1.20        1.25        1.30        1.35
                                             0.45        3.736%      3.730%      3.724%      3.718%      3.712%
                                             0.50        3.794%      3.788%      3.782%      3.775%      3.769%
                                             0.55        3.856%      3.849%      3.843%      3.836%      3.830%
                                             0.60        3.921%      3.914%      3.907%      3.900%      3.894%
----------------------------------------------------------------------------------------------------------------

    Table 2C presents the variation in the calculated average loss rate 
across combinations of DCS and LTV ratios, holding DA constant at the 
level in the final rule. Under these combinations, the average loss 
rate varies within a range of just 3 basis points.

[[Page 77251]]



                                                    Table 2C
----------------------------------------------------------------------------------------------------------------
 
----------------------------------------------------------------------------------------------------------------
             DA proxy = 0.50                                            DSC proxies
----------------------------------------------------------------------------------------------------------------
LTV proxies.............................  ..........     1.15        1.20        1.25        1.30        1.35
                                             0.60        3.763%      3.757%      3.751%      3.745%      3.739%
                                             0.65        3.776%      3.770%      3.764%      3.757%      3.751%
                                             0.70        3.794%      3.788%      3.782%      3.775%      3.769%
                                             0.75        3.820%      3.814%      3.807%      3.801%      3.795%
----------------------------------------------------------------------------------------------------------------

    Rather than focusing on the distribution of underwriting ratios in 
the existing loan data sets through time, we instead chose proxy values 
that are near the conservative limits of the range of values that are 
acceptable to Farmer Mac under its underwriting standards for different 
types of loans (including, but not limited to, full-time farm loans). 
In addition, we took into consideration that Farmer Mac can accept 
underwriting ratios that exceed the stated ranges of its underwriting 
standards.\10\ We intended that the proxy values be sufficiently 
conservative to avoid underestimating the risk in the portfolio, but 
not at the extremes of Farmer Mac's underwriting standards. This 
approach recognizes that Farmer Mac would be unlikely to underwrite 
loans at its underwriting limits in each ratio category. These values 
are acceptable to Farmer Mac for underwriting purposes, as demonstrated 
by both its policies and its practices. Therefore, we believe that the 
proxy values are realistic as well as conservative and reflect Farmer 
Mac's actual business practices.
---------------------------------------------------------------------------

    \10\ We note Farmer Mac's actual practice of accepting loans 
with ratios that are outside of the ranges, as permitted under its 
underwriting standards when the loan has compensating strengths in 
other ratios or risk indicators.
---------------------------------------------------------------------------

    Farmer Mac's comment that the proposed proxy values ``likely will'' 
distort or misrepresent the risks of its business, as well as create 
unintended incentives for or against particular classes of loans, did 
not make clear exactly what unintended incentives or what classes of 
loans Farmer Mac had in mind. We would agree that, on an individual 
loan basis, using proxy data will ``misrepresent'' the loan to the 
extent that the proxy values understate or overstate the level of 
actual risk in the loan. The problem of the likely inexactitude in the 
calculation is necessitated by, and a direct result of, the uncertainty 
created by the missing data. This uncertainty is itself one component 
of the risk in Farmer Mac's loan portfolio. We believe that applying 
conservative proxy values is a way to consider adequately the actual 
risk in the loan as well as the added risk associated with this 
uncertainty. With respect to unintended incentives, it is true that, 
however we decide to treat loans with missing data--for example, if we 
were to apply proxy values based on historical loan data or related to 
underwriting standards, or even if we entirely removed loans with 
missing data from the model--we could create incentives for Farmer Mac 
and its business partners to expand or contract one or more lines of 
business or to modify program requirements. Indeed, in the model's 
current treatment of loans with missing data, one could argue that 
using state-level loss estimates may have been a disincentive for 
Farmer Mac to collect loan origination data in some cases. We believe 
that the proxy values in the final rule will minimize any potential 
incentive not to collect loan origination data on the great majority of 
loans, without providing inappropriate incentives to continue or 
terminate worthy and needed loan products.
    As we described above, Farmer Mac offered to work with FCA to 
develop an appropriate RBCST submodel for part-time farm loans since, 
in Farmer Mac's stated view, the DA and DSC ratios are ``not relevant'' 
to its underwriting standards for such loans.\11\ FCA weighed the added 
complexity of a submodel against potential benefits in improved 
accuracy of the RBC model's output, as well as the potential 
disincentive that might be created to underwrite part-time farm 
business in the absence of such a submodel. By our calculation of 
Farmer Mac-submitted data, the part-time farm loan volume is a very 
small percentage of the total modeled portfolio as of June 30, 2006. We 
do not consider this amount to be substantial and, therefore, do not 
see a compelling reason to add complexity to the model by adding a 
submodel at this time. We could consider a submodel in the future if 
the Corporation's part-time farm loan volume grows. We believe that the 
selected proxy data values appropriately balance the risk of a 
disincentive to underwrite part-time loans with the risk of an 
inappropriate incentive to underwrite more loans of this type with risk 
characteristics that exceed those of the proxy values.
---------------------------------------------------------------------------

