Federal Agricultural Mortgage Corporation; Risk-Based Capital Requirements

SUMMARY: This proposed rule amends Farm Credit Administration (FCA)

regulations, through the Office of Secondary Market Oversight (OSMO),

by establishing risk-based capital requirements for the Federal

Agricultural Mortgage Corporation (Corporation or Farmer Mac). The

proposed regulations: Set forth the risk-based capital rules for Farmer

Mac, including definitions, methods, parameters and guidelines for

developing and implementing the risk-based capital stress test; specify

capital calculation, reporting, and compliance requirements; delineate

our monitoring, examination, supervisory, and enforcement activities;

and, prescribe certain policy requirements for business and capital

planning.

DATES: Please send your comments to us by March 13, 2000.

ADDRESSES: You may mail or deliver written comments to Carl A.

Clinefelter, Director, Office of Secondary Market Oversight, Farm

Credit Administration, 1501 Farm Credit Drive, McLean, Virginia 22102-

5090, or send them by facsimile transmission to (703) 734-5784. You may

also send comments via electronic mail to ``[email protected]'' or

through the Pending Regulations section of our website at

``www.fca.gov.'' You may review copies of all comments we receive in

the Office of Policy and Analysis, Farm Credit Administration.

FOR FURTHER INFORMATION CONTACT:

Carl A. Clinefelter, Director, Office of Secondary Market Oversight,

Farm Credit Administration, McLean, VA 22102-5090, (703) 883-4280, TDD

(703) 883-4444, or Dennis K. Carpenter, Senior Policy Analyst, Office

of Policy and Analysis, Farm Credit Administration, McLean, VA 22102-

5090, (703) 883-4498, TDD (703) 883-4444,

or

Joy Strickland, Senior Attorney, Office of General Counsel, Farm Credit

Administration, McLean, VA 22102-5090, (703) 883-4020, TDD (703) 883-

4444.

SUPPLEMENTARY INFORMATION:

I. Objective

The purpose of this proposed regulation is to establish a risk-

based capital stress test for the Corporation as required by section

8.32 of the Farm Credit Act of 1971, as amended (Pub. L. 92-181)(Act).

Section 8.32 of the Act requires us to establish a risk-based capital

stress test that will determine the level of regulatory capital \1\

necessary for the Corporation to maintain positive capital during a 10-

year period where stressful credit and interest rate conditions occur.

The proposed rule contains specific information on the structure of the

risk-based capital stress test, including guidelines for its

implementation, monitoring, reporting and examination. The rule also

includes requirements for business and capital planning. The guidelines

and procedures for implementation of the stress test are available to

the public through the proposed regulation, technical appendix to part

650, subpart B, and an electronic version of the risk-based capital

stress test (spreadsheet-based) that is available on our website

``www.fca.gov'' or on written request. The technical appendix contains

details on how to construct the risk-based capital stress test,

including basic assumptions used in the test.

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

\1\ ``Regulatory capital'' is defined in section 8.31(5) of the

Act as core capital plus an allowance for losses and guarantee

claims (in accordance with generally accepted accounting principles

(GAAP)). For the purposes of this definition, regulatory capital

includes any allowance or reserve accounts that Farmer Mac maintains

for losses on loans that are held in portfolio and for losses on

securities it has guaranteed, particularly, reserves required by

section 8.10 of the Act.

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

II. Farmer Mac Organization

Farmer Mac is a federally chartered instrumentality of the United

States (U.S.) established on January 6, 1988 by the Agricultural Credit

Act of 1987 (Pub. L. 100-233)(1987 Act), which amended the Act. The

Corporation's status as a Government-sponsored enterprise (GSE)

requires it to fulfill the public policy mission of providing a

secondary market for agricultural real estate loans. The Corporation is

charged with increasing liquidity to rural lenders, increasing

available long-term credit to farmers and ranchers at stable interest

rates, and enhancing the ability of individuals in rural communities to

get financing for moderately priced homes. Congress established the

Corporation as part of its efforts to resolve the agricultural crisis

of the 1980s. Congress believed that a secondary market for

agricultural mortgages would increase available mortgage credit to

America's farmers, ranchers and rural homeowners. Farmer Mac serves

this role mainly by buying and securitizing ``qualified loans'' \2\

from lenders, thereby restoring the lenders' availability of funds to

make new loans. Although created by Congress, Farmer Mac is privately

owned with its common stock publicly traded on the New York Stock

Exchange.

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

\2\ A qualified loan is a loan secured by a first lien, fee

simple mortgage or a long-term leasehold mortgage on agricultural

real estate or rural housing that is located in the U.S.

Agricultural real estate is defined by Farmer Mac as a parcel or

parcels of land, which may be improved by buildings or other

structures permanently affixed to the parcel or parcels, that (1)

Are used for the production of one or more agricultural commodities,

and (2) consist of a minimum of five acres or are used in the

production of agricultural receipts of at least $5,000. In

accordance with the Act, the maximum principal amount of a qualified

loan secured by agricultural real estate is indexed to inflation and

currently is $3.49 million, unless the loan is secured by 1,000

acres or less, in which case the maximum loan size is set by Farmer

Mac at $6.0 million.

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

III. Corporation Authorities and Statutory Requirements for Risk-

Based Capital

Farmer Mac's statutory authority, which was established under title

VIII of the Act, has been substantively amended three times since its

origination in 1988 (i.e., 1990, 1991, and 1996). The 1990 amendments

authorized the Farmer Mac II program at the request of the United

States Department of Agriculture (USDA). The 1991 amendments authorized

the Farmer Mac linked portfolio program.\3\ The 1991 amendments created

OSMO and established the FCA, acting through OSMO, as the regulator of

Farmer Mac. The 1991 amendments also set forth definitions for core

capital \4\ and regulatory capital. The 1991 amendments also

established minimum

[[Page 61741]]

capital \5\ and critical capital \6\ levels and required us to

establish risk-based capital requirements for Farmer Mac.

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

\3\ The linked portfolio authority allows Farmer Mac to purchase

guaranteed securities that have been issued by Farmer Mac or another

authorized issuer and hold the securities indefinitely in its

portfolio.

\4\ ``Core Capital'' is defined in section 8.31(2) of the Act as

the sum (as determined in accordance with GAAP) of: (1) The par

value of outstanding common stock; (2) the par value of outstanding

preferred stock; (3) paid-in capital; and (4) retained earnings.

\5\ The Corporation's ``minimum capital'' requirements are

described under section 8.33 of the Act. The minimum capital level

for the Corporation is an amount of core capital equal to the sum

of: (1) 2.75 percent of the aggregate on-balance sheet assets of the

Corporation, as determined in accordance with GAAP; and (2) 0.75

percent of the aggregate off-balance sheet obligations of the

Corporation which include: (a) The unpaid principal balance of

outstanding securities that are guaranteed by the Corporation and

backed by pools of qualified loans; (b) instruments that are issued

or guaranteed by the Corporation and are substantially equivalent to

(a); and (c) other off-balance sheet obligations. These minimum

statutory capital standards will continue in effect after the risk-

based capital rule becomes effective.

\6\ The Corporation's ``critical capital level'' is described in

section 8.34 of the Act. The critical capital level for the

Corporation is an amount of core capital equal to 50 percent of the

total minimum capital amount determined under section 8.33 of the

Act.

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

The 1996 amendments served to streamline Farmer Mac's operating

structure to be more competitive. Specifically, and most importantly,

Farmer Mac was allowed to buy loans directly from lenders and issue

guaranteed securities representing 100 percent of the principal of the

purchased loans. This amendment removed the previous requirement for

poolers to hold at least a 10-percent subordinated interest against

loan losses on pools of loans securitized by Farmer Mac.

The Food, Agriculture, Conservation, and Trade Act Amendments of

1991 (Pub. L. 102-237)(1991 Act) required us to develop and issue a

risk-based capital stress test for the Corporation. The Farm Credit

System Reform Act of 1996 (Pub. L. 104-105)(1996 Act) further amended

the Act by prohibiting us from establishing a risk-based capital stress

test prior to February 10, 1999, 3 years following the effective date

of the 1996 Act. The risk-based capital stress test required by the

1991 Act determines the amount of capital necessary for the Corporation

to preserve positive capital while undergoing stressful credit and

interest rate risk conditions during a 10-year period. The 1991 Act

also required an added amount of capital to cover management and

operational risk.

Section 8.32 of the Act requires that the risk-based capital stress

test subject the Corporation to credit losses on agricultural mortgages

it owns or guarantees. The frequency of loan default and severity of

losses must be reasonably related to a ``benchmark'' with the highest

rate of default and severity of agricultural mortgage losses

experienced during a historical period of at least 2 consecutive years.

The credit losses also must be related to those experienced in

contiguous areas of the U.S. containing at least 5 percent of the total

U.S. population. The establishment of the benchmark loss experience is

more fully discussed later in this preamble.

The 1991 Act also required us to incorporate in the risk-based

capital stress test an interest rate risk stress scenario based on

rising and falling interest rates on Treasury obligations of various

terms. Under the interest rate stress scenario, current rates on

Treasury obligations are instantaneously shocked up and down. For the

first 12 months of the 10-year stress period, rates either increase or

decrease by: (1) 50 percent of the average rates on various Treasury

obligations during the 12-month period preceding the stress period, or

(2) 600 basis point (bp), whichever is less. The rates must remain at

the increased or decreased level for the remainder of the 10-year

stress period.

In addition to the risk-based capital level required as a result of

the credit loss and interest rate change components of the risk-based

capital stress test, Farmer Mac is required to maintain additional

capital to protect against management and operational risk. This

additional capital level is specified in the Act to be 30 percent of

the capital level required for the sum of the credit loss and interest

rate change components of the risk-based capital stress test.

In developing the risk-based capital stress test, the Act permits

us to take into account appropriate distinctions relative to various

types of agricultural mortgage products, varying terms of Treasury

obligations, and any other factors considered appropriate. We may also

consider credit loss protection provided by retained subordinated

participation interests, which were required for guaranteed securities

under section 8.6(b)(2) of the Act prior to the enactment of the 1996

Act.\7\ The 1991 Act does not require a specific adjustment for any of

these factors, but allows us to determine how best to account for them.

Unlike the risk-based capital stress tests applicable to other GSEs,

the 1991 Act does not contain specific requirements for addressing new

business and other corporate activities during the stress period,

including growth, product types, and pricing.\8\

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

\7\ Section 8.32 of the Act states that we must also conform

loan loss data to the geographic and commodity diversification

standards that the Corporation loan pools had to meet based on

provisions of the 1991 Act. Because the geographic and commodity

diversification requirement was eliminated in the 1996 Act, this

consideration is no longer applicable.

\8\ See 12 U.S.C. 4611(a)(3).

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

Our risk-based capital regulations must contain specific

information on the requirements, definitions, methods and parameters

used in implementing the risk-based capital stress test in order to

enable others to apply the test in a similar manner. Finally, we must

ultimately make available to the public any statistical model used to

implement the risk-based capital stress test.

IV. Philosophy and Development of the Risk-Based Capital Stress

Test

The principal objective of the risk-based capital standard is to

ensure that Farmer Mac has sufficient capital to remain solvent in the

face of extreme economic conditions. We believe that effective capital

standards should also permit Farmer Mac to fulfill its public policy

mission while pursuing prudent business practices and strategies.

Although the risk-based capital stress test can produce a single

capital requirement, it effectively creates marginal capital

requirements, that is, incremental requirements based on the riskiness

of each additional dollar of business for every type of product that

Farmer Mac guarantees or holds in its portfolio. Marginal capital

requirements for mortgages held in portfolio will vary depending on the

interest rate and credit risk associated with the mortgages as well as

the Farmer Mac's funding strategy. These marginal capital requirements

may have significant bearing on how Farmer Mac implements its business

strategies.

We developed the risk-based capital stress test to closely reflect

the risks inherent in Farmer Mac's various business activities. We

incorporated, to the extent permitted by the Act, consistent

relationships between the economic environment of the stress period and

Farmer Mac's business activities. This required modeling Farmer Mac's

assets, liabilities, and off-balance sheet positions at a sufficient

level of detail to capture their various risk characteristics.

Our philosophy guiding the development of the risk-based capital

stress test was that it should:

Be consistent with the requirements of the statute, i.e.,

it should reflect worst-case credit conditions and interest rate

movements, as defined in the Act;

Reflect Farmer Mac's regulatory capital needs for credit

and interest rate risks measured under stressful conditions;

Be internally consistent;

[[Page 61742]]

Not create inappropriate economic incentives;

Aim for simplicity; and

Reflect, to the extent practical and meaningful, Farmer

Mac's current operating policies and practices.

In developing the risk-based capital regulations, we also compared

our statutory requirements with the Basle Accord risk-based capital

framework. Although the current Basle Accord and our risk-based capital

framework significantly differ, both strive to equate risk with an

appropriate capital requirement. We note that the proposed direction of

the Basle Committee suggests an increasing reliance and acceptance of

econometric and statistical models for measuring credit and market risk

and allocating capital.\9\ Additionally, we both advocate that

proactive regulatory measures, such as our risk-based capital stress

test, should be complemented by effective monitoring, supervision, and

examination. For these reasons, we believe our risk-based capital

framework is consistent with the current opinions of the Basle

Committee.

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

\9\ A New Capital Adequacy Framework is a consultative paper

issued by the Basle Committee on Banking Supervision. A copy of this

paper can be found at www.bis.org.

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

In developing the risk-based capital stress test, we engaged in

three distinct activities that vary in complexity and time horizons:

Identification of the benchmark loss experience;

Construction of the risk-based capital stress test; and

Examination and oversight.

