Emergency Disaster Farm Loans: Government's Financial Risk Could Be
Reduced (Letter Report, 03/29/96, GAO/RCED-96-80).

GAO analyzed the financial condition of the Department of Agriculture's
multi-billion dollar farm loan portfolio, focusing on: (1) the factors
that contribute to the financial risk associated with these loans; and
(2) the extent to which farmers attempted to minimize the need for farm
loans by purchasing insurance to protect their farming operations.

GAO found that: (1) since 1989, the Farm Service Agency (FSA) has
forgiven over $6 billion in emergency disaster farm loans and interest;
(2) additional losses should be expected because 80 percent of the $3
billion in loan debt is held by borrowers who are already delinquent or
have had difficulty repaying their loans; (3) weak FSA lending policies
and practices add to the inherent risk of emergency loans; (4) FSA does
not consistently implement lending safeguards to protect federal
financial interests; (5) few borrowers purchase insurance to protect
their property, preferring to rely on government assistance.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-96-80
     TITLE:  Emergency Disaster Farm Loans: Government's Financial Risk 
             Could Be Reduced
      DATE:  03/29/96
   SUBJECT:  Agricultural programs
             Insurance
             Debt collection
             Farm credit
             Farm income stabilization programs
             Farm subsidies
             Delinquent loans
             Loan defaults
             Emergency loans
             Loan repayments
IDENTIFIER:  USDA Emergency Disaster Farm Loan Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

March 1996

EMERGENCY DISASTER FARM LOANS -
GOVERNMENT'S FINANCIAL RISK COULD
BE REDUCED

GAO/RCED-96-80

Reducing the Risk of Emergency Disaster Farm Loans

(150327)


Abbreviations
=============================================================== ABBREV

  ASCS - Agricultural Stabilization and Conservation Service
  CAR - Coordinated Assessment Review
  FCIC - Federal Crop Insurance Corporation
  FSA - Farm Service Agency
  GAO - General Accounting Office
  OIG - Office of Inspector General
  USDA - U.S.Department of Agriculture

Letter
=============================================================== LETTER


B-271392

March 29, 1996

The Honorable Richard G.  Lugar
Chairman
The Honorable Patrick J.  Leahy
Ranking Minority Member
Committee on Agriculture, Nutrition,
 and Forestry
United States Senate

The Honorable Pat Roberts
Chairman
The Honorable E (Kika) de la Garza
Ranking Minority Member
Committee on Agriculture
House of Representatives

We have issued a series of reports highlighting the significant
financial risk associated with the U.S.  Department of Agriculture's
(USDA) multibillion-dollar farm loan portfolio.\1 This report focuses
on one component of the portfolio--emergency disaster loans.  The
emergency disaster loan program, administered by USDA's Farm Service
Agency (FSA)\2 provides assistance to farmers who sustain losses
resulting from adverse weather conditions or other natural disasters. 
The objectives of our work were to analyze the financial condition of
the emergency loan portfolio and identify factors that contribute to
the financial risk associated with these loans.  We also examined the
extent to which farmers attempted to minimize the need for this type
of government loan by purchasing insurance to protect their farming
operations. 

We conducted our work at FSA headquarters and at state and county
offices in California, Florida, Iowa, Michigan, and New York.  Our
work included analyzing information in FSA's centralized computer
databases on the payment status of emergency loans.  In addition, we
mailed a survey to FSA county officials to obtain detailed
information about a random sample of 600 loans made during fiscal
years 1992-94 (see apps.  II and III).  This sample allowed us to
make national estimates of certain characteristics of all emergency
loans made during that period. 


--------------------
\1 See "Related GAO Products" at the end of this report. 

\2 FSA consolidates and delivers services formerly provided by
several predecessor agencies, including the farm loan programs
administered by the Farmers Home Administration. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Over the past 7 years, the Farm Service Agency (FSA), under its
emergency disaster farm loan program, has forgiven over $6 billion in
unpaid principal and interest.  More losses can be expected because
80 percent of the $3 billion in outstanding loan principal is held by
borrowers who are delinquent or who have previously had difficulty
repaying emergency or other farm program loans. 