    \11\ Notwithstanding Farmer Mac's assertion that the DA and DSC 
ratios are not relevant, not all part-time farm loans are missing 
those data in Farmer Mac's submission of the RBCST as of June 30, 
2006.
---------------------------------------------------------------------------

    AgStar commented specifically that the proxy values would reflect 
an especially unrealistic risk estimate on seasoned loans. We disagree 
with the comment because the model's loan seasoning adjustment occurs 
after loss rates are estimated. Therefore, the risk in seasoned loans 
in Farmer Mac's portfolio would continue to be adjusted downward in 
accordance with Section 2.2 of Appendix A. We expect the impact of the 
seasoning adjustment to be similar in magnitude in the revised RBCST 
model regardless of whether the proxy values are applied. The reason is 
that the model recognizes substantial risk mitigation through its 
seasoning adjustment component. However, we note that when a loan's 
origination date is among the missing data, and therefore age is not 
determinable, the final rule will substitute the ``cut off'' date for 
the origination date. In such cases, if a loan were several years old 
and only recently taken into Farmer Mac's portfolio, the risk-
mitigation of its true age could not be recognized. We believe our 
approach recognizes the risk created when a loan's origination date is 
not collected in a low-documentation loan program.
    AgStar also noted that recent unseasoned loans placed in the 
Standby program are better quality than the proxy values would 
estimate. While AgStar may have good information to substantiate this 
claim, if these loan records do not contain that information, the 
Agency must address the resulting uncertainty (i.e., risk). If a 
primary lender consistently has such information on Standby loans, it 
could benefit from including these data in the loan data submitted 
under the Standby program regardless of whether such data are required 
under the Standby program.

C. Calculation of Miscellaneous Income and Gain on Sale of AMBS--
Appendix A, Section 4.2(3)

    Farmer Mac commented that more accurate moving average calculations 
of miscellaneous income and gain on sale of AMBS would be achieved by 
first

[[Page 77252]]

calculating individual ratios, annualizing the ratios and then 
computing the moving average over the appropriate time horizon. We do 
not agree that Farmer Mac's suggested approach would be more accurate. 
Our approach provides a volume-weighted measure of miscellaneous income 
that is more accurate and generally less sensitive to variations in 
asset volumes than the Farmer Mac-suggested approach. Under Farmer 
Mac's suggested approach, each individual observation has the same 
weight regardless of the level of the relevant assets. The weighted 
average approach to AMBS avoids counting each undefined (0/0) ratio as 
an individual observation which would skew the average.
    Similarly, in the case of gain on sale of AMBS, we believe our 
approach to generating the weighted average rate of gain is less 
potentially volatile than the Farmer Mac-suggested approach. Moreover, 
Farmer Mac's suggestion that the calculated amount be annualized would 
be incorrectly applied in this case, regardless of the method adopted, 
because the calculated rate is as applicable and appropriate on an 
annual basis as it is on a quarterly basis. To multiply the calculated 
rate by 4 would overstate the rate of gain.