A. Identification of the Benchmark Loss Experience

Our first initiative was to identify the worst-case historical loss

experience as required by the Act. We published our results for comment

in the Federal Register on July 28, 1998 (63 FR 40282). This study

entitled, ``Risk-based Capital Regulations for Farmer Mac: Loan Loss

Estimation Procedures,'' is available through our website at

(www.fca.gov/pubs/farmmac). The study was prepared by Barry &

Associates,\10\ consultants who also assisted us in all facets of

development of the risk-based capital stress test. The following is a

brief summary of our efforts to locate agricultural mortgage loan data

and identify the worst-case loss rates to serve as a benchmark for the

loss rates used in the risk-based capital stress test.

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

\10\ Barry & Associates is a consulting group that conducts

research and education projects in agricultural finance on behalf of

industry, policy, and non-profit organizations. Projects have

included analyses of capital regulations for financial institutions,

insurance modeling, risk pricing of loans, community banks' access

to agency market funds, and Farm Bill changes. Principal members of

Barry & Associates are Peter Barry (Managing Partner), Bruce

Sherrick, Paul Ellinger, and Del Banner. Each of these members is

affiliated with the Center for Farm and Rural Business Finance at

the University of Illinois, Urbana-Champaign.

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

1. Available Loan Data

We were unable to use Farmer Mac loan data for establishing the

benchmark loss experience because Farmer Mac is a relatively new

enterprise and did not have historical data. Therefore, we searched for

other possible data sources, including the Economic Research Service

(ERS) of the USDA, commercial banks, life insurance companies, and

System banks. After an exhaustive search, we identified the Farm Credit

Bank of Texas (FCBT) and the former Farm Credit Bank of St. Paul

(FCBSP) as the only data sources with available historic loan-level

performance data on Farmer Mac-eligible loans that satisfied the

statutory provisions.

After an extensive evaluation of the available data, we concluded

that the FCBT had the most relevant data available for developing a

benchmark loss experience for use in a risk-based capital stress test.

Data from the FCBT is the most useful available for determining

benchmark losses because losses were taken expeditiously as charge-offs

and are thus clearly measurable. In contrast, although the FCBSP

experienced substantial financial distress during the 1980s, charge-off

rates were relatively low due to the FCBSP's strategy of forbearance

and restructuring of problem loans. Thus, the FCBSP experienced most of

its financial stress through reduced earnings on loans and increased

servicing costs. This stress is more difficult to measure in the form

of loan default rates and severity of defaults, which the statute

requires us to measure. Also, we proposed to use only the FCBT loans

that would have met the Farmer Mac underwriting standards that were in

place at loan origination because non-conforming loans could present

significantly different credit, market, and institutional risks.

2. Identification of Worst-Case Losses

According to the USDA, Texas ranked fourth among the 50 states in

terms of farm financial stress in the 1980s. In addition, our

experience with the System banks indicates that the FCBT did not

experience the worst historical losses on agricultural real estate

mortgages. Therefore, the only usable loan data we were able to

identify, the FCBT data, did not represent the worst-case agricultural

mortgage loss experience. We, thus, found it necessary to consider how

Texas loss rates related to other states and regions for determining

worst-case loss experience.

We employed a statistical procedure to expand the loan loss rates

for the FCBT to estimate loss rates for other states and regions of the

U.S. This procedure is explained in detail in the study published for

public comment. Briefly, the preferred regression equation identified

by Barry & Associates was based on the relationship between FCBT loss

rates and the annual percentage change in Texas farmland values over

the next 2 years. This regression equation was used to estimate

historical loss rates for every state from 1976 to 1993. Then a ranking

was compiled of 2-year loss rates for contiguous regions representing

at least 5 percent of the 1990 U.S. population. Our study concluded

that the worst-case region was found to contain Minnesota, Iowa and

Illinois during the 1983-1984 time period with a 2-year loan loss rate

of 4.18 percent. This region represents 7.5 percent of the U.S.

population. Our experience in overseeing FCS institutions with severe

credit problems and high default rates during this period also points

to the upper mid-west as the focal point of agricultural stress.

3. Use of the Benchmark Losses

Following our identification of the worst-case benchmark loss rate,

we began our development of the risk-based capital stress test. The Act

requires that the risk-based capital stress test use losses that occur

throughout the U.S. The identified losses are to be at a rate of

default and severity ``reasonably related'' to the rate and severity

that occurred for at least 2 years in contiguous areas of the U.S.

containing not less than 5 percent of the U.S. population.

The published study used loss rates extrapolated from the FCBT data

to identify the worst-case region. The primary variable used in the

extrapolation was the change in farmland values. However, we are not

proposing to use the benchmark loss rate in the risk-based capital

stress test. The extrapolation method used in the study was an

appropriate method for estimating aggregate loss rates on agricultural

mortgages. The method proposed here allows us to incorporate the

current risk characteristics of Farmer Mac's portfolio, including loan-

level data, in addition to the farmland value changes for the worst-

stress time period. These loan-level risk characteristics include loan

size, loan-to-value ratio (LTV), debt service coverage ratio

[[Page 61743]]

(DSCR), and debt-to-asset (D/A) ratio. The statistical method proposed

makes it easier for Farmer Mac to implement and for us to examine the

results.

Although we are not using the benchmark loss rate identified in the

published study, we use the percentage changes in farmland values from

the published study as a primary variable in estimating the loss rates

used in the stress test. Using the farmland value changes from the

published benchmark worst-case region of Minnesota, Iowa, and Illinois

as input in the credit risk portion of the risk-based capital stress

test is a direct linkage to the benchmark loss rate. The loss rates

used in the risk-based capital stress test are closely related to the

benchmark loss rate, because both are limited to changes in farmland

values. The changes in farmland values identified in the published

benchmark loss rate study are primary variables in the default equation

used in the risk-based capital stress test. Changes in farmland values,

as used in the risk-based capital stress test and the benchmark loss

rate study, represent the combined effects of the level and growth

rates of farm income, interest rates, and inflationary expectations.

More detailed information on the procedure for calculating loss rates

in the risk-based capital stress test is presented later in this

supplementary information and in the technical appendix.

4. Comments on the Benchmark Loss Report

As noted earlier, in July 1998 we published a ``Notice of

availability of study and request for comment'' on the loan loss study

completed by Barry & Associates. (See 63 FR 40282, July 28, 1998.)

Through the Notice, we made the results of the study available for

public comment in expectation that it would lead to improved input for

the credit risk component of the risk-based capital stress test. Due to

the complexity of the study and the importance of the benchmark loss

experience in the risk-based capital stress test, several parties

requested that we extend the comment period on the Notice from

September 15, 1998, to January 4, 1999, which we did.

We received five letters on the study from a variety of interested

parties. The commenters were Farmer Mac (2 letters), AgFirst Farm

Credit Bank, the American Bankers Association, and the Independent

Bankers Association of America (now the Independent Community Bankers

of America). We have considered these comments in the development of

this proposed risk-based capital regulation. Many of the comments were

related to the benchmark loss rates that were identified rather than

the loan loss data we used as the starting point or the land value

changes we identified. Because we are not using the benchmark loss

rates as the loss rates in the risk-based capital stress test, we do

not believe that a detailed response to each comment is relevant in

this supplementary information. Thus, we are providing a summary of,

and our response to, the primary comments relevant to this proposed

rule.

First, the commenters stated that the statute does not permit

extrapolation procedures in identifying the worst-case loss data. The

commenters asserted that the statute required us to use as benchmark

losses, the worst-case data that are available to us, i.e., the losses

from the FCBT portfolio. We disagree with this interpretation of the

statute. The statute directs us to use the worst-case data, not the

worst-case data ``that are available.'' Congress directed us to use

loan loss rates in the risk-based capital stress test that are

reasonably related to the area of the U.S. that experienced the

``highest rates of default and severity.'' Therefore, our first step

was to determine the benchmark worst-case losses pursuant to this

requirement.

Second, the commenters stated that the study failed to account for

appropriate credit risk distinctions between the historical FCBT data

and Farmer Mac's current portfolio. We believe it was appropriate to

use only those agricultural mortgages that would have been Farmer Mac-

eligible loans at the time the study was conducted so that the

benchmark losses would reflect losses on relevant loans. We reviewed

the process for selecting the screening criteria used in the study and

found it to be appropriate given the underwriting data of Farmer Mac's

portfolio, the statutory criteria for loan eligibility, and the

limitations of the FCBT data set. We also reviewed the eligible loan

set obtained from the FCBT data and determined that all variables were

within the values found in Farmer Mac's current portfolio. Thus, we

believe the data used for the benchmark study are appropriate. We also

consider it appropriate to account for Farmer Mac's current portfolio

risk factors in applying the loss rates in the risk-based capital

stress test. Thus, we consider the current make-up of Farmer Mac's

portfolio when we apply the loan loss default equation to determine the

loss rates used in the risk-based capital stress test. Later

discussions in this supplementary information and the technical

appendix further explain how the risk characteristics of Farmer Mac's

portfolio are incorporated in determining the loss rates used in the

risk-based capital stress test.

B. Construction of the Statutory Stress Test

Our second major undertaking was to develop a financial model to

represent Farmer Mac's assets, liabilities and off-balance sheet

positions at a sufficient level of detail to capture important risk

characteristics and project Farmer Mac's financial performance over a

hypothetical period of stress lasting 10 years. The focus of our

efforts was to determine the appropriate parameters and economic

relationships necessary for the risk-based capital stress test to

fulfill the statutory requirements. To accomplish this task, we worked

in consultation with Barry & Associates. Additionally, Farmer Mac and

PriceWaterhouseCoopers provided information relative to loan data,

Farmer Mac's operations, and economic relationships and statistical

methodologies for use in measuring various types and levels of risk.

A summary of the major components of the risk-based capital stress

test is provided in a subsequent section of this preamble and the

regulation. A more thorough discussion of all the technical aspects of

the risk-based capital stress test can be found in the technical

appendix to the proposed regulation. Due to the nature and complexity

of the risk-based capital stress test, we are also making an electronic

version of the risk-based capital stress test available to the public

through our website at www.fca.gov.

The proposed rule specifies the basic structure and parameters of

the risk-based capital stress test and allows Farmer Mac to implement

the stress test internally using a model built according to our

specifications to determine its risk-based capital level. During the 1-

year period following adoption of the final risk-based capital

regulation and on an ongoing basis thereafter, we will examine and

verify Farmer Mac's implementation of the risk-based capital stress

test to ensure compliance with the regulation, including the

specifications identified in the technical appendix to the regulation.

Furthermore, we are proposing that Farmer Mac have its implementation

of the risk-based capital stress test verified and audited once every 3

years by an external independent party. The audit should ensure that

the financial data used in the stress test are accurate and that stress

test is implemented in accordance with our regulations and procedures.

We note that because of the proprietary nature of specific, transaction

level loan and financial data used in the risk-based

[[Page 61744]]

capital stress test, it is unlikely that results of the test will be

fully reproducible by parties other than Farmer Mac and us. Other

parties will, however, be able to approximate the test results on an

aggregate basis using publicly available information.

C. Examination and Oversight

From a regulatory perspective, the ongoing nature of the risk-based

capital stress test facilitates our understanding of how changes in

Farmer Mac's business activities will affect its risk profile and

resulting capital requirements. A risk-based capital stress test,

because it is based on statistical relationships, is limited in a

number of important ways that must be understood before it can be used

as an effective regulatory tool. Foremost, the risk-based capital

stress test uses econometric relationships based on historical data to

estimate potential loss rates. Past historical data, even though

required by the statute, may not be the best basis for projecting the

performance of new agricultural mortgages originated using a different

set of underwriting criteria and subject to a different set of economic

conditions. As a result, we may need a significant period of time to

collect and analyze new data to appropriately update the risk-based

capital stress test procedures. Furthermore, the effectiveness of the

risk-based capital stress test may be influenced by changes in Farmer

Mac's operations, underwriting standards or products and services

offered. Lastly, the risk-based capital stress test best measures

identifiable and quantifiable risks.

Therefore, our ongoing monitoring and on-site examination will be

integral in assessing Farmer Mac's capital adequacy. Our monitoring and

examination program will help ensure that Farmer Mac appropriately

implements the risk-based capital stress test and aid in identifying

non-quantifiable risks that the risk-based capital stress test cannot

measure. Together, the ongoing monitoring and examination by OSMO will

enable us to provide effective regulatory oversight and ensure the

adequacy of regulatory capital standard set by the risk-based capital

stress test.

V. Risk-Based Capital Stress Test

The risk-based capital stress test is intended to be forward-

looking and sensitive to fluctuations in the economy, as well as to

changes in Farmer Mac's asset composition, funding strategies, and on-

and off-balance sheet exposures. The risk-based capital level, unlike

simple leverage ratios, is tailored to specific risks in Farmer Mac's

book of business. In designing the risk-based capital stress test, we

sought to identify and incorporate all significant credit and interest

rate risks to which Farmer Mac is exposed.

Given the risk-based capital stress test's sensitivity to changing

risk conditions, the risk-based capital requirement bears no direct

relationship to the statutory minimum capital requirements. Based on a

Farmer Mac condition of relatively low risk exposure, the risk-based

capital stress test could produce a risk-based capital requirement

below that of the statutory minimum standard. When this occurs, Farmer

Mac must continue to meet its statutory minimum capital level.

Econometric models are used to project the effects of stressed

conditions on Farmer Mac's assets, liabilities and off-balance sheet

activities. The risk-based capital stress test will project credit

losses from defaults on agricultural mortgages and loss severities

comparable to the worst historical agricultural mortgage default loss

experience in any region of the country.

The risk-based capital stress test is designed to capture Farmer

Mac's specific exposure to credit and interest rate risks under

stressed conditions. Economic conditions of the stress scenario affect

Farmer Mac's agricultural mortgage performance, earnings and market

values, and ultimately required capital. For example, movement in

farmland values, which reflect changes in farm income and interest

rates, influence mortgage credit loss rates, which in turn affect

Farmer Mac's cashflows and capital accretion or depletion. By requiring

the risk-based capital stress test to be conducted on a quarterly

basis, we will strive to identify changes in capital needs before such

economic events as declining farmland values can impact Farmer Mac's

balance sheet to any significant degree. Thus, the risk-based capital

stress test is more dynamic than simple leverage ratios because the

entire business profile of Farmer Mac from assets and liabilities to

off-balance sheet obligations is modeled.