Although emergency loans are inherently risky, several lending
policies and practices have added to their risk.  For example,
current legislation does not prohibit borrowers who have received
debt forgiveness on past loans from obtaining new emergency loans. 
Under another policy, borrowers with minimal projected cash flow
margins are eligible to obtain loans--as long as their expected
income equals their expected expenses, they qualify, and no cushion
is required for unexpected difficulties.  Besides having weak lending
policies, the agency does not consistently implement lending
requirements that are intended to safeguard federal financial
interests.  For example, the agency does not always verify the
accuracy of the information in loan applications. 

Although crop insurance was generally available, few recipients of
emergency loans obtained coverage to protect their crops against the
risk of natural disaster.  Instead, they relied on the federal
government for assistance.  Recent legislation strengthens the
requirement that emergency loan borrowers have crop insurance at the
time of the disaster loss in order to qualify for loans.  However, in
most years, the Congress has waived this requirement. 


   BACKGROUND
------------------------------------------------------------ Letter :2

As the government's lender of last resort to the nation's farmers,
FSA lends federal funds directly to farmers who are unable to obtain
financing elsewhere at reasonable rates and terms.  It also
guarantees loans made by other agricultural lenders. 

The emergency disaster loan program--in operation since 1949--is one
of several FSA loan programs.  Emergency disaster loans cover actual
physical and production losses so that farmers can return to normal
farming operations.  The emergency loan program consists entirely of
subsidized loans, offered at interest rates below those offered by
other federal farm loan programs.  As of January 1996, the emergency
loan interest rate was 3.75 percent, while rates for other types of
loans, such as operating and farm ownership loans, were 6.5 percent
and 7.0 percent, respectively. 

Emergency loans are made available in specific counties when
disasters are declared by the President, the Secretary of
Agriculture, or the FSA Administrator.  Farmers may use loan funds
for a variety of purposes, including land restoration, farm operating
costs, family living expenses, and debt refinancing.  Under the Food
Security Act of 1985, applicants are ineligible for emergency loans
for crop losses if they did not purchase insurance for crops planted
or harvested after December 31, 1986, assuming it was available. 
However, in most years, the Congress has waived this requirement. 

Currently, the maximum emergency loan assistance available to a
farmer for any single disaster is the amount necessary to restore the
farm to its condition preceding the disaster, not to exceed (1) the
sum of 80 percent of the production losses plus 100 percent of the
physical losses not reimbursed by other sources or (2) $500,000,
whichever is less.  Farmers can qualify for additional emergency
loans following subsequent disasters.  There is no limit on the total
amount of emergency loan debt that a farmer may accrue. 

Over the years, changes have occurred in the emergency disaster loan
program.  The program was expanded in the 1970s to achieve objectives
other than the recovery from actual disaster losses, such as helping
farmers survive during periods of financial stress or make major
adjustments in their farming operations.  However, during the 1980s,
restrictions were placed on the availability of emergency loans,
including the $500,000 limit for each disaster.  As shown in table 1,
the number and total value of emergency loans have varied greatly. 



                                Table 1
                
                Amount of Emergency Loan Obligations by
                       Year, Fiscal Years 1975-95

                         (Dollars in millions)

                                                        Amount of loan
Fiscal year                        Number of loans         obligations
------------------------------  ------------------  ------------------
1975                                        43,675              $735.0
1976                                        18,453              $519.2
1977                                        35,769            $1,178.4
1978                                        89,119            $3,411.8
1979                                        62,913            $2,871.6
1980                                        54,394            $2,226.9
1981                                       138,990            $5,112.3
1982                                        42,863            $2,173.4
1983                                         8,771              $565.9
1984                                        34,997            $1,052.0
1985                                        14,060              $490.9
1986                                         5,584              $217.8
1987                                         2,548              $113.6
1988                                           554               $29.9
1989                                         2,806               $73.5
1990                                         2,609              $101.5
1991                                         1,181               $81.5
1992                                         1,602               $74.9
1993                                           885               $58.6
1994                                         3,815              $145.7
1995                                         1,526               $68.8
----------------------------------------------------------------------
Source:  FSA's obligation reports. 