D. Operating Expense Regression Equation--Appendix A, Section 4.2(3)

    In the RBCST's operating expense regression equation, we proposed a 
change that would remove the dummy variable from the equation and 
include multiple variables to account for different business 
activities.
    Farmer Mac agreed in principle with the extension of the 
independent variables in the regression and the elimination of the 
dummy variable but argued that the intent of the proposed regression 
was to provide marginal impacts of different activities to the 
operating expenses. It observed that the individual coefficient signs 
are not entirely consistent with expected relationships and offered two 
alternative proposals to enable projections of their operating expenses 
to be applied within the model. The first alternative proposed involves 
calculation of a simple average of recent operating expenses applied as 
a constant in the model. They refer to this approach as being analogous 
to that used to estimate MI rates and gains on AMBS rates.
    The second approach offered by Farmer Mac involves a regression 
framework across similar expense categories as proposed by us, but 
expresses these in cost share form. Their proposed approach contains 
similar drawbacks as those Farmer Mac raised regarding FCA's proposed 
approach and suffers specific problems in expressing logarithms of 
values which may be zero at times.
    In light of the recent evolution of their cost structures and 
changing relative scales of their program activities, we agree with the 
comment that an approach to accurately reflect their cost structures 
can be obtained from recent data and applied forward within the 
existing constructs of the model. Farmer Mac proposes the use of 
average expenses to reflect future experiences. We note that in periods 
of increasing costs, the recent average will have a negative bias, and 
during periods of decreasing costs, that there will be a positive bias. 
We accept the moving average application of expenses and agree that it 
is consistent with the spirit of the calculations of the rates for 
miscellaneous income and gains on sales of AMBSs. In specific 
application, we require that the operating expense rate be calculated 
as the average of operating expense rates calculated as the annualized 
expenses as shares of the sum of on-balance sheet assets and off-
balance sheet program activities over the most recent 4 quarters 
inclusive of the current submission date. This average rate is applied 
to the current quarter's on-balance sheet assets and off-balance sheet 
program activities. That share will then be applied forward to the 
balances of the same categories throughout the 10-year period of the 
RBCST model.

E. Change to Disclosure Regulations

    We proposed to clarify Sec.  655.50(c) to state that Farmer Mac 
must provide FCA with copies of its substantive correspondence with the 
Securities and Exchange Commission (SEC). We received no comments on 
this proposal and adopt it without change in the final rule.

V. Issues Not Addressed in Final Rule

A. Carrying Costs of Troubled Loans--Appendix A, Section 4.2(3)

    We proposed to improve estimates of carrying costs of troubled 
loans by revising the Loan Loss Resolution Timing to reflect that 
problem loans may take longer than the 1 year assumed in the existing 
model's loss-severity rate. Farmer Mac commented that it agreed with 
aspects of the proposed change but had concerns about some of the 
modifications, as well as the validity of certain assumptions we made.
    The Agency has elected to address this revision in a future 
rulemaking out of a desire to review further the scaling factor applied 
to loan loss volume in order to estimate the amount of associated 
unpaid principal balance, and to review any new information that may be 
available from Farmer Mac regarding its actual loan resolution timing. 
The proposed scaling factor is derived from the average principal 
amortization of loans in the current portfolio and would be 
recalculated on a quarterly basis. While we received no comments on the 
scaling factor, we believe that the principal amortization of actual 
nonperforming loans at Farmer Mac might provide an opportunity to 
improve the estimate of unpaid principal balance associated with 
nonperforming loans during the LLRT period.