The goal of the risk-based capital stress test is to align capital

requirements with risk and avoid creating incentives for the

Corporation to engage in inappropriately risky activities. The stress

test approach also provides greater flexibility to meet regulatory

requirements than is available in traditional capital requirements. For

instance, the stress test approach recognizes risk-mitigating

activities. As an example, Farmer Mac may meet its risk-based capital

needs by reducing risk and/or increasing capital.

Proposed Sec. 650.24 describes the main components of the risk-

based capital stress test that Farmer Mac must apply to its current

operations. The technical appendix to the regulation provides details

on the specification and estimation of the statistical (econometric)

model used to project Farmer Mac performance over the 10-year stress

period. Additionally, the technical appendix discusses how the

statistical model is applied in the proposed risk-based capital stress

test. The key stress test components, include data requirements,

specifications of credit risk, interest rate movements, the cashflow

generator, and the capital calculation.

The following discussions provide a general overview of the risk-

based capital stress test components and explanation of the concepts

underlying the stress test.

A. Data Requirements

Historical loan data from the FCBT are used to determine

appropriate relationships among mortgage-risk factors, rates of loan

default and loss occurrence. Data on Farmer Mac's current book of

business are used to establish Farmer Mac's initial balance sheet

structure, financial position and risk profile for the start of the

risk-based capital stress test. Current interest rate information as

described in the technical appendix is needed for the interest rate

component of the risk-based capital stress test.

Farmer Mac will be required to provide additional information in

its quarterly financial reports to us. Specific details regarding the

new requirements will be provided through modifications to our Call

Report instructions. Although we are allowing Farmer Mac the

flexibility to determine its risk-based capital level and report its

results to us, we believe it is essential for FCA to retain the

capability to also determine Farmer Mac's risk-based capital level.

Therefore, we intend to modify the current Call Report instructions to

accommodate our data and information needs.

B. Farmer Mac Programs and Risk Characteristics

Farmer Mac operates a variety of secondary market programs with

varying amounts of risk to fulfill its mission. A brief description of

these programs follows.

[[Page 61745]]

1. Farmer Mac I--Direct Loan Purchases

Farmer Mac purchases from approved lenders\11\ qualified loans

secured by a first mortgage on agricultural real estate, including

part-time farms that meet specified credit standards. Farmer Mac

provides liquidity to the agricultural mortgage market by: (1)

Purchasing newly originated qualified loans directly from lenders on a

continuing basis through its ``cash window''; (2) exchanging qualified

loans for securities issued and guaranteed by Farmer Mac (Farmer Mac

Guaranteed Securities) through ``swap'' transactions; and (3)

purchasing portfolios of existing loans on a negotiated basis.\12\

Qualified loans purchased by Farmer Mac are aggregated into pools that

back Farmer Mac Guaranteed Securities, which are periodically issued

and sold to investors in the capital markets. Farmer Mac also has the

authority to purchase these securities and hold them in its portfolio.

Farmer Mac receives income from guarantee fees on securities it

guarantees. Farmer Mac also receives interest income on securities it

holds.

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

\11\ Approved lenders are financial institutions that have met

Farmer Mac's technical, financial and stock ownership requirements.

\12\ See Farmer Mac's 1998 Annual Report.

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

2. Farmer Mac I--Long-term Standby Purchase Commitment

Under a standby purchase commitment agreement with the lender,

Farmer Mac receives an annual fee in return for its commitment to

purchase certain loans in the future. The lender services the loans and

retains them on its books in a segregated pool. This program allows

approved lenders to reduce credit risk and free capital to make

additional loans.

3. Farmer Mac I--AgVantage Bond Sales

AgVantage bonds are highly collateralized corporate debt issued by

approved lenders to Farmer Mac, which in turn guarantees the bonds. The

approved lenders pledge qualified loans and other securities (Treasury

securities) as collateral, which are retained by the lenders. This

program provides approved lenders with another means to fund qualified

loans and generates revenue for Farmer Mac. Farmer Mac receives revenue

in the form of interest income on AgVantage bonds.

4. Farmer Mac II

Under Farmer Mac II, the Corporation purchases the portions of

loans guaranteed by the USDA. The Food, Agriculture, Conservation, and

Trade Act of 1990 (Pub. L. 101-624) (1990 Act) gives Farmer Mac the

authority to operate a secondary market for certain USDA-guaranteed

loans. The guaranteed portions of loans are pooled and securitized by

Farmer Mac or other certified facilities. Farmer Mac then guarantees

the repayment of the securities. Farmer Mac receives income from

guarantee fees on securities it guarantees. Farmer Mac also receives

interest income on securities it holds.

5. Farmer Mac--Rural Housing

Home mortgages from lenders in rural areas and small communities

are eligible for sale to Farmer Mac for pooling and securitization.

Rural housing is defined by Farmer Mac as a one-to-four family, owner-

occupied principal residence that is a moderately priced dwelling

located in a community having a population of 2,500 or fewer

inhabitants; the dwelling (excluding the land to which it is affixed)

cannot have a purchase price or current appraised value of more than

$100,000 (adjusted annually for inflation). This figure is currently

$133,000. In addition to the dwelling, a rural housing loan can be

secured by land associated with the dwelling having an appraised value

of no more than 50 percent of the total appraised value of the combined

property. As of August 31, 1999, Farmer Mac had not issued any

securities backed by rural home mortgages.

6. Risk Characteristics

Farmer Mac's primary exposure to credit risk is the risk of loss

resulting from the inability of borrowers to repay their mortgages.

Farmer Mac is exposed to credit risk on the loans it holds or

guarantees against default, as well as securities it guarantees. Farmer

Mac guarantees the timely payment of principal, including any balloon

payments, and interest on securities. Loans held or guaranteed by

Farmer Mac can be divided into three groups: (1) Pre-1996 Act Farmer

Mac I loans; (2) post-1996 Act Farmer Mac I loans; and (3) Farmer Mac

II loans. Within these general groupings, Farmer Mac, as previously

discussed, operates other programs. Each of these programs carries

different amounts of credit risk that must be appropriately reflected

in the credit risk component of the risk-based capital stress test.

For pre-1996 Act loans, subordinated interests mitigate Farmer

Mac's credit risk exposure. Before Farmer Mac incurs a credit loss,

recourse must be taken against the subordinated interest. At December

31, 1998, the subordinated interest of each outstanding security on

pre-1996 Act Farmer Mac I loans was equal to or greater than 10

percent. The 1996 Act eliminated the subordinated interest requirement.

As a result, Farmer Mac assumes 100 percent of the credit risk exposure

on the post-1996 Act Farmer Mac I loans. Farmer Mac mitigates the

credit risk related to these loans through the application of its

underwriting standards and by requiring collateral in the form of real

estate. Farmer Mac's credit exposure on Farmer Mac II loans is covered

by the ``full faith and credit'' of the U.S. Government by virtue of

the USDA guarantee of the principal and interest on all guaranteed

portions.

There is very limited, if any, credit risk exposure on the pre-1996

Act loans due to the subordinated interest, or on the Farmer Mac II

loans because of the USDA guarantee. For this reason, we are not

requiring Farmer Mac to project any credit losses on these programs

during the stress period of the test. Farmer Mac's credit risk exposure

on post-1996 Act Farmer Mac I loans is fully reflected in the risk-

based capital stress test. Farmer Mac I rural home loans are subject to

the same loss rates as agricultural mortgages. Rural home loan loss

rates are not computed independently given the lack of data and the

fact that there is no outstanding loan volume held on the balance sheet

or guaranteed.

Farmer Mac is also exposed to institutional credit risk relating

to: (1) Issuers of AgVantage bonds and other investments held by Farmer

Mac; (2) sellers and servicers; and (3) interest-rate contract

counterparties. We decided not to model these sources of institutional

credit risk for several reasons. AgVantage bonds are general

obligations of the AgVantage bond issuers and secured by collateral in

an amount ranging from 120 percent to 150 percent of the bond amount.

In addition to requiring collateral, Farmer Mac mitigates credit risk

related to AgVantage bonds by evaluating and monitoring the financial

condition of the AgVantage issuers. Farmer Mac manages institutional

credit risk related to sellers and servicers by requiring such

institutions to meet certain standards and by monitoring their

financial condition and servicing performance. The credit risk inherent

in the investment portfolio is mitigated by Farmer Mac's policy of

investing in highly rated institutions and by establishing

concentration limits, which reduce exposure to any one counterparty.

Furthermore, the short-term nature of Farmer Mac's investment portfolio

limits credit risk. Farmer Mac mitigates credit risk arising from

interest-rate swaps by dealing only with counterparties with high

credit ratings,

[[Page 61746]]

establishing and maintaining collateral requirements, and entering into

netting agreements.

We are proposing to capture Farmer Mac's institutional credit risk

exposure through the 30-percent management and operations risk add-on

required by the statute. At this time, we believe modeling

institutional credit risk presents many challenges that would

unnecessarily complicate the risk-based capital stress test. However,

if Farmer Mac significantly increases its credit risk exposure in these

areas through modifications to its current operating policies, we will

reconsider how to best reflect institutional credit risk exposure in

the risk-based capital stress test.

Farmer Mac is also exposed to credit risk concentration in the

mortgages it holds and guarantees. Farmer Mac's current policy is to

limit its credit exposure in a particular geographic region or

commodity to a percentage of total principal amount of all loans

outstanding. Additionally, Farmer Mac employs more stringent

underwriting criteria in regions with higher loan volume

concentrations. Such underwriting criteria consider the credit quality

of the loans in a particular geographic region or commodity based on

the borrower's LTV, DSCR, equity-to-asset and working capital-to-

current asset ratios. The effectiveness of Farmer Mac's underwriting

standards is specifically measured in the risk-based capital stress

test through our model of loan losses as described more thoroughly in

the next section as well as in the technical appendix.

C. Credit Risk

A statistical methodology is used to model the stress conditions

described in the statute. Econometric models are used to estimate the

probability of mortgage defaults and the severity of loss under

stressed conditions. Detailed instructions for measuring credit risk

are provided in the technical appendix.

1. Estimation of Default.

A logistic model is used to estimate the frequency of defaults from

historical FCBT loan data. Logistic models are widely accepted as an

appropriate methodology for modeling loan-level mortgage defaults.

There are several well-known predictors of mortgage default, including

loan age, payment burden, LTV, and interest rates. Additionally, there

are other variables that are specific to farm mortgages, such as

farmland prices, net farm income, commodity prices and the D/A ratio.

These variables, in addition to a host of others, were considered in

the process of modeling defaults of FCBT loans. After extensive

statistical analyses, the final equation for estimating the frequency

of default includes the following variables:

Maximum decline in farmland values

LTV ratio

DSCR

Original loan balance in 1997 dollars

D/A ratio

These variables have logical relationships to the incidence of loan

default and loss.

a. Farmland values. Changes in farmland prices are an important

factor in the model because they directly affect the likelihood of

mortgage defaults and the magnitude of potential losses. In estimating

the default frequency equation, the largest annual percentage decline

in farmland values resulted in the strongest relationship between an

economic variable and default frequency. For stress test purposes, we

used the largest decline in farmland values from the benchmark loss

experience as an input variable.

Because the lives of loans are unknown at origination and differ

among loan observations, annual economic variables or annual changes

throughout the life of the loan cannot be consistently applied across

all loans. Economic variables need to be expressed in a form that can

be applied to loans regardless of the life of the loan. For example,

geometric average lifetime changes, minimum changes, or maximum changes

could be considered. The maximum percentage decline in annual Texas

farmland values resulted in the strongest relationship among economic

variables considered in the estimated equation for default frequency.

b. Loan-to-value Ratio. Another important variable known to drive

defaults is the LTV ratio. LTV is equal to the loan amount divided by

the appraised value of the underlying property. This variable is one of

the primary underwriting ratios that Farmer Mac uses in its loan

purchasing decisions. LTV indicates the relative safety of

collateralized debt. Large equity investments represent a substantial

incentive for a borrower to continue making mortgage payments to

safeguard their equity position in their property. Furthermore, if an

income problem does arise, lower LTVs provide a significant cushion for

the borrower to sell the underlying asset at a price that is sufficient

to cover accrued interest and the remaining outstanding principal.

Conversely, high LTV loans are more likely to default.

Ideally, an updated LTV could be calculated each year, which would

take into account amortization and changes in property value. While we

cannot obtain updates in the market value of the underlying collateral,

the USDA Texas farmland value series could serve as a proxy for helping

to update the denominator of the LTV. Since the amortization schedule

of the FCBT data was not available, however, updating the numerator

would require making payment assumptions as well. Rather than making

additional assumptions, we opted to use LTV at origination.

c. Debt Service Coverage Ratio. A key factor in our assessment of

potential frequency of default is the DSCR of a loan at origination.

The numerator in the DSCR is net income plus depreciation, interest on

capital debt, capital lease payments, and net off-farm income less

living expenses and income taxes. The denominator is the sum of the

total annual debt service requirements. Loans with low DSCRs have a

higher expected frequency of default because borrowers cannot be

expected to fund losses indefinitely. Conversely, loans with high DSCRs

have a lower likelihood of default because they have an excess cashflow

buffer, which would have to erode before the borrower would experience

losses and consider defaulting. This variable would ideally be updated

over the term of the loan. However, because of data limitations and the

cyclical nature of agricultural receipts, DSCR at origination is used

as the variable.

d. Origination Loan Balance. The beginning loan balance also proved

to be a significant factor of default. We adjusted the origination loan

balance for current constant dollars when we estimated the default

equation and applied it to Farmer Mac's data. The base year we selected

is 1997 and we adjusted the current dollars based on the consumer price

index. It has been observed in the agricultural mortgage market that

larger loans tend to have higher default rates. Liquidity constraints

are the likely cause of this phenomenon. For a borrower who is

experiencing financial difficulty, one of the alternatives is to sell

the property and prepay the loan. Considering that there is relatively

less demand for large farm properties, the owners of these properties

may have difficulty in selling their properties in such an illiquid

market. This inability to sell and then prepay eventually limits the

farmers' alternatives to default.

e. Debt-to-asset ratio. The D/A ratio at origination is the last

explanatory variable used in the default frequency estimation model.