      OUR PREVIOUS REPORTS HAVE
      HIGHLIGHTED THE RISKS IN
      FSA'S FARM LOAN PROGRAMS
---------------------------------------------------------- Letter :2.1

Problems with FSA's farm loan portfolio are not new.  In 1990, we
identified FSA's farm loan programs as one of 17 high-risk areas
especially vulnerable to waste, fraud, abuse, and mismanagement.  In
1992, we reported that the taxpayers' interests were not being
protected in these programs, the agency had evolved into a source of
continuous farm credit for many borrowers, and billions of dollars in
debt were being written off.\3 Large losses continue to plague the
programs. 

We also issued two reports that focused specifically on emergency
disaster loans.  In November 1987, we reported that delinquencies in
the emergency disaster loan program were increasing, and we suggested
that the Congress consider whether credit, particularly less
restrictive credit, was the proper vehicle for providing disaster
relief and whether a proper balance of risk existed between the
farmer and the government.\4 In September 1989, we reported on the
federal government's strategies for responding to natural disasters
affecting agriculture--direct cash payments, subsidized loans, and
subsidized insurance (crop insurance).\5 We concluded that crop
insurance was the most effective of the three disaster assistance
strategies in part because it could minimize the government's costs. 


--------------------
\3 Farmers Home Administration:  Billions of Dollars in Farm Loans
Are at Risk (GAO/RCED-92-86, Apr.  3, 1992). 

\4 Farmers Home Administration:  Problems and Issues Facing the
Emergency Loan Program (GAO/RCED-88-4, Nov.  30, 1987). 

\5 Disaster Assistance:  Crop Insurance Can Provide Assistance More
Effectively Than Other Programs (GAO/RCED-89-211, Sept.  20, 1989). 


   LARGE LOSSES IN THE EMERGENCY
   LOAN PROGRAM ARE LIKELY TO
   CONTINUE
------------------------------------------------------------ Letter :3

Although the amount of the outstanding emergency loan principal and
the amount owed by delinquent borrowers have declined, FSA's
emergency loan program continues to exhibit the problems that we
first identified in 1987.  From 1989 through 1995, the program has
lost over $6.1 billion in debt forgiveness.  Moreover, much of the
remaining portfolio is at high risk of failure. 


      BILLIONS OF DOLLARS HAVE
      BEEN LOST
---------------------------------------------------------- Letter :3.1

As shown in table 2, from fiscal years 1989 through 1995,\6

FSA forgave over $6.1 billion in principal and interest held by over
35,000 borrowers who did not meet their payment obligations. 



                                Table 2
                
                Amount of Emergency Loan Debt Forgiven,
                          Fiscal Years 1989-95

                         (Dollars in millions)

                                         Amount of
                                    emergency loan           Number of
Fiscal year\a                        debt forgiven           borrowers
------------------------------  ------------------  ------------------
1989                                      $1,617.3              11,102
1990                                       1,330.4               8,400
1991                                         815.0               4,525
1992                                         788.2               3,910
1993                                         647.8               4,032
1994                                         543.3               2,592
1995                                         403.5               1,769
======================================================================
Total                                   $6,145.6\b            35,439\c
----------------------------------------------------------------------
Note:  Judicial decisions in 1984 and 1987 prevented FSA's
predecessor agency--the Farmers Home Administration--from foreclosing
on delinquent borrowers.  As a result, the losses since 1989 are
probably higher than they would have been if the losses had been
spread out over the previous years when the agency could not
foreclose. 

\a Debt forgiven in any year may consist of loans made in previous
years. 

\b The figures in the column do not add to the total because of
rounding. 

\c The figures in the column do not add to the total because
individual borrowers are counted only once and some borrowers
received more than one loan over the 7 years. 

Source:  GAO's analysis of FSA's data. 

In commenting on this report, agency officials noted that many of the
past losses, as well as much of the risk in the current portfolio,
are primarily attributable to past lending policies that have since
been changed (see further discussion in the agency comments section). 
Also, they noted that the federal government's overall exposure to
risk has decreased because the size of the program has declined. 