B. Spreadsheet Linkage for Funding Off-Balance Sheet Loans

    This comment from Farmer Mac deals with a component of the revision 
dealing with the carrying cost of nonperforming loans. Because the 
Agency has elected to address this revision in a future rulemaking for 
reasons explained in ``A'' above, we do not address this comment here.

C. Adding a Component To Reflect Counterparty Risk--Appendix A, Section 
4.1e.

    The proposed rule's provisions related to the estimation of 
counterparty risk are not included in the final rule and will be 
addressed by the Agency in a future rulemaking. Specifically, while we 
received no comments on the approach to identifying or applying the 
counterparty risk component, we have elected to review the Office of 
Federal Housing Enterprise Oversight (OFHEO) haircut levels, confirm 
the applicability of the OFHEO haircut schedules for application to 
yields rather than individual cash flows, and consider the formal 
development of a calculation tool with fixed-category investment 
instrument definitions.
    In the preamble to the proposed rule, we requested comment on 
potential methods to incorporate three specific risks into the model in 
future proposed regulations. The three risks are: The risk associated 
with the AgVantage portfolio; the risk of a stress-induced increase in 
Farmer Mac's cost of funds; and the counterparty risk associated with 
the derivatives portfolio and specifically the replacement cost of 
defaulted derivative contracts. However, we received no comments on 
these topics.

[[Page 77253]]

VI. Other Comments Received

A. Method of Historical Loss Estimation

    Farmer Mac reiterated comments it made to our first rule 
implementing the RBCST that was published in the Federal Register on 
November 12, 1999. (See 64 FR 61740.) The comments criticize the 
methodology employed to quantify the worst-case historical benchmark 
loss experience, stating that it is unsubstantiated by actual loss 
experience. In this rulemaking, we proposed no changes related to this 
aspect of the RBC model and are, therefore, not adopting Farmer Mac's 
recommended changes in the final rule. However, we note that the 
Agency's position on this issue remains consistent with our response 
that was published in the final rule implementing the RBC model on 
April 12, 2001. (See 66 FR 19048.)

B. Spreadsheet Financial Statement Formats

    In its comment letter, Farmer Mac asked us to update the RBCST's 
Balance Sheet and Income Statement categories. Farmer Mac commented 
that populating financial statement data has become time-consuming for 
its staff due to changes in its SEC reporting formats that are not 
reflected in the RBC model. While we would prefer to make the 
submission preparation process as efficient as possible, we have 
observed that Farmer Mac's financial statements have been changing 
format with relative frequency over recent years. For that reason, we 
hesitate to expend resources to modify the formats in the model if 
these could become outdated relatively soon. However, we agree that 
such updates should be done periodically in order to keep the formats 
reasonably close. For that reason, while we have made no changes to the 
financial statement formats in this rule, we would expect to make such 
changes in consultation with Farmer Mac through the technical change 
process (i.e., without rulemaking).

VII. Technical Changes to the RBCST in the Final Rule

    In section 4.2b(3)(E) of the Appendix, we have deleted specific 
guarantee fee values for post-1996 Farmer Mac I assets, pre-1996 Farmer 
Mac I assets, and Farmer Mac II assets because specific values are not 
applied in the stress test. The stress test applies quarterly updates, 
supplied by Farmer Mac, of the weighted average guarantee rates for 
each category of assets.

VIII. Impact of Final Rule Changes on Required Risk-Based Capital

    The table below provides an indication of the impact of the 
revisions in the quarter ended June 30, 2006. Lines 1 through 4 present 
the impacts if only that revision were made to the current version and 
the column labeled ``Difference'' calculates the impact of that 
individual change for the quarter ended June 30, 2006, compared to the 
minimum requirement calculated using the currently active Version 1.25. 
Line 5 presents the impact of all of the revisions in Version 2.0 (the 
model as revised in this final rule).\12\
---------------------------------------------------------------------------

    \12\ Please note that Farmer Mac announced on October 6, 2006, 
that it intends to restate certain financial results for several 
recent reporting periods. The calculations in the table could change 
based on the restatement.