The D/A ratio

[[Page 61747]]

indicates the borrower's total amount of financial leverage. This is an

important factor in agricultural lending because agricultural producers

typically have significant amounts of debt from operations in addition

to farm real estate debt. Borrowers with high D/A ratios experience

higher default rates because they have limited capacity to withstand

adverse conditions.

2. Loss Severity

Loss severity is a key element in the estimation of loan losses.

Loss severity is defined as the total dollar amount of losses on a

defaulted loan expressed as a percentage of origination loan balance.

The loss severity rate is estimated with the same FCBT data employed in

the estimation of defaults. To estimate loss severity, we searched for

a significant statistical relationship between loss rate and various

independent indicators in the FCBT loan-level data. We concluded, after

extensive analysis, that the data set was insufficient to estimate an

acceptable loss severity rate. As a result, the loss severity rate is

calculated by taking the weighted average loss of defaulted loans. The

resulting loss severity rate is 20.9 percent. When a more extensive

data set becomes available, loss severity can be re-estimated.

3. Age Adjustment

Mortgage seasoning (aging) is widely accepted as an important

determinant in default frequency. The probability of default is low in

the early life of a loan and increases as the loan ages until it peaks

in years 6 to 8. After this peak period, the borrower has developed

greater equity in the property and the likelihood of default tapers

off. Therefore, we adjust loan-level loss to reflect the differences in

loss occurrence attributed to loan seasoning. We used FCBT data to

estimate the distribution function for loan seasoning assuming that the

loans have a 14-year average life.

4. Time Pattern of Loss Occurrence

The age-adjusted losses are then distributed through time on a

deterministic path that is representative of a stressful scenario. The

loss rates estimated in the credit risk component of the risk-based

capital stress test are based on an origination year concept. Under

this approach, all losses arising from loans originated in a particular

year are expressed as a percent of that year's originated loan volume

irrespective of when the losses actually occur. The stress test must

adjust the origination loss rates to an exposure year concept, in which

losses occurring in any 1 year are related to the total outstanding

loan volume in that year. Because all loans held at any time are not

all originated in that year (or, conversely, loan principal balances

are reflected on more than 1 year's balance sheet), the origination

year loss rates must be adjusted to exposure year rates. To adjust from

origination to exposure year losses, we apply a deterministic time path

for loss occurrence during the 10-year stress period. The deterministic

time path for converting from origination year to exposure year was

determined by calculating exposure year losses in the FCBT data and

expressing such losses as proportions of total losses for each

origination year. The maximum 1-, 2-, 3-, and 4-year commutative

proportions of total origination loss observed in the FCBT data are

used in the first four periods of the stress test. The remaining losses

are equally allocated to years 5 through 10 of the stress test.

D. Interest Rate Risk

The statute requires the risk-based capital stress test to

incorporate an interest rate risk component. Interest rate risk is the

risk that interest rate changes could materially affect Farmer Mac's

market value of equity and future earnings. Farmer Mac may be exposed

to interest rate risk through any product or activity that is sensitive

to changes in interest rates. Farmer Mac is exposed to three primary

sources of interest rate risk: (1) Farmer Mac I and Farmer II

securities; (2) other assets held for investment; and (3) loans held

for securitization.

Farmer Mac's primary strategy to manage interest rate risk related

to Farmer Mac I and II securities and other assets held for investment

is to fund them with liabilities that have similar durations or average

cashflow patterns over time.\13\ To achieve the desired liability

duration, Farmer Mac uses a mix of short-term discount notes and

callable and non-callable medium term notes. By using a mix of

liabilities that includes callable debt, the duration of the

liabilities will tend to increase or decrease as interest rates change

in a manner similar to the changes in the duration of assets. Farmer

Mac also uses a variety of off-balance sheet derivative financial

instruments to manage its interest rate risk exposure.

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

\13\ See Farmer Mac's 1998 Annual Report.

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

The Treasury yield curve represents the market's view of risk-free

borrowing over a range of maturities. As such, it serves as a

foundation for all other market rates. In the context of the risk-based

capital stress test, the general level of interest rates will directly

affect major components of Farmer Mac's business, including borrowing

costs and earnings on mortgages and investments.

The statute specifically describes how Treasury rates must vary

during the 10-year risk-based capital stress test period. While the Act

provides a fairly specific description of how rates change in the

stress scenario, it is silent on how we are to measure the financial

effect of those rate changes. Accordingly, we are proposing the

following procedures for implementing the statutory stress test and

measuring Farmer Mac's exposure to interest rate risk.

Measurement of Farmer Mac's interest rate risk exposure requires

the ability to estimate the sensitivity of the Corporation's assets and

liabilities to interest rate risk. Vulnerability to interest rate risk

is expressed through the degree of match between an institution's rate

sensitive assets and liabilities, or between the durations of its

assets and liabilities. More closely matched positions reduce the

vulnerability to interest rate risk. By determining its duration gap,

as measured by the difference between the duration of assets and

liabilities under various parallel and instantaneous shifts in the

yield curve, Farmer Mac can assess the potential effects of mismatches

between the durations of its assets and liabilities. Farmer Mac derives

its interest rate sensitivity measures using a commercially developed

model, current market information, and other proprietary information.

We are proposing to use Farmer Mac's duration measures as inputs

into the stress test to capture the cumulative effects of the

Corporation's interest rate risk exposure under the interest rate shock

scenarios required by the Act. We have two reasons for using this

approach. One, we routinely assess Farmer Mac's interest rate risk

measurement and management through our examination process. During this

process, we closely evaluate the assumptions and inputs used in Farmer

Mac's interest rate risk sensitivity measures. Therefore, we can

validate the process and obtain the necessary confidence in the

accuracy and integrity of the results to permit us to use them as

inputs into the stress test. Our second reason for using Farmer Mac's

internal duration measures is that it reduces the complexity of the

stress test, thereby increasing the efficiency in implementing the

model.

To estimate the effects of the interest rate shocks (up and down

scenarios) on Farmer Mac's equity position, we compute effective

duration over each

[[Page 61748]]

interest rate shock scenario using information supplied by Farmer Mac.

The duration measure is then used as a proxy for market value effects

under each interest rate scenario. We consider Farmer Mac's assets and

liabilities to be available for sale under GAAP. Thus, Farmer Mac must

record changes in market values as increases or decreases to equity on

its balance sheet. Finally, Farmer Mac must determine its risk-based

capital level based on the rate movement (increase or decrease) that

results in the highest level of required capital.

As noted, Farmer Mac is subject to interest rate risk on all assets

held for investment because of the timing differences in the cashflows

of the assets and related liabilities. This risk is primarily related

to Farmer Mac I and II securities because of the ability of borrowers

to repay their mortgages. Mortgage prepayments can cause fluctuations

in the value of Farmer Mac securities to the extent they change

cashflows of Farmer Mac's on- and off-balance sheet assets. Yield

maintenance provisions associated with many of the loans underlying

Farmer Mac securities significantly reduce, but do not eliminate, this

risk.\14\

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

\14\ Yield maintenance provisions require borrowers to make an

additional payment to Farmer Mac when they repay their loans.

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

Although the effects of increasing and decreasing prepayments are

captured in Farmer Mac's market value results, we are also proposing to

use prepayment rates as input variables for generating balance sheet

cashflows. We provide Farmer Mac the option to use their actual

prepayment experience or assumed prepayment rates estimated from other

data sources for the reasons explained below.

Prepayment rates often are estimated statistically by measuring

relationships between prepayment and a set of independent variables

that have been found to influence the prepayment rate. For several

reasons, however, statistical relationships were not feasible to

estimate in this case. During the historic time period in which the

FCBT data were compiled, the FCBT priced its farm real estate loans

with floating interest rates that adjusted annually in response to

changes in the FCBT's average cost of funds. The resulting loan rates

followed changes in market rates but with slower and lower rates of

change. Because Farmer Mac does not engage in average cost pricing,

estimating a prepayment function from the FCBT data would bear little,

if any, relationship to prepayment rates experienced by Farmer Mac in

the future. Moreover, an explicit or implicit prepayment measure was

not available or obtainable from the FCBT data.\15\

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

\15\ The FCBT data set has a ``status'' variable that indicates

whether the loan was active, foreclosed, re-amortized, paid in full

or merged with a new loan.

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

Implementation of the interest rate shock requires several steps.

The statutory interest rate shock is applied to the initial interest

rate level, which is the preceding 3-month average of the 10-year

Constant Maturity Treasury (CMT) rate. The 10-year CMT is frequently

used for financial modeling of GSEs \16\ since it is viewed as a good

index for the cost of funds. Previous studies by Barry & Associates

found the 10-year CMT to be a reliable index for System funding costs,

and the 10-year CMT did not suffer random or unexplainable variations

observed at shorter-term points on the yield curve. Thus, using a 3-

month average avoids the possibility of unusual and extreme short-term

movements in interest rates unduly influencing the results of the test

and Farmer Mac's risk-based capital requirement. The interest rates

resulting after the rate shock serve as the index needed to simulate

mortgage and investment performance over the stress period and to

calculate the risk-based capital level. Because many different interest

rates affect Farmer Mac's business performance, we allow the use of

other non-Treasury yield curves to simulate the financial effects of

the interest rate shock on Farmer Mac's cashflows, income statement,

and balance sheet.

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

\16\ The Office of Federal Housing Enterprise Oversight also

used the 10-year CMT as a component of their previously published

proposed risk-based capital rule. (See 64 FR 18083, April 13, 1999.)

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

Subject to our concurrence, Farmer Mac may use additional indexes,

such as the London Interbank Offer Rates (LIBOR), in the risk-based

capital test as long as the relationships between those indexes and the

10-year CMT are based on standard, widely used term structure modeling

relationships. Farmer Mac may use these relationships to compute the

cost of new debt, yields on investments, and coupon rates on mortgages

purchased or guaranteed by Farmer Mac. The interest rate index and rate

shock procedures described in the technical appendix are minimum

guidelines, and although Farmer Mac can use additional indexes, the

resulting risk-based capital level cannot be lower than it would be if

only the 10-year CMT were used.

E. Cashflows

Our spreadsheet based model projects cashflows from all of Farmer

Mac's assets, liabilities and off-balance sheet activities. Farmer Mac

may use its own internal cashflow generator system and programming for

this aspect of the risk-based capital stress test. However, Farmer Mac

must first obtain our concurrence for any internal cashflow generator

system and follow the procedures described in the technical appendix to

this regulation.

There are numerous modeling constructs and assumptions used in the

proposed cashflow component of the stress test. For the test,

investments are aggregated into the following categories: (i) Cash and

money market securities; (ii) commercial paper; (iii) certificates of

deposit; (iv) Agency mortgage-backed securities and collateralized

mortgage obligations; (v) and other investments. With our concurrence,

Farmer Mac is permitted to more finely disaggregate these categories.

Any new category deemed material to its operation in the future will

also be required to be added as a separate account. The level of

aggregation must appropriately reflect the contributions of revenues

and expenses of major program activities. For each asset class, we must

be able to discern the earnings rate and funding cost. Loan items

requiring separate accounts include: (i) Farmer Mac I program assets,

post-1996 Act; (ii) Farmer Mac I post-1996 Act swap balances; (iii)

Farmer Mac I pre-1996 Act loans; (iv) Farmer Mac I AgVantage

securities; (v) loans held for securitization; and (vi) Farmer Mac II

loans.

During the stress test, the balance sheet remains a constant size

over the 10-year period. This reflects a steady state scenario, meaning

that when on-balance sheet assets and liabilities and off-balance sheet

obligations amortize or pay-down, they are replaced with similar

assets, liabilities, and obligations. However, as discontinued loan

programs (e.g., pre-1996 Act Farmer Mac I program) amortize, they are

assumed to be replaced by current loan programs to more appropriately

reflect Farmer Mac's current operations.

We use effective years to maturity to simulate the amortization of

financial instruments, such as loans and other investments. A constant

prepayment rate (CPR) is used for all assets that have embedded

prepayment options. Together, the effective years to maturity and the

CPR are used to establish a roll-off rate and generate cashflows

reflecting a steady state over the stress

[[Page 61749]]

period. All cashflows and losses are computed on an annual basis.

To construct pro forma income statements for each period of the

stress test, it is necessary to establish rules and relationships for

deriving future income and expense items. Information from the first

period balance sheet is used in conjunction with the earnings and cost-

spread relationships from Farmer Mac supplied data to generate the

first period's income statement. In our spreadsheet model, each

investment account, loan item, and liability account can be specified

as either: (i) A fixed rate investment; or (ii) an instrument with a

fixed spread to Treasury with initial rates determined by actual data.

The specific spreads (weighted average yield less initial 10-year CMT)

by category are calculated directly in the stress test from the

weighted average yield data supplied by Farmer Mac in accordance with

the data requirements described in the technical appendix.

For non-interest income items, we follow certain decision rules for

generating earnings over time. For example, gains on agricultural

mortgage-backed security (AMBS) sales are a function of the amount of

new AMBS being issued. We based the relationship on historical

financial information. Expense items, such as fixed cost and variable

cost, are measured using a regression equation where operating expenses

is the dependent variable and the sum of investments and Farmer Mac

program assets held on-balance sheet is the independent variable. We

use the historical relationship of reserves to loan assets (that are

subject to reserves, post-1996 Act Farmer Mac I loan items, on- and

off-balance sheet) to simulate the loan reserves over time. The

corporate tax rate is estimated from actual Farmer Mac financial data.