--------------------
\6 We selected fiscal year 1989 as the starting point for our review
because it was the first year for which computerized data were
available. 


      FUTURE LOSSES COULD BE
      SIGNIFICANT
---------------------------------------------------------- Letter :3.2

Although FSA has forgiven billions of dollars in emergency loans to
borrowers who have had problems repaying their debt, the portfolio
continues to present a high risk of substantial losses to the
government.  Much of the portfolio is held by borrowers who are
delinquent or who have previously had difficulty repaying emergency
or other types of FSA loans.  Furthermore, borrowers have already had
difficulty repaying recent loans, which reflect the most current
changes to the agency's emergency lending policies and practices. 

As of September 30, 1995, FSA had over $3 billion in outstanding
emergency loan principal to 42,093 borrowers.  Of that amount, about
$1.8 billion, or 58.6 percent, was held by borrowers who were
delinquent on emergency loans, a percentage that has been fairly
consistent over the last several years, as shown in table 3. 



                                Table 3
                
                Outstanding Emergency Loan Principal and
                Amount Owed by Delinquent Emergency Loan
                  Borrowers at Fiscal Year End, Fiscal
                             Years 1985-95

                         (Dollars in millions)

                               Outstanding     Principal    Percentage
                                 emergency       owed by       owed by
                                      loan    delinquent    delinquent
Fiscal year end                  principal     borrowers     borrowers
----------------------------  ------------  ------------  ------------
1985                              $9,862.2      $6,116.1          62.0
1986                              $9,373.4      $5,928.3          63.2
1987                              $8,639.6      $5,680.5          65.7
1988                              $8,413.5      $5,932.2          70.5
1989                              $7,682.6      $5,252.2          68.4
1990                              $6,057.3      $3,802.0          62.8
1991                              $5,296.2      $3.348.0          63.2
1992                              $4,526.2      $2,847.7          62.9
1993                              $3,876.1      $2,279.6          58.8
1994                              $3,435.1      $1,914.9          55.7
1995                              $3,046.3      $1,785.3          58.6
----------------------------------------------------------------------
Source:  GAO's analysis of FSA's data. 

Payment status alone, however, does not provide a complete measure of
the potential risk associated with the portfolio.  This indicator
excludes borrowers who may be current on emergency loan payments but
who have previously had problems repaying emergency loans or other
types of FSA farm loans.  These problems required FSA to restructure
payment schedules or forgive debt.  As of September 30, 1995, such
borrowers held approximately $665 million, or 21.8 percent, of the
outstanding emergency loan principal.  When this principal is
combined with the principal owed by delinquent borrowers, about $2.5
billion, or 80.4 percent, of the outstanding emergency loan principal
is held by borrowers who are at risk. 

This overall picture of potential risk in the emergency loan
portfolio reflects both past and present lending policies and
practices.  To better assess the risk associated with current loans,
we examined the status of loans made during fiscal years 1992 through
1994. 

FSA made 6,302 emergency loans totaling $279 million to 5,753
borrowers during this period.  Although these loans are relatively
new--from 1 to 4 years old--repayment problems have already surfaced. 
Through September 30, 1995, FSA had forgiven $1.2 million in
uncollectible principal and interest on 41 of these loans.  In
addition, on the basis of our sample of 600 loans, we estimate that
payments were delinquent on approximately 25 percent of the loans as
of January 10, 1995.\7


--------------------
\7 We chose Jan.  10 as a cutoff date to give the FSA county staff
time to record payments at the beginning of the new year. 


   WEAK LENDING POLICIES AND
   PRACTICES ADD TO INHERENT RISK
   OF EMERGENCY LOANS
------------------------------------------------------------ Letter :4

The likelihood that farmers will repay emergency loan debt is
diminished by the nature of the loans.  Emergency loans are
inherently riskier than other types of farm loans because they are
made to help farmers recover from losses rather than generate new
income.  This problem is compounded by weak lending policies and the
agency's failure to implement existing lending requirements. 