------------------------------------------------------------------------
  Calculated regulatory minimum capital      6/30/2006      Difference
------------------------------------------------------------------------
RBCST Version 1.25 (calculated as of 6/           67,660  ..............
 30/2006)
RBCST 2.0 Individual Change Impacts:
    (1) CLM Changes: Data Proxies and             93,523          25,862
     Standby Treatment..................
    (2) Miscellaneous Income Treatment..          59,932          -7,728
    (3) Gain on Sale of AMBS............          67,660               0
    (4) Operating Expenses..............          95,297          27,637
    (5) Total RBCST Version 2.0 Impact..         113,431          45,771
------------------------------------------------------------------------

    As shown in the table, implementation of the data proxies and the 
revised operating expense estimation result in the greatest impact on 
the calculated risk-based capital requirements.

IX. Regulatory Flexibility Act

    Pursuant to section 605(b) of the Regulatory Flexibility Act (5 
U.S.C. 601 et seq.), FCA hereby certifies the rule will not have a 
significant economic impact on a substantial number of small entities. 
Farmer Mac has assets and annual income over the amounts that would 
qualify them as small entities. Therefore, Farmer Mac is not considered 
a ``small entity'' as defined in the Regulatory Flexibility Act.

List of Subjects

12 CFR Part 652

    Agriculture, Banks, banking, Capital, Investments, Rural areas.

12 CFR Part 655

    Accounting, Agriculture, Banks, banking, Accounting and reporting 
requirements, Disclosure and reporting requirements, Rural areas.

0
For the reasons stated in the preamble, parts 652 and 655 of chapter 
VI, title 12 of the Code of Federal Regulations are amended as follows:

PART 652--FEDERAL AGRICULTURAL MORTGAGE CORPORATION FUNDING AND 
FISCAL AFFAIRS

0
1. The authority citation for part 652 continues to read as follows:

    Authority: Secs. 4.12, 5.9, 5.17, 8.11, 8.31, 8.32, 8.33, 8.34, 
8.35, 8.36, 8.37, 8.41 of the Farm Credit Act (12 U.S.C. 2183, 2243, 
2252, 2279aa-11, 2279bb, 2279bb-1, 2279bb-2, 2279bb-3, 2279bb-4, 
2279bb-5, 2279bb-6, 2279cc); sec. 514 of Pub. L. 102-552, 106 Stat. 
4102; sec. 118 of Pub. L. 104-105, 110 Stat. 168.

0
2. Revise subpart B to part 652 to read as follows:
Subpart B--Risk-Based Capital Requirements
Sec.
652.50 Definitions.
652.55 General.
652.60 Corporation board guidelines.
652.65 Risk-based capital stress test.
652.70 Risk-based capital level.
652.75 Your responsibility for determining the risk-based capital 
level.
652.80 When you must determine the risk-based capital level.
652.85 When to report the risk-based capital level.
652.90 How to report your risk-based capital determination.
652.95 Failure to meet capital requirements.
652.100 Audit of the risk-based capital stress test.
Appendix A--Subpart B of Part 652--Risk-Based Capital Stress Test

Subpart B--Risk-Based Capital Requirements


Sec.  652.50  Definitions.

    For purposes of this subpart, the following definitions will apply:

[[Page 77254]]

    Farmer Mac, Corporation, you, and your means the Federal 
Agricultural Mortgage Corporation and its affiliates as defined in 
subpart A of this part.
    Our, us, or we means the Farm Credit Administration.
    Regulatory capital means the sum of the following as determined in 
accordance with generally accepted accounting principles:
    (1) The par value of outstanding common stock;
    (2) The par value of outstanding preferred stock;
    (3) Paid-in capital, which is the amount of owner investment in 
Farmer Mac in excess of the par value of stock;
    (4) Retained earnings; and,
    (5) Any allowances for losses on loans and guaranteed securities.
    Risk-based capital means the amount of regulatory capital 
sufficient for Farmer Mac to maintain positive capital during a 10-year 
period of stressful conditions as determined by the risk-based capital 
stress test described in Sec.  652.65.