Guarantee fee rates are obtained from actual guarantee fees charged by

Farmer Mac on its loan programs.

F. Financial Reports

Pro forma financial statements showing the resulting capital levels

for each period of the stress test are developed. Annual pro forma

balance sheets and income statements are generated for the stress

period using Farmer Mac's starting position, the stress conditions,

resulting cashflow outputs, and current operating strategies and

policies, as well as other assumptions. The proposed regulation

provides Farmer Mac with the option to use its own financial software

to produce the projected financial statements using the risk-based

capital stress test specifications and parameters described in the

technical appendix to this regulation. Projected financial statements

must comply, to the extent practical, with GAAP.

G. Capital Calculation

The risk-based capital stress test determines the amount of

starting capital Farmer Mac must hold to maintain a positive amount of

capital throughout the stress period using an iterative methodology.

Also, Farmer Mac must add on an additional 30 percent to this amount to

account for management and operational risk. Section 8.31(5) of the Act

defines regulatory capital as core capital (the par value of

outstanding common and preferred stock, paid-in capital, and retained

earnings), plus the allowance for losses and guarantee claims, as

determined in accordance with GAAP.

More specifically, to calculate the risk-based capital, our model

includes a section to solve for the minimum initial capital amount that

results in at least zero capital at the end of each period of the 10-

year stress test. In solving for initial capital, it is assumed that

reductions or additions to the initial capital accounts are made in the

retained earnings accounts and are balanced in the debt accounts at

levels proportionate to initial balances (same relative proportion of

long- and short-term debt as existing initial proportions). Because the

initial capital position affects the earnings, and hence capital

positions and appropriate discount rates through time, the initial and

future capital are simultaneously determined and must be solved using

an iterative process.

H. Future Changes to the Stress Test

Farmer Mac's performance over the stress period reflects its

current operating policies and other assumptions about its operations

to make the model functional. Due to significant data limitations

relating to a variety of issues, we were required to make a number of

simplifying assumptions. We recognize this may require us to revisit a

number of issues, particularly as more data become available from

Farmer Mac's own operations. Therefore, we will continually monitor the

risk-based capital stress test results and consider whether

modifications to the risk-based capital stress test are warranted. In

particular, we anticipate that as more data from agricultural mortgage

losses, especially those loans currently securitized by the Corporation

become available, changes may be required in the risk-based capital

stress test through amendment of this subpart.

Through our ongoing evaluations of the risk-based capital stress

test, we also may find it necessary to make technical modifications to

the risk-based capital stress test procedures. If we modify the

procedures for implementing the risk-based capital stress test, we will

notify Farmer Mac and provide them with written instructions to

implement the changes. We will make these modifications available to

the public on a quarterly basis on our web site.

VI. Statistical Properties of the Default Equation

This section provides further details about the credit risk

component, including the underlying theory, analytical methods, data,

model specifications, econometric estimation results, and conformance

with the worst-case conditions specified in the statute. This section

is intended for readers who desire further information on the

measurement of credit risk and the statistical properties of the

default estimation equation.

A. Estimation Methods

Historic time series on the frequency and severity of losses on

farm mortgage loans are compiled from available loan-level data.\17\

The measures of frequency and severity are related to selected loan-

level characteristics and macroeconomic conditions through

appropriately specified regression equations in order to account

explicitly for the collective effects of these characteristics on

frequency and severity of loss. The resulting regression equations are

applied to estimate Farmer Mac's future credit risk position by

substituting the respective values of their loan level characteristics

and macroeconomic (farmland value changes) measures into the estimated

regression equations, calculating the results, and determining the

performance implications.

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

\17\ As previously discussed, the historic loan data was

obtained from FCBT.

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

Several estimation approaches are possible, although the ultimate

choice depends on the degree of conformance between the characteristics

of the data and the properties of the respective regression methods. In

the case of frequency of default and loan loss occurrence, the loan

outcome is a qualitative, binary variable--default and loss either

occur or do not occur. Therefore, the appropriate regression procedure

must also accommodate qualitative characteristics (e.g., loan default

and loss is coded with a value of 1, while successful loan performance

is coded with a value of 0).

[[Page 61750]]

Regression approaches with qualitative dependent variables include

the linear probability model, logit, and probit. The linear probability

model has a number of shortcomings and is rarely used. Under the linear

probability model, estimates can occur outside the 0-1 interval

resulting in nonsense probabilities and negative variances. Logit and

probit are the most commonly used approaches. The primary difference

between the two approaches is the assumed underlying probability

distribution. The probit model assumes a normal distribution while the

logit model uses the logistic distribution. Logit is used more

frequently in modeling loan defaults, and is utilized in the credit

risk component of the stress test.

In the case of severity of loss, the resulting magnitude of loan

loss does not have a qualitative characteristic. Rather, magnitude of

loss occurs in a continuous form, bounded at zero, thus requiring a

different modeling approach.

The two equations for frequency and severity could be estimated

independently. If they are estimated independently, but are in fact

related, inconsistent estimates result. A method to accommodate

possible dependence is to use a Heckman two-step approach that first

assesses the probability of default then subsequently estimates the

level of loss based on similar or different covariates. Accordingly,

Heckman's two-step or ``Heckit'' estimation method was also explored.

The Heckman two-step approach is estimated using method of moments

techniques that results in consistent estimates.\18\ Basically, it is a

discrete model estimated on the basis of sample selection criteria.

Then, a linear regression is performed in the second step. Examples of

the two-step estimation procedure include Miller, Barry, Ellinger, and

Lajili and Nakosteen and Zimmer.\19\ The method utilizes an asymptotic

covariance for the two-step estimation and results in a consistent

estimator for variance (e2). The system

specification is appropriate in this case because of the relationship

between the two equations. That is, severity of loss only occurs on

defaulted loans. Default is required in order for loss to occur. While

we explored this approach, the two-step procedure did not yield

significant results for estimating a loss severity equation. Severity

was not found to vary systematically and considered constant across the

tested loan characteristics and lending conditions. Therefore, the

simple weighted average by loss volume of 20.9 percent is used in the

stress test.

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

\18\ Greene, W. A. Economic Analysis, 3rd ed., Prentice Hall,

1997.

\19\ Miller, L.P.N. Ellinger, P.J. Barry, and K. Lajili. ``Price

and Non-Price Management of Agricultural Credit Risk,'' Agricultural

Finance Review, 53 (1993): 28-41. Nakosteen, R. and M. Zimmer.

``Migration and Income: The Question of Self-Selection,'' Southern

Economic Journal, 7 (1980): 840-851.

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

Due to Farmer Mac's relatively short history, its own loan-level

data are insufficiently developed for use in estimating default

frequency and loss severity equations. In the future, however,

expansions in both the scope and historic length of Farmer Mac's

lending operations likely will warrant use of its data in estimating

the regression equations.

B. Model Specifications

Agricultural credit and residential mortgage literature suggest

several independent financial variables to consider in modeling loan

default. These include the D/A ratio, LTV ratio, DSCR, age of the loan,

payment burden, interest rate changes, land price changes, net farm

income and changes in commodity prices. These variables were each

considered in modeling the default experiences of FCBT loans. Standard

goodness-of-fit measures and the credibility of outcomes were used to

select the final equation used to estimate the loss probabilities.

The FCBT farm real estate loans were included in the estimating

data if they satisfied at least three of four underwriting standards

currently utilized by Farmer Mac. The four standards specify that: (1)

The D/A ratio must be less than or equal to 0.50, (2) the LTV ratio

must be less than or equal to 0.70,(3) the DSCR must exceed 1.25, (4)

and the current ratio must exceed 1.0. Farmer Mac may waive complete

compliance with these standards if a loan is judged to have appropriate

offsetting strengths. Accordingly, the approach employed in the 1998

study requires that loans satisfy at least three of the four specified

standards.

Furthermore, the D/A and LTV ratios were restricted to be less than

or equal to 0.85. It is unlikely that Farmer Mac would waive these

standards if the ratios exceeded these values. Inspection of a portion

of Farmer Mac's loan portfolio indicated several instances where the D/

A and LTV ratios exceed .50 and .70, respectively, with values of both

ratios rarely exceeding 0.85. In the Farmer Mac data, 3.3 percent of

the loans and 3.1 percent of the current outstanding loan balances have

LTV ratios exceeding 0.70. The use of the maximum values for LTV and D/

A and the three-out-of-four standards requirement for passing standards

is intended to emulate Farmer Mac's underwriting standards, and

includes the practice of waiving selected standards.

Several limitations in the FCBT loan-level data affect construction

of the default function. The data contained loans that were originated

between 1979 and 1992, but there were virtually no losses during the

early parts of the sample period. As a result, losses attributable to

specific loans are only available from 1986 through 1992. In addition,

no prepayment information was available in the data.

The data set used for estimation also includes loans that were re-

amortized, paid in full, or merged with a new loan as performing loans.

Including these loans may lead to an underestimation of defaults, if

some of the re-amortized, paid, or merged loans default and incur

losses. In contrast, when the loans that are re-amortized, paid in full

or merged are excluded from the analysis, the default rates are

overestimated if a higher proportion of loans that are re-amortized,

paid in full, or combined (merged) into a new loan are non-default

loans compared to live loans. Excluding loans with defaults, 11,527

loans were active and 7,515 loans were paid in full, re-amortized or

merged as of 1992. Application of a t-test \20\ for differences in the

means for these two groups indicated that active loans had

significantly higher D/A and LTV ratios, and lower current ratios than

other loans. These results indicate that, on average, active loans have

potentially higher risk than loans that were re-amortized, paid in

full, or merged.

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

\20\ The t-test used evaluates the hypothesis that the means

from the two loan groups are statistically different. The t-test

uses a statistic derived from the student distribution.

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

C. Estimation Results

From a statistical perspective, models utilizing information based

on origination information and subsequent economic information were

consistently more reliable than models using loans that are transformed

into multiple observations.

The structure of the historical FCBT data supports estimation of

defaults based on origination information and economic conditions.

Under an origination year approach each observation is used only once

in estimating loan default. The underwriting variables at origination

and economic factors that occur over the life of the loan are used to

estimate loan default.

The final estimated equation for loss frequency is:

[[Page 61751]]

[GRAPHIC] [TIFF OMITTED] TP12NO99.000

where p is the probability that a loan defaults and has positive losses

(Pr(Y=1|x)); B0 to B5 are the estimated

coefficients for the intercept and variables X1 to

X5; X1 is the LTV ratio at loan origination

raised to the power 5.38027; \21\ X2 is the largest annual

percentage decline in FCBT farmland values during the life of the loan

discounted by 4.8 percent per year; \22\ X3 is the DSCR at

loan origination; X4 is the original loan balance in 1997

dollars; and X5 is the D/A ratio at loan origination. It is

commonly accepted that farmland values at any point in time reflect the

discounted present value of expected returns to the land.\23\ Thus,

changes in land values, as expressed in the default equation, represent

the combined effects of the level and growth rates of farm income,

interest rates, and inflationary expectations--each of which is

accounted for in the discounted, present value process.

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

\21\ Loss probability is likely to be more sensitive to changes

in LTV at higher values of LTV. To test and implement this non-

linearity, the model was first estimated with 8 dummy variables at

LTV intervals of 0 to 0.399, 0.400 to 0.499, 0.500 to 0.599, 0.650

to 0.699, 0.700 to 0.749, 0.750 to 0.799 and 0.800 to 0.850. A power

functional form for the LTV variable was fit to replace the

individual dummy variables. The result using generalized least

squares (GLS) was LTV 5.38027. The default equation is

re-estimated with the power function. The power function increases

the degrees of freedom for the model and provides a continuous

relationship between LTV and defaults.

\22\ Discounting reflects the declining effect that the maximum

land value decline has on the probability of default when it occurs

later in a loan's life. The value of 4.8 percent was determined by

iteratively solving the default equation with the default equation

dummy variable ranging from 0 percent to 10 percent. The 4.8 percent

rate yielded the highest goodness-of-fit values.

\23\ Barry, P. J., P. N. Ellinger, J. A. Hopkin, and C. B.

Baker. Financial Management in Agriculture, 5th ed., Interstate

Publishers, 1995.

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

These variables have logical relationships to the incidence of loan

default and loss, as evidenced by the findings of numerous credit

scoring studies in agricultural finance.\24\ Each of these anticipated

directions of relationship signifies greater risk for the borrower, and

thus greater credit risk and incidence of loan loss for the lender. The

frequency of loan default was found to differ significantly across all

of the loan characteristics and lending conditions, as indicated by the

results of the logit equation. The estimated logit coefficients and p-

values are:

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

\24\ Splett, N.S., P. J. Barry, B. Dixon, and P. Ellinger. ``A

Joint Experience and Statistical Approach to Credit Scoring,''

Agricultural Finance Review, 54 (1994):39-54.

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

Coefficients p-value

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

Intercept.................................... -9.7267 0.0001

X1: LTV...................................... 2.7337 0.0001

X2: Max farmland value decline............... -0.3138 0.0001

X3: DSCR..................................... -0.1822 0.0003

X4: Loan size................................ 8.222E-7 0.0001

X5: D/A ratio................................ 2.3229 0.0001

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

The low p-values on each coefficient indicate a highly significant

relationship between loan default and the respective independent

variables. Other goodness-of-fit indicators are:

Hosmer and Lemeshow goodness-of-fit p-value--0.2232

Max-rescaled R2--0.1204

Concordant--79.4%

Disconcordant--16.5%

Tied--4.1%

Explanations of these indicators follow:

1. Hosmer and Lemeshow Goodness-of-Fit Test divides subjects into

deciles based on predicted probabilities, then computes a chi-square

test statistic from observed and expected frequencies. A probability

(p) value is computed from the chi-square distribution with 8 degrees

of freedom to test the fit of the logistic model. If the Hosmer and

Lemeshow goodness-of-fit test statistic is .05 or less, the null

hypothesis that there is no difference between the observed and

predicted values of the dependent is rejected. If it is greater, the

null hypothesis of no difference is not rejected, implying that the

model's estimates fit that data at an acceptable level. This result

does not, however, indicate that the model necessarily explains much of

the variance in the dependent variable. Because the p-value of 0.22 is

greater than 0.05, the null hypothesis of no difference between the

observed and predicted values cannot be rejected. No other information

about the default equation's goodness-of-fit is provided by this

statistic.