      LENDING POLICIES ARE WEAK
---------------------------------------------------------- Letter :4.1

Three lending policies expose the government to potential losses by
allowing borrowers who have poor credit histories or are in a very
weak financial position to obtain loans.  First, the provisions of
the Consolidated Farm and Rural Development Act as amended (P.L. 
87-128, Aug.  8, 1961) do not prohibit borrowers who received prior
FSA debt forgiveness from obtaining additional farm loans, including
emergency loans.  We identified 293 borrowers who obtained $11.6
million in emergency loans during fiscal years 1992 through 1994
after having about $51 million in unpaid debt forgiven.  As of
September 30, 1995, 27 percent of these borrowers were already
experiencing problems repaying the recent emergency loans.  The
following example illustrates this problem: 

  A New York vineyard owner, an FSA farm loan borrower since 1978,
     received a 1993 emergency loan of $9,640 for crop losses
     resulting from a drought.  FSA made this loan after having
     forgiven approximately $207,000 on five other farm loans in
     1990.  In January 1995, the borrower was delinquent on the 1993
     emergency loan.  FSA classified this borrower as being unlikely
     to repay the loan. 

Second, FSA's method for determining an applicant's ability to repay
a loan does not provide for contingencies.  The current "cash flow"
lending criterion requires only that an applicant have an estimated
income equal to the estimated expenses in order to qualify for a
loan.  It does not provide a cushion for any unanticipated expense
that may occur during the life of the loan.\8 On the basis of our
sample of loans made during fiscal years 1992 through 1994, we
estimate that FSA made about 62 percent of the emergency loans to
borrowers whose anticipated income exceeded their expenses by less
than 10 percent.  Furthermore, about 17 percent of these borrowers
attained these minimal cash flows only because FSA rescheduled or
forgave debt on which the borrowers had failed to honor repayment
schedules.  As of September 30, 1995, approximately 37 percent of the
borrowers with cash flow margins of less than 10 percent were behind
on their loan payments or had required debt restructuring or
forgiveness after receiving their emergency loans.  For borrowers
with cash flow margins of 10 percent or more, this percentage dropped
to about 28 percent.  The following example shows the kinds of
problems that can arise when cash flow margins are minimal: 

  A Michigan fruit producer applied for a 1992 emergency loan because
     of freeze damage.  FSA approved the application, even though the
     applicant--an FSA borrower since 1985--had a poor payment
     history and a projected income exceeding his projected expenses
     by only $2.  This borrower's cash flow margin was low even after
     FSA rescheduled existing loans on which he could not make
     payments.  In 1993, the 1992 emergency loan was rescheduled
     because the borrower had not made payments; in 1994, the debt
     was forgiven; and, as of July 1995, the borrower was again
     delinquent on other FSA loan payments, and the county supervisor
     expected additional losses to the government. 

Finally, there is no limit on the amount of emergency loan
indebtedness an individual borrower may accumulate.  Although
assistance for a single disaster is limited to the amount necessary
to restore a farm to its predisaster condition or $500,000, whichever
is less, a farmer can obtain additional emergency loans for
subsequent disasters.  As of September 30, 1995, 696 borrowers had
each accumulated emergency loan principal in excess of $500,000, with
a cumulative total in excess of $800 million.  These borrowers had a
higher rate of delinquency than those with less outstanding emergency
loan principal.  Specifically, borrowers with more than $500,000 in
emergency loan principal had a delinquency rate of 82 percent, while
those with $500,000 or less in emergency loan principal had a
delinquency rate of about 30 percent.  The following example
illustrates the large emergency loan indebtedness that a borrower can
accumulate and the types of repayment problems that can result: 

  A Maryland farmer who operated a dairy and produced multiple crops
     had seven emergency loans with outstanding principal balances
     totaling approximately $850,000 as of September 30, 1995. 
     According to an FSA county official, the borrower has been
     unable to repay the emergency loans on schedule, and FSA has
     deferred payments for 5 years.  We noted that FSA compounded
     this problem in 1992 by providing an emergency loan that
     exceeded the borrower's eligibility level for the disaster by
     $101,000.  FSA did not reduce the borrower's eligible losses, as
     required, by the amount of reimbursements received from the
     USDA's Federal