Sec.  652.55  General.

    You must hold risk-based capital in an amount determined in 
accordance with this subpart.


Sec.  652.60  Corporation board guidelines.

    (a) Your board of directors is responsible for ensuring that you 
maintain total capital at a level that is sufficient to ensure 
continued financial viability and--provide for growth. In addition, 
your capital must be sufficient to meet statutory and regulatory 
requirements.
    (b) No later than 65 days after the beginning of Farmer Mac's 
planning year, your board of directors must adopt an operational and 
strategic business plan for at least the next 3 years. The plan must 
include:
    (1) A mission statement;
    (2) A review of the internal and external factors that are likely 
to affect you during the planning period;
    (3) Measurable goals and objectives;
    (4) Forecasted income, expense, and balance sheet statements for 
each year of the plan; and,
    (5) A capital adequacy plan.
    (c) The capital adequacy plan must include capital targets 
necessary to achieve the minimum, critical and risk-based capital 
standards specified by the Act and this subpart as well as your capital 
adequacy goals. The plan must address any projected dividends, equity 
retirements, or other action that may decrease your capital or its 
components for which minimum amounts are required by this subpart. You 
must specify in your plan the circumstances in which stock or equities 
may be retired. In addition to factors that must be considered in 
meeting the statutory and regulatory capital standards, your board of 
directors must also consider at least the following factors in 
developing the capital adequacy plan:
    (1) Capability of management;
    (2) Strategies and objectives in your business plan;
    (3) Quality of operating policies, procedures, and internal 
controls;
    (4) Quality and quantity of earnings;
    (5) Asset quality and the adequacy of the allowance for losses to 
absorb potential losses in your retained mortgage portfolio, securities 
guaranteed as to principal and interest, commitments to purchase 
mortgages or securities, and other program assets or obligations;
    (6) Sufficiency of liquidity and the quality of investments; and,
    (7) Any other risk-oriented activities, such as funding and 
interest rate risks, contingent and off-balance sheet liabilities, or 
other conditions warranting additional capital.


Sec.  652.65  Risk-based capital stress test.