2. Max-rescaled R-squared. Several measures often are used to

develop R-squared measures with logistic regression. The logistic

measures do not specifically measure the degree of variation explained

by the model. However, the measures often are scaled from 0 to 1.0 to

provide a relative index of degree of fit. The one reported here is a

modification of the Cox and Snell coefficient that compares the

likelihood function with the intercept only with the likelihood

function with all the variables. Nagelkerke proposed normalizing the

value between 0 and 1 by dividing by the likelihood function with

intercepts only. The specific formula is:

[GRAPHIC] [TIFF OMITTED] TP12NO99.001

where L(0) is the likelihood of the intercept-only model, L(()

is the likelihood of the specified model and n is the sample size. The

quantity R\2\ achieves a maximum of less than 1 for discrete models,

where the maximum is given by:

[GRAPHIC] [TIFF OMITTED] TP12NO99.002

The Nagelkerke adjustment to normalize the value between 0 and 1

is:

[GRAPHIC] [TIFF OMITTED] TP12NO99.003

While the default equation has strong statistical significance, the

Max-rescaled R-squared value of 0.1204 indicates that other variables

and factors not included in the default equation may also influence

default rates. Limitations on availability and quality of data,

however, restrict the access to, and use of, other variables.

Other statistical measures that are indicative of a model's

performance for correctly estimating the probability of loan default,

include concordant, disconcordant, and ties. Generally, model

performance is superior when the concordant measure is high and the

other two measures are low. Each measure is discussed below.

3. Concordant. The predicted values for each possible pair of non-

defaulted and defaulted loans are compared. The number of possible

pairs is equal to the number of non-default loans times the number of

defaulted loans. The percent of pairs that have defaulted loans with

higher predicted default rates than predicted default rates for non-

defaulted loans are included in the concordant category. Given all

possible non-default/default pairwise combinations, the concordant

percentage is the proportion that has defaulted loans with a higher

predicted percentage than non-default loans. The concordant value for

the default equation of 79.4 percent

[[Page 61752]]

indicates a relatively high incidence of correct rankings for the

possible pairs of defaulted and non-defaulted loans, when the estimated

coefficients of the default equation are used to estimate default

rates.

4. Disconcordant. The disconcordant value is the proportion of

pairwise estimates that have higher predicted default rates for non-

default loans than defaulted loans. The discordant value of 16.5

percent indicates a relatively low incidence of incorrect rankings of

default and non-default loans.

5. Ties. The proportion of pairwise estimated probabilities that

are equal between non-default and default is 4.1 percent, which is

relatively low.

D. Comparison of Actual to Predicted Losses

We compared the actual and predicted loss rates based on

origination date and the 20.9 percent severity rate applied to all FCBT

loans for the years 1979 to 1992. The largest discrepancies between the

series occur on loans originated in 1984 and 1987. A problem associated

with errors on specific loans is the application of an average severity

value. Using an average severity rate underestimates losses on specific

loans that have actual severity rates exceeding 20.9 percent. Using the

average severity rate restricts the maximum estimated loss on any loan

in the portfolio to 20.9 percent of the origination loan balance.

Application of the estimated loss equations to the FCBT data

results in total estimated loss over the entire sample period equal to

$9,417,704. Actual losses incurred total $9,805,472. The average of the

predicted loss rates is 0.48 percent while the average of the actual

default rates during 1979-92 is 0.50 percent. The maximum 1- and 2-year

loss rates are 1.54 percent and 2.17 percent in 1985 and 1984-85,

respectively. The maximum 1- and 2-year loss rates estimated by the

model are 1.20 percent in 1984 and 1.85 percent in 1984-85.

VII. Sensitivity of Risk-Based Capital Requirement

The stress test is responsive to changes in the risk profile

inherent in Farmer Mac's financial positions. The stress test requires

higher levels of risk-based capital when Farmer Mac's risk levels

increase and a lower requirement when risk levels decrease. Risk

increases or decreases when Farmer Mac modifies its loan underwriting

standards and/or interest rate risk exposure through various funding

and hedging strategies. In addition, the mix and volume of assets and

liabilities, both on- and off-balance sheet, affect risk levels as does

the initial market interest rate used in the stress test. Some assets

such as high quality investments, Farmer Mac II program mortgages,

AgVantage, and Farmer Mac I pre-1996 Act mortgages present little or no

loss exposure (lower credit risk exposure assets), while other assets

such as Farmer Mac I post-1996 Act mortgages present greater levels of

credit risk (higher credit risk exposure assets).

We evaluated the sensitivity of the stress test using two different

initial financial positions. Financial position one is consistent with

Farmer Mac's current business activities and risk profile. Financial

position two is a hypothetical portrayal of Farmer Mac as a more mature

business. For this scenario, we increased Farmer Mac's size, business

activities, and risk profile. We specifically designed financial

position two to evaluate the sensitivity of the stress test assuming

additional growth in Farmer Mac program I assets. The characteristics

of financial position one and two are as follows.

Table 1. Financial Positions Used in Performing the Sensitivity Analysis

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

Financial Financial

Financial Component (in millions) Position Position

1 2

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

Assets............................................ $2,566 $3,206

Liabilities....................................... 2,481 3,095

Capital........................................... 84 111

Off-Balance Sheet Assets..........................

Overall Portfolio Characteristics 828 3,187

Lower Credit Risk Exposure Assets................. 1,931 2,133

Higher Credit Risk Exposure Assets................ 1,459 4,260

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

We used these two hypothetical financial positions as our initial

starting positions. For each initial position, we calculated a ``base''

case risk-based capital requirement. We then increased or decreased

Farmer Mac's risk levels by varying:

Mortgage factors that influence loss performance (D/A

ratio, LTV ratio, DSCR, loan size, and loan age);

Interest rate risk exposure as measured by Farmer Mac in a

market value framework;

The initial interest rate environment;

Spread relationships of interest earning assets to

interest rate index used in the stress test; and

Guarantee fee charged by Farmer Mac.

We then recalculated the risk-based capital requirement for each

varied condition and compared the results to the ``base'' case. The

results of this analysis follow.

A. Sensitivity to Changes in Mortgage Risk Factors

The stress test calculates increases in the risk-based capital

requirement when the risk increases in Farmer Mac's mortgage portfolio

of held and guaranteed loans. We found that, if Farmer Mac increases

risk by loosening origination loan underwriting standards, the stress

test calculates a higher capital requirement. Conversely, if Farmer Mac

tightens its underwriting standards, the stress test calculates a lower

capital level. As shown in the following table, the stress test

consistently produces these results when mortgage characteristics are

changed individually or on a combined basis.

Table 2.--Changes in Risk-Based Capital Requirements for Changes in

Mortgage Characteristics

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

Risk-Based Capital Requirement

Sensitivity Cases (in -------------------------------------------

millions) Financial Position Financial Position

1 2

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

1. Base Case................ $29.5 $43.2

2. Origination D/A Ratios 38.3 65.8

Increase...................

3. Origination LTV Ratios 37.2 63.1

Increase...................

4. Origination DSCR Decrease 30.3 45.3

5. Origination Loan Size 65.2 141.6

Increases..................

[[Page 61753]]

6. Increases Stated 2 to 5 109.5 266.9

Above Occur Simultaneously.

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

The mortgage factors were increased from the base case on a loan-

by-loan basis to increase risk levels in Farmer Mac's current

portfolio. In each case, the increase in a mortgage factor was limited

to the maximum permitted under Farmer Mac's underwriting standards or

the unadjusted existing loan origination value, whichever was greater.

We used the existing origination values in Farmer Mac's current

portfolio as our starting point and then increased and decreased

individual loan underwriting ratios to perform our sensitivity testing.

The sensitivity tests are:

1. Base case;

2. D/A ratio for individual loans was increased 50 percent

resulting in an increase in the portfolio-weighted average ratio to 56

percent from 37 percent;

3. LTV ratio for individual loans was increased 25 percent

resulting in an increase in the portfolio-weighted average ratio to 70

percent from 56 percent with the maximum individual loan increase

capped at 85 percent;

4. DSCR for individual loans was decreased 25 percent resulting in

a decrease in the portfolio-weighted average ratio to 1.26 from 1.71;

5. Origination size for each loan in Farmer Mac's current portfolio

was doubled resulting in an increase in the portfolio average to $956

thousand from $478 thousand with the maximum individual increase capped

at $3.49 million; and

6. All increases stated in tests 2 to 5 occurring simultaneously.

Loan age affects the level of risk-based capital required by the

stress test. Older loans represent lower credit risk and, therefore,

reduce the risk-based capital requirement while the opposite is true

for new loans. We evaluated how the capital requirement changes for an

increase in loan age of 1 year. The results show a reduced risk-based

capital requirement from the base case as follows:

Table 3.--Changes in Risk-Based Capital Requirements for Changes in Loan

Age

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

Risk-Based Capital Requirement

Sensitivity Cases (in -------------------------------------------

millions) Financial Position Financial Position

1 2

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

1. Base Case................ $29.5 $43.2

2. Loan Age Increases by 1 26.9 35.8

year.......................

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

B. Sensitivity to Changes in Interest Rate Risk Exposure and the

Initial Rate Environment

The stress test requires Farmer Mac to hold more capital as it

increases its interest rate risk exposure and less capital as it

decreases exposure. The stress test uses Farmer Mac's market value

measurement of interest rate risk to quantify the effects that changes

in interest rates have on risk-based capital. Farmer Mac can change its

market value exposure by varying its funding, asset holdings, and

hedging strategies. We evaluated the effect on the risk-based capital

requirement if Farmer Mac pursues strategies that either increase or

decrease its interest risk exposure as measured by the market value

methodology. For the increase in interest rate risk scenario, we assume

Farmer Mac doubles its interest rate risk exposure. In this scenario, a

277 bp movement in interest rates caused the loss to capital to

increase by $25.1 million compared to the base case. We also evaluated

the situation where Farmer Mac's interest rate risk exposure declines

50 percent from the base case. The results of our sensitivity tests are

summarized below.

Table 4.--Changes in Risk-Based Capital Requirements for Changes in

Interest Rate Risk Exposure

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

Risk-Based Capital

Requirement

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

Sensitivity Cases (in millions) Financial Financial

Position Position

1 2

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

1. Base Case...................................... $29.5 $43.2

2. IRR Exposure Increases......................... 54.7 74.9

3. IRR Exposure Decreases......................... 16.9 27.3

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

The interest rate environment affects stress test results. When

interest rates are low, the rate change used in the stress test is

relatively small compared to when interest rates are high. Clearly,

interest rates can change by a greater degree when they are high

compared to when they are low. In addition, the large changes in

interest rates expose Farmer Mac to greater risk. The stress test,

therefore, requires higher risk-based capital in rate environments

where interest rates are high relative to low rate environments as

indicated in the following table:

[[Page 61754]]

Table 5.--Changes in Risk-Based Capital Requirements for Different Initial Rates

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

Risk-Based Capital

Requirement

Sensitivity Cases (in millions) Initial Rate -------------------------------

(percent) Financial Financial

Position 1 Position 2

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

1. Base Case.................................................... 5.54 $29.5 $43.2

2. Higher Initial Rate.......................................... 11.08 62.6 75.8

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

C. Sensitivity to Changes in Spread Relationships and Guarantee Fees

The stress test requires higher risk-based capital when earnings

are under pressure from a tightening in spreads on interest earning

assets or a reduction in guarantee fees charged by Farmer Mac. On the

other hand, the risk-based capital requirement would be lower when

yield spreads widen or Farmer Mac increases its guarantee fees. The

stress test incorporates earnings when calculating risk-based capital.

We evaluated the sensitivity of the stress test for decreases in

spreads on interest earning assets of 5 bp and 10 bp. The stress test

uses current spreads (i.e., the difference in current yields and the

interest rate index used in the model) to determine asset yields when

interest rates are changed. Therefore, a tightening in spreads will

reduce asset yields used to generate earnings. We also evaluated stress

test results assuming Farmer Mac reduced guarantee fees currently

charged by half. The stress test calculated a higher risk-based capital

requirement under diminished earnings capacity as follows:

Table 6.--Changes in Risk-Based Capital Requirements for Changes in

Earning Spreads and Guarantee Fees

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

Risk-Based Capital

Requirement

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

Sensitivity Cases (in millions) Financial Financial

Position Position

1 2

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

1. Base Case...................................... $29.5 $43.2

2. Spread Tighten by 5 bp......................... 31.0 44.2

3. Spread Tighten by 10 bp........................ 33.8 45.2

4. Guarantee Fee Decrease......................... 38.4 59.6

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

VIII. Impact of the Risk-Based Capital Stress Test on Farmer Mac

The impact of the stress test depends on Farmer Mac's risk profile

and starting capital position. High-risk assets and unhedged interest

rate risk will result in larger risk-based capital requirements.

Conversely, if Farmer Mac maintains a low risk profile, the stress test

will produce a low capital requirement. Given Farmer Mac's current

financial position and risk profile, the proposed stress test would not

require Farmer Mac to increase its capital. The risk-based capital

requirement for Farmer Mac produced by the proposed stress test is

below the statutory minimum and critical capital standards.

Furthermore, Farmer Mac's current capital level exceeds both the

statutory minimum and critical capital standards. We emphasize that

this result is only based on Farmer Mac's current financial position

and risk profile. If Farmer Mac accepts more risk as it grows into a

mature business, the risk-based capital requirement could exceed the

statutory minimum and critical capital standards as well as current

capital level. In such a situation there are several options available

to Farmer Mac, including:

Issue additional stock,

Increase guarantee fees to build earnings and capital,

Reduce credit risk through modifications to loan

underwriting standards or obtain credit enhancements,

Mitigate interest rate risk through funding and hedging

strategies.

IX. Reporting Requirements

Proposed Secs. 650.25 and 650.26 outline Farmer Mac's basic

responsibilities for determining its risk-based capital level and

reporting the results to us. Farmer Mac must determine its risk-based

capital level in accordance with the procedures in Sec. 650.24 and the

technical appendix of the subpart. Farmer Mac must at all times

maintain compliance with the risk-based capital levels established by

the risk-based capital stress test and must be able to determine its

risk-based capital level at any time. If, at any time, the risk-based

capital level computed using the risk-based capital stress test

procedures is less than the minimum capital requirements set forth in

section 8.33 of the Act, Farmer Mac must maintain the statutory minimum

capital level.

Proposed Sec. 650.26 requires Farmer Mac to determine its risk-

based capital level at least quarterly. However, changing circumstances

that may have a significant effect on capital may necessitate that

Farmer Mac determine its risk-based capital level more frequently than

quarterly. For example, we may require the Corporation to determine its

risk-based capital level and report the results to us more frequently

than quarterly if:

1. The Corporation is receiving special supervisory attention;

2. The Corporation has, or is expected to have, losses resulting in

capital depletion;

3. The Corporation has significant exposure due to operational

risk, the risks from concentrations of credit, certain risks arising

from other products, services, or related activities, or management's

overall inability to monitor and control financial risks;

4. The Corporation is exposed to a high volume of, or particularly

severe, problem loans;

5. The Corporation is growing rapidly;

6. The Corporation may be adversely affected by the activities or

the condition of other institutions with which it has significant

business relationships or in which it has significant investments; or

7. The Corporation has significant exposure to declines in net

income or in the market value of its capital due to a change in

interest rates and/or the exercise of embedded or explicit options.

In addition, if Farmer Mac anticipates entering into any new

business activity that could have a significant effect on capital, it

must determine a pro forma risk-based capital level that includes the

new business activity. Farmer Mac must provide the pro forma

determination to us 10 days prior to implementation of the new business

program. Proposed Secs. 650.27 and 650.28 provide further instructions

on how and when to report the risk-based capital level.

X. Business and Capital Plans

Well-conceived strategic and operational business and capital plans

promote safety and soundness and are essential ingredients in meeting

institutional objectives. The process of identifying, measuring and

controlling

[[Page 61755]]

an institution's risks and the resulting capital requirements starts

with the development of the institution's goals and objectives. Such

goals and objectives should identify the direction in which an

institution wants to proceed, its stated mission, business structure,

and how it intends to achieve its stated goals.

We expect that any strategic and operational business and capital

plans will address the long-term purpose and mission of the business.

In addition, we believe that such plans should include quantifiable

goals and objectives, and recognize and discuss internal and external

factors that are likely to influence the future operations of the

business. We also expect that the strategic planning process will

include an appropriate capital adequacy plan.

Proposed Sec. 650.22 sets forth the responsibilities of the

Corporation's board to ensure that the Corporation maintains its

capital at a level that is sufficient to sustain continued financial

viability and provide for growth. The Board must take appropriate

measures so that the Corporation's capital is not only adequate to meet

formal regulatory standards, but is also sufficient to support the

Corporation's business objectives and strategies. This requires the

Board to set explicit goals for capitalization with respect to risk and

return objectives. The capital adequacy target levels should be part of

the Corporation's internal process for evaluating capital adequacy. The

Board should annually review and approve the Corporation's capital

adequacy target and composition of capital.

Proposed Sec. 650.22(b) requires the Board to adopt a 3-year

strategic and operational business plan. The plan must contain the

elements of both a basic strategic and operational business plan as

well as a capital adequacy plan. Among other items listed in proposed

Sec. 650.22(b), the capital adequacy plan must include any projected

dividends, equity retirements, or other action that may decrease the

Corporation's capital. The Board should also consider other relevant

factors that may affect Farmer Mac's capital adequacy, such as the

capability of management to measure, manage, and control risk, the

development of new lines of business or Farmer Mac's continued ability

to access the market at favorable rates.

XI. Supervision and Notification

Section 8.35(a) of the Act describes the various levels (I-IV) of

enforcement under which the Corporation will be classified by the OSMO

Director. Proposed Sec. 650.29 establishes the regulatory procedure for

the OSMO Director to notify Farmer Mac of a determination that it is

not meeting the risk-based capital level calculated by the Corporation

as required by Sec. 650.23 or the minimum or critical capital

requirements specified by sections 8.33 and 8.34 of the Act. Proposed

Sec. 650.29 provides for the submission of a capital restoration plan,

as appropriate, when it has been determined that the Corporation is not

meeting the required capital levels.

List of Subjects in 12 CFR Part 650

Agriculture, Banks, banking, Conflicts of interest, Rural areas.

For the reasons stated in the preamble, part 650 of chapter VI,

title 12 of the Code of Federal Regulations is proposed to be amended

to read as follows:

PART 650--FEDERAL AGRICULTURAL MORTGAGE CORPORATION

1. The authority citation for part 650 is revised 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.

2. Subpart B is added to read as follows:

Subpart B--Risk-Based Capital Requirements

Sec.

650.20 Definitions.

650.21 General.

650.22 Corporation board of directors guidelines.

650.23 Risk-based capital stress test.

650.24 Risk-based capital level.

650.25 Your responsibility for determining the risk-based capital

level.

650.26 When you must determine the risk-based capital level.

650.27 When to report the risk-based capital level.

650.28 How to report your risk-based capital determination.

650.29 Failure to meet capital requirements.

650.30 Effective date for compliance with regulation.

650.31 Audit of the risk-based capital stress test.

Appendix A to Subpart B. of Part 650--Risk-Based Capital Stress

Tests.

Sec. 650.20 Definitions.

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

(a) Farmer Mac, Corporation, you, and your means the Federal

Agricultural Mortgage Corporation and its affiliates as defined in

subpart A of this part.

(b) Our, us or we means the Farm Credit Administration.

(c) 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 the

Corporation in excess of the par value of stock;

(4) Retained earnings; and

(5) Any allowances for losses on loans and guaranteed securities.

(d) Risk-based capital means the amount of regulatory capital

sufficient for the Corporation to maintain positive capital during a

10-year period of stressful conditions as determined by the risk-based

capital stress test described in Sec. 650.23.

Sec. 650.21 General.

You must hold risk-based capital in an amount determined in

accordance with this subpart.

Sec. 650.22 Corporation board of directors 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 30 days after the beginning of each calendar

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) Pro forma financial statements for each year of the plan;

(5) A detailed operating budget for the first year of the plan;

and,

(6) 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

[[Page 61756]]

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. 650.23 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 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 estimate 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) and the applicable Internal Revenue Service

corporate income tax schedule, as further described in the technical

appendix.

(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 bp 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 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 the technical appendix to this subpart

and available electronically at www.fca.gov. You may also use any

reliable 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 the

Corporation'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 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 www.fca.gov.

Sec. 650.24 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. 650.23.

(b) Management and operations risk. Thirty (30) percent of the

amount of risk-based capital determined by the risk-based capital test

in Sec. 650.23.

Sec. 650.25 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

[[Page 61757]]

prescribed in Sec. 650.28. At any time, however, we may determine your

risk-based capital level using the procedures in Sec. 650.23 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. 650.26 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. 650.27 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. 650.26.

(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. 650.24.

(c) If you make any changes to the data used to calculate your

risk-based capital requirement that causes 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. 650.28 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 media as we require.

Sec. 650.29 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. 650.23, 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. 650.30 Effective date for compliance with regulation.

For the 12-month period beginning on the effective date of this

regulation, you must determine a risk-based capital level by

implementing the risk-based capital stress test as described in

Sec. 650.23 and appendix A to this subpart, and must report the results

to us as described in Sec. 650.28. During this 12-month period, you

will not be required to maintain capital at the risk-based capital

level, but you must maintain your minimum capital level as prescribed

in section 8.33 of the Act. Beginning on the day following the 12-month

period, you must comply with all provisions of this subpart.

Sec. 650.31 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 to Subpart B of Part 650--Risk-Based Capital Stress

Tests

1.0 Introduction.

2.0 Credit Risk.

2.1 Loss Frequency and Severity Models.

2.2 Loan Seasoning Adjustment.

2.3 Example Calculation of Dollar Loss on One Loan.

2.4 Treatment of Long-term Standby Purchase Commitments.

2.5 Calculation of Loss Rates for Use in the Stress Test.

3.0 Interest Rate Risk.

3.1 Process for Calculating the Interest Rate Movement.

4.0 Elements Used in Generating Cashflows.

4.1 Data Inputs.

4.2 Assumptions and Relationships.

4.3 Risk Measures.

4.4 Loan and Cashflow Accounts.

4.5 Income Statements.

4.6 Balance Sheets.

4.7 Capital.

5.0 Capital Calculations.

5.1 Method of Calculation.

1.0 Introduction

a. This technical appendix provides details about the risk-based

capital stress test (stress test) for Farmer Mac. The stress test is

a deterministic portrayal of Farmer Mac's annual capital needs for

10 years. The stress test calculates the risk-based capital level

required by statute under stipulated conditions of credit risk and

interest rate risk. The stress test uses loan-level data from Farmer

Mac's agricultural mortgage portfolio, as well as quarterly Call

Report and related information to generate pro forma financial

statements and calculate a risk-based capital requirement. The

stress test also uses historic agricultural real estate mortgage

performance data, relevant economic variables, and other inputs in

its calculations.

b. The key components of the stress test include the specifications

of credit risk, interest rate risk, the cashflow generator, and the

capital calculation. Linkages among the components ensure that the

measures of credit and interest rate risk pass into the cashflow

generator. The linkages also transfer cashflows through the financial

statements to represent values of assets, liabilities, and equity

capital. We designed the 10-year projection to reflect a steady state

in the scope and composition of Farmer Mac's assets. This technical

appendix provides details about the credit risk, interest rate risk,

cashflow generator, and capital components of the stress test.

2.0 Credit Risk

Computing credit risk requires loan loss rates. We determined

loan loss rates by applying loss frequency and severity equations to

Farmer Mac loan-level data. From these equations, we calculated loan

losses under stressful economic conditions and loss rates assuming

Farmer Mac's portfolio remains at a ``steady state.'' Steady state

assumes the underlying characteristics and, therefore, risks of

Farmer Mac's

[[Page 61758]]

portfolio remain constant over the 10 years of the stress test. From

estimated dollar losses, we computed loss rates for use in the

stress test. The loan volume subject to loss throughout the stress

test is then multiplied by the loss rate. Lastly, the stress test

allocates losses to each of the 10 years assuming a time pattern for

loss occurrence as discussed in section 4.3 of this appendix

entitled Risk Measures.

2.1 Loss Frequency and Severity Models

a. We modeled credit risk using historical time series loan-

level data to measure the frequency and severity of losses on

agricultural mortgage loans. The model relates frequency and

severity to loan-level characteristics and economic conditions

through appropriately specified regression equations in order to

account explicitly for the collective effects of these

characteristics on loan losses. We can then estimate loan losses for

Farmer Mac with the resulting regression equations by substituting

the respective values of Farmer Mac's loan-level data and using a

stressful economic input.

b. The loss frequency and severity equations were estimated from

historical agricultural real estate mortgage loan data from the Farm

Credit Bank of Texas (FCBT). To estimate the equations, the data

used included FCBT loans if they satisfied at least three of four

underwriting standards Farmer Mac currently uses (estimation data).

The final estimated equation for loss frequency is:

[GRAPHIC] [TIFF OMITTED] TP12NO99.004

Where:

p is the probability that a loan defaults and has positive

losses (Pr (Y=1|x)),

X1 is the loan-to-value ratio (LTV) at loan

origination raised to the power 5.38027,1

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

\1\ To test and implement the non-linear relationship between

loss probability and LTV, the model was first estimated with 8 dummy

variables at LTV intervals of 0 to 0.399, 0.400 to 0.499, 0.500 to

0.599, 0.650 to 0.699, 0.700 to 0.749, 0.750 to 0.799 and 0.800 to

0.850. Using generalized least squares, a power function of LTV

5.38027 was fit to replace the individual dummy variables, and the

equation was re-estimated. The power function increases the degrees

of freedom for the model and provides a continuous relationship

between LTV and defaults.

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

X2 is the annual percentage decline in farmland

values during the life of the loan discounted by 4.8 percent per

year, 2

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

\2\ We determined the 4.8 percent by iteratively solving the

default equation using dummy variables ranging from 0 percent to 10

percent. The 4.8-percent rate yielded the highest goodness-of-fit

values.

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

X3 is the DSCR at loan origination,

X4 is the origination loan balance stated in

1997 dollars based on the consumer price index, and

X5 is the debt-to-asset ratio (D/A) at loan

origination.

c. When applying the equation to Farmer Mac's portfolio, you

must get the input values for X1, X3,

X4, and X5 for each loan on the stress test

run date. For the variable X2, the stressful input value

from the benchmark loss experience is -23.52 percent. You must apply

this input to all Farmer Mac loans subject to loss to calculate loss

frequency under stressful economic conditions.3 The

maximum land value decline stressed input from the benchmark loss

experience is the simple average of annual land value changes for

Iowa, Illinois, and Minnesota for the years 1984 and 1985.