    You will perform the risk-based capital stress test as described in 
summary form below and as described in detail in Appendix A to this 
subpart. The risk-based capital stress test spreadsheet is also 
available electronically at https://www.fca.gov. The risk-based capital 
stress test has five components:
    (a) Data requirements. You will use the following data to implement 
the risk-based capital stress test.
    (1) You will use Corporation loan-level data to implement the 
credit risk component of the risk-based capital stress test.
    (2) You will use Call Report data as the basis for Corporation data 
over the 10-year stress period supplemented with your interest rate 
risk measurements and tax data.
    (3) You will use other data, including the 10-year Constant 
Maturity Treasury (CMT) rate and the applicable Internal Revenue 
Service corporate income tax schedule, as further described in Appendix 
A to this subpart.
    (b) Credit risk. The credit risk part estimates loan losses during 
a period of sustained economic stress.
    (1) For each loan in the Farmer Mac I portfolio, you will determine 
a default probability by using the logit functions specified in 
Appendix A to this subpart with each of the following variables:
    (i) Borrower's debt-to-asset ratio at loan origination;
    (ii) Loan-to-value ratio at origination, which is the loan amount 
divided by the value of the property;
    (iii) Debt-service-coverage ratio at origination, which is the 
borrower's net income (on- and off-farm) plus depreciation, capital 
lease payments, and interest, less living expenses and income taxes, 
divided by the total term debt payments;
    (iv) The origination loan balance stated in 1997 dollars based on 
the consumer price index; and,
    (v) The worst-case percentage change in farmland values (23.52 
percent).
    (2) You will then calculate the loss rate by multiplying the 
default probability for each loan by the estimated loss-severity rate, 
which is the average loss of the defaulted loans in the data set (20.9 
percent).
    (3) You will calculate losses by multiplying the loss rate by the 
origination loan balances stated in 1997 dollars.
    (4) You will adjust the losses for loan seasoning, based on the 
number of years since loan origination, according to the functions in 
Appendix A to this subpart.
    (5) The losses must be applied in the risk-based capital stress 
test as specified in Appendix A to this subpart.
    (c) Interest rate risk. (1) During the first year of the stress 
period, you will adjust interest rates for two scenarios, an increase 
in rates and a decrease in rates. You must determine your risk-based 
capital level based on whichever scenario would require more capital.
    (2) You will calculate the interest rate stress based on changes to 
the quarterly average of the 10-year CMT. The starting rate is the 3-
month average of the most recent CMT monthly rate series. To calculate 
the change in the starting rate, determine the average yield of the 
preceding 12 monthly 10-year CMT rates. Then increase and decrease the 
starting rate by:
    (i) 50 percent of the 12-month average if the average rate is less 
than 12 percent; or
    (ii) 600 basis points if the 12-month average rate is equal to or 
higher than 12 percent.
    (3) Following the first year of the stress period, interest rates 
remain at the new level for the remainder of the stress period.
    (4) You will apply the interest rate changes scenario as indicated 
in Appendix A to this subpart.
    (5) You may use other interest rate indices in addition to the 10-
year CMT subject to our concurrence, but in no event can your risk-
based capital level be less than that determined by using only the 10-
year CMT.
    (d) Cashflow generator. (1) You must adjust your financial 
statements based on the credit risk inputs and interest rate risk 
inputs described above to

[[Page 77255]]

generate pro forma financial statements for each year of the 10-year 
stress test. The cashflow generator produces these financial 
statements. You may use the cashflow generator spreadsheet that is 
described in Appendix A to this subpart and available electronically at 
https://www.fca.gov. You may also use any reliable cashflow program that 
can develop or produce pro forma financial statements using generally 
accepted accounting principles and widely recognized financial modeling 
methods, subject to our concurrence. You may disaggregate financial 
data to any greater degree than that specified in Appendix A to this 
subpart, subject to our concurrence.
    (2) You must use model assumptions to generate financial statements 
over the 10-year stress period. The major assumption is that cashflows 
generated by the risk-based capital stress test are based on a steady-
state scenario. To implement a steady-state scenario, when on- and off-
balance sheet assets and liabilities amortize or are paid down, you 
must replace them with similar assets and liabilities. Replace 
amortized assets from discontinued loan programs with current loan 
programs. In general, keep assets with small balances in constant 
proportions to key program assets.
    (3) You must simulate annual pro forma balance sheets and income 
statements in the risk-based capital stress test using Farmer Mac's 
starting position, the credit risk and interest rate risk components, 
resulting cashflow outputs, current operating strategies and policies, 
and other inputs as shown in Appendix A to this subpart and the 
electronic spreadsheet available at http: //www.fca.gov.
    (e) Calculation of capital requirement. The calculations that you 
must use to solve for the starting regulatory capital amount are shown 
in Appendix A to this subpart and in the electronic spreadsheet 
available at https://www.fca.gov.


Sec.  652.70  Risk-based capital level.

    The risk-based capital level is the sum of the following amounts:
    (a) Credit and interest rate risk. The amount of risk-based capital 
determined by the risk-based capital test under Sec.  652.65.
    (b) Management and operations risk. Thirty (30) percent of the 
amount of risk-based capital determined by the risk-based capital test 
in Sec.  652.65.


Sec.  652.75  Your responsibility for determining the risk-based 
capital level.

    (a) You must determine your risk-based capital level using the 
procedures in this subpart, Appendix A to this subpart, and any other 
supplemental instructions provided by us. You will report your 
determination to us as prescribed in Sec.  652.90. At any time, 
however, we may determine your risk-based capital level using the 
procedures in Sec.  652.65 and Appendix A to this subpart, and you must 
hold risk-based capital in the amount we determine is appropriate.
    (b) You must at all times comply with the risk-based capital levels 
established by the risk-based capital stress test and must be able to 
determine your risk-based capital level at any time.
    (c) If at any time the risk-based capital level you determine is 
less than the minimum capital requirements set forth in section 8.33 of 
the Act, you must maintain the statutory minimum capital level.


Sec.  652.80  When you must determine the risk-based capital level.

    (a) You must determine your risk-based capital level at least 
quarterly, or whenever changing circumstances occur that have a 
significant effect on capital, such as exposure to a high volume of, or 
particularly severe, problem loans or a period of rapid growth.
    (b) In addition to the requirements of paragraph (a) of this 
section, we may require you to determine your risk-based capital level 
at any time.
    (c) If you anticipate entering into any new business activity that 
could have a significant effect on capital, you must determine a pro 
forma risk-based capital level, which must include the new business 
activity, and report this pro forma determination to the Director, 
Office of Secondary Market Oversight, at least 10-business days prior 
to implementation of the new business program.


Sec.  652.85  When to report the risk-based capital level.

    (a) You must file a risk-based capital report with us each time you 
determine your risk-based capital level as required by Sec.  652.80.
    (b) You must also report to us at once if you identify in the 
interim between quarterly or more frequent reports to us that you are 
not in compliance with the risk-based capital level required by Sec.  
652.70.
    (c) If you make any changes to the data used to calculate your 
risk-based capital requirement that cause a material adjustment to the 
risk-based capital level you reported to us, you must file an amended 
risk-based capital report with us within 5-business days after the date 
of such changes;
    (d) You must submit your quarterly risk-based capital report for 
the last day of the preceding quarter not later than the last business 
day of April, July, October, and January of each year.


Sec.  652.90  How to report your risk-based capital determination.

    (a) Your risk-based capital report must contain at least the 
following information:
    (1) All data integral for determining the risk-based capital level, 
including any business policy decisions or other assumptions made in 
implementing the risk-based capital test;
    (2) Other information necessary to determine compliance with the 
procedures for determining risk-based capital as specified in Appendix 
A to this subpart; and
    (3) Any other information we may require in written instructions to 
you.
    (b) You must submit each risk-based capital report in such format 
or medium, as we require.


Sec.  652.95  Failure to meet capital requirements.

    (a) Determination and notice. At any time, we may determine that 
you are not meeting your risk-based capital level calculated according 
to Sec.  652.65, your minimum capital requirements specified in section 
8.33 of the Act, or your critical capital requirements specified in 
section 8.34 of the Act. We will notify you in writing of this fact and 
the date by which you should be in compliance (if applicable).
    (b) Submission of capital restoration plan. Our determination that 
you are not meeting your required capital levels may require you to 
develop and submit to us, within a specified time period, an acceptable 
plan to reach the appropriate capital level(s) by the date required.


Sec.  652.100  Audit of the risk-based capital stress test.

    You must have a qualified, independent external auditor review your 
implementation of the risk-based capital stress test every 3 years and 
submit a copy of the auditor's opinion to us.

Appendix A--Subpart B of Part 652--Risk-Based Capital Stress Test

1.0 Introduction.
2.0 Credit Risk.
2.1 Loss-Frequency and Loss-Severity Models.
2.2 Loan-Seasoning Adjustment.
2.3 Example Calculation of Dollar Loss on One Loan.
2.4 Calculation of Loss Rates for Use in the Stress Test.
3.0 Interest Rate Risk.
3.1 P