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

\3\ On- and off-balance sheet Farmer Mac I agricultural mortgage

program assets booked after the 1996 amendments are subject to the

loss calculation.

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

d. The loss frequency (default) equation is non-linear and,

therefore, using inputs outside the estimation data requires special

treatment to implement the non-linear nature of the equation. While

the estimation data embody Farmer Mac values for various loan

characteristics, the maximum farmland price decline experienced in

Texas was 16.69 percent, far below the benchmark experience of 23.52

percent. Applying the more severe benchmark loss experience to the

increasing non-linear loss frequency equation could result in

unreasonably large loss rates. The rates could get too large if the

actual relationship between loss rates and land value declines is

lower than calculated from the estimation data. To account for this

effect you must apply a procedure that restricts the slope of all

the independent variables to that observed at the maximum land value

decline observed in the estimation data. Essentially, you must

approximate the slope of each variable and use the measurement to

adjust the probability of loan default and loss occurrence to

reflect the more severe benchmark land value change. The adjustment

procedure is shown in step 4 of section 2.3 of this appendix

entitled Example Calculation of Dollar Loss on One Loan.

e. Loss severity is a weighted average rate of 20.9 percent

where the weight is loss volume.4 You must multiply loss

severity with the probability estimate computed from the loss

frequency equation to determine the origination loss rate for a

loan.

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

\4\ We calculated the weighted average severity from the

estimation data.

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

f. Using origination data results in estimated probabilities of

loss frequency over the life of a loan. To account for loan

seasoning, you must apply the loan seasoning distribution and

subtract the cumulative distribution of loss exposure already

experienced by each loan as discussed in section 2.3 of this

appendix entitled Loan Seasoning Adjustment. This subtraction is

based on loan age and reduces the loss estimated by the loss

frequency and severity equations. The result is an age-adjusted

dollar loss that can be used in subsequent calculations of loss

rates as discussed in section 2.5 of this appendix entitled

Calculation of Loss Rates for Use in the Stress Test.

2.2 Loan Seasoning Adjustment

a. You must use the seasoning distribution to adjust each Farmer

Mac loan for the cumulative loss exposure already experienced based

on age. The estimated seasoning distribution for a 14-year average

loan life and estimated values of p = 5.0875 and q = 13.6376 is: \5\

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

\5\ We estimated the loan seasoning distribution from portfolio

aggregate charge-off rates from the estimation data. To do so, we

arrayed all defaulting loans where loss occurred according to the

time from origination to default. Then, a beta distribution,

(p, q), was fit to the estimation data scaled to the

maximum time a loan survived (14 years).

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

Proportion

Year of loss

(percent)

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

1......................................................... 0.58

2......................................................... 8.30

3......................................................... 21.98

4......................................................... 27.56

5......................................................... 21.99

6......................................................... 12.45

7......................................................... 5.18

8......................................................... 1.57

9......................................................... 0.33

10......................................................... 0.05

11......................................................... 0.00

12......................................................... 0.00

13......................................................... 0.00

14......................................................... 0.00

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

b. How you must use the loan seasoning distribution is shown in

step 7 of section 2.3 of this appendix entitled Example Calculation

of Dollar Loss on One Loan.

2.3 Example Calculation of Dollar Loss on One Loan

Following is an example of how to calculate the loss for an

individual loan that has the following independent characteristics

and input values: \6\

Loan Origination Year.................................... 1996

Loan Origination Balance................................. $1,250,000

LTV at Origination....................................... 0.5

D/A at Origination....................................... 0.5

DSCR at Origination...................................... 1.3984

Maximum Percentage Land Price Decline (MAX............... -23.52

\6\ In the example calculations, we rounded numbers. However,

the stress test does not use rounded numbers.

Step 1: Convert 1996 Origination Value to 1997 dollar value

(LOAN) based on the consumer price index as follows: $1,278,750 =

$1,250,000 1.023

[[Page 61759]]

Step 2: Calculate the default probabilities using -16.69 percent

and -16.79 percent land value declines as follows:\7\

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

\7\ This process facilitates the approximation of slope needed

to adjust the loss probabilities for land value declines greater

than observed in the estimation data.

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

Where,

[GRAPHIC] [TIFF OMITTED] TP12NO99.005

Step 3: Calculate the slope adjustment. You must calculate slope

by subtracting the difference between ``Default Probability @ -16.69

percent'' and ``Default Probability @ -16.79 percent'' and dividing

by -0.1 (the difference between -16.69 percent and -16.79 percent)

as follows:

[GRAPHIC] [TIFF OMITTED] TP12NO99.006

Step 4: Make the linear adjustment. You make the adjustment by

increasing the ``default probability @ -16.69 percent'' computed in

Step 2 to reflect the stressed farmland value input, appropriately

discounted. As discussed previously, the stressed land value input

is discounted to reflect the declining effect that the maximum land

value decline has on the probability of default when it occurs later

in a loan's life.\8\ The linear adjustment is the difference between

the -16.69 percent land value decline and the adjusted stressed

maximum land value decline input of -23.52 multiplied by the slope

estimated in Step 3 as follows:

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

\8\ The discount period is the number of years from the

beginning of the origination year to the current year (i.e., January

1, 1996 to January 1, 2000, is 4 years).

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

Discounted Maximum Land Price Decline = -19.50 = (-23.52)(1.048)-\4\

Slope Adjustment = 0.06575 = 0.02340 (-16.69---19.50)

Loan Default Probability = 0.144026 = 0.078276 + 0.06575

Step 5: Multiply loan default probability times the average

severity of 0.209 as follows:

0.03010 = 0.144026 0.209

Step 6: Multiply the loss rate times the origination loan

balance as follows:

$37,625=$1,250,000 x 0.03010

Step 7: Adjust the dollar losses for 4 years of loan seasoning

as follows:

$15,644=$37,625-($37,625 x 0.584215)

b. The loan seasoning adjustment factor is obtained from the

beta distribution, previously discussed, for the age of the loan,

where age is determined from loan origination to the run date of the

test.

2.4 Treatment of Long-term Standby Purchase Commitments.

a. The default equation cannot directly compute the loss

exposure on loans covered by a long-term standby purchase commitment

(standbys) because complete origination underwriting standards for

these loans are unavailable. Instead, the loss rate applied to each

standby loan is the respective state-level loss rate unadjusted for

loan seasoning. You must calculate state-level loss rates from non-

standby loans as total dollar loan losses before the loan seasoning

adjustment divided by total origination loan balances. Then you must

multiply the origination loan balance of each standby loan by the

appropriate loss rate to calculate estimated dollar losses. You must

now adjust the resulting standby loan-level dollar losses adjusted

for loan seasoning as was done for non-standby loans. For example,

consider a $1,000,000 standby loan originated in Idaho in 1990. And,

suppose the unadjusted loss rate for Idaho is 3 percent. The loss

for this loan is:

($1,000,000 x 0.03) = $30,000.

The loan is 7 years old, thus the estimated age-adjusted loss rate

is:

Estimated standby loan loss=$30,000*(0.02)=$600. As previously

noted, the loan seasoning adjustment factor is obtained from the

beta distribution for the age of the loan, where age is determined

from loan origination to the run date of the test.

c. This treatment may not be used for loans that exhibit risk

characteristics that, at the time Farmer Mac makes the commitment,

disqualify the loan from being placed in the lowest risk category of

the internal credit classification systems of both guarantor and

guarantee. In the credit component of the stress test, such loans

must be treated in the same manner as a new loan in any standard

Farmer Mac I program. Thus, the risk characteristics of the loan at

the time Farmer Mac enters into the standby commitment are treated

as loan origination characteristics for calculating credit losses.

2.5 Calculation of Loss Rates for Use in the Stress Test.

a. You must compute loss rates by state (based on Farmer Mac's

loan portfolio distribution) after you calculate dollar loan losses

for each loan subject to loss in Farmer Mac's portfolio. The

estimated origination year lifetime losses adjusted for loan

seasoning for non-standby loans are computed as total dollar loan

losses divided by total origination loan balances for each state.

Similarly, you must calculate the estimated origination year

lifetime losses adjusted for loan seasoning for standby loans. This

calculation is total dollar loan losses divided by total scheduled

current loan balances for each state. You must then blend the

resulting state-level loss rates for non-standby and standby loans

by calculating the weighted average loss rate for each state. For

instance, the state-level loss rates you would calculate on Farmer

Mac's current loan portfolio are:

[[Page 61760]]

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

Blended rate for

Non-standby loans Standby loans stress test use

(percent) (percent) (percent)

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

All States.................................... 3.24 0.14 2.42

Alaska........................................ 3.24 0.00 0.00

Alabama....................................... 4.58 0.14 4.58

Arkansas...................................... 1.97 0.14 1.97

Arizona....................................... 2.32 0.14 1.68

California.................................... 3.89 0.33 3.83

Colorado...................................... 2.78 0.14 2.78

Connecticut................................... 3.24 0.14 2.42

Delaware...................................... 1.90 0.14 1.90

Florida....................................... 1.46 0.00 1.42

Georgia....................................... 3.78 0.14 3.78

Hawaii........................................ 3.24 0.44 0.44

Iowa.......................................... 3.81 0.14 3.81

Idaho......................................... 2.88 0.12 1.57

Illinois...................................... 3.95 0.31 3.86

Indiana....................................... 3.31 0.14 3.31

Kansas........................................ 1.92 0.00 1.92

Kentucky...................................... 1.46 0.14 1.46

Louisiana..................................... 2.06 0.14 2.06

Massachusetts................................. 3.24 0.14 2.42

Maryland...................................... 1.40 0.14 1.40

Maine......................................... 3.24 0.00 0.00

Michigan...................................... 2.42 0.00 2.41

Minnesota..................................... 2.46 0.00 2.46

Missouri...................................... 2.96 0.14 2.96

Mississippi................................... 3.62 0.14 3.62

Montana....................................... 2.09 0.10 0.82

North Carolina................................ 2.31 0.00 2.12

North Dakota.................................. 2.04 0.14 2.04

Nebraska...................................... 1.89 0.14 1.89

New Hampshire................................. 3.24 0.14 2.42

New Jersey.................................... 3.24 0.81 0.81

New Mexico.................................... 3.79 0.00 3.73

Nevada........................................ 4.74 0.00 4.62

New York...................................... 1.17 0.33 1.06

Ohio.......................................... 2.05 0.14 2.05

Oklahoma...................................... 2.13 0.14 2.13

Oregon........................................ 2.84 0.15 1.13

Pennsylvania.................................. 3.24 0.14 2.42

Rhode Island.................................. 3.24 0.14 2.42

South Carolina................................ 3.24 0.14 2.42

South Dakota.................................. 1.49 0.14 1.49

Tennessee..................................... 1.25 0.14 1.25

Texas......................................... 4.53 0.71 4.51

Utah.......................................... 2.39 0.39 2.29

Virginia...................................... 3.55 0.29 2.40

Vermont....................................... 3.24 0.14 2.42

Washington.................................... 2.93 0.13 1.65

Wisconsin..................................... 6.72 0.14 6.72

West Virginia................................. 3.24 0.14 2.42

Wyoming....................................... 2.61 0.00 2.48

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

b. How the stress test uses the blended loss rates is discussed

in section 4.3 of this appendix entitled Risk Measures.

3.0 Interest Rate Risk.

The stress test explicitly accounts for Farmer Mac's

vulnerability to interest rate risk from the movement in interest

rates specified in the statute. The stress test considers Farmer

Mac's interest rate risk position through the current structure of

its balance sheet, reported interest rate risk shock-test

results,\9\ and other financial activities. The stress test

calculates the effect of interest rate risk exposure through market

value changes of interest-bearing assets and liabilities, and thus

equity capital. The stress test also captures this exposure through

the cashflows on rate-sensitive assets and liabilities. We discuss

how to calculate the dollar impact of interest rate risk in section

4.0 of this appendix entitled Elements Used in Generating Cashflows.

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

\9\ See paragraph c of section 4.1 of this appendix entitled

Data Inputs for a description of the IRR shock-reporting

requirement.

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

3.1 Process for Calculating the Interest Rate Movement.

a. The stress test uses the 10-year Constant Maturity Treasury

(10-year CMT) released by the Federal Reserve in their publication

HR. 15 Selected Interest Rates, which is available on their website

at www.frb.gov. The stress test uses the 10-year CMT to generate

earnings yields on assets, expense rates on liabilities, and changes

in the market value of assets and liabilities. For stress test

purposes, the starting rate for the 10-year CMT is the 3-month

average of the most recent monthly rate series published by the

Federal Reserve and available through their website. The 3-month

average is calculated by summing the monthly series of the 10-year

CMT and dividing by 3. For instance, you would calculate the initial

rate on June 30, 1999, as:

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

10-year

Month end CMT monthly

series

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

04/1999.................................................... 5.18

[[Page 61761]]

05/1999.................................................... 5.54

06/1999.................................................... 5.90

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

Average.................................................. 5.54

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

b. The amount by which the stress test shocks the initial rate

up and down is determined by calculating the 12-month average of the

10-year CMT monthly series. If the resulting average is less than 12

percent, the stress test shocks the initial rate by an amount

determined by multiplying the 12-month average rate by 50 percent.

However, if the average is greater than or equal to 12 percent, the

stress test shocks the initial rate by 600 bp. For example, you

would determine the amount by which to increase and decrease the

initial rate for June 30, 1999 as:

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

10-year CMT

Month End Monthly

Series

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

07/1998.................................................... 5.46

08/1998.................................................... 5.34

09/1998.................................................... 4.81

10/1998.................................................... 4.53

11/1998.................................................... 4.83

12/1998.................................................... 4.65

01/1999.................................................... 4.72

02/1999.................................................... 5.00

03/1999.................................................... 5.23

04/1999.................................................... 5.18

05

This text is long and has been trimmed here. Open the source document for the complete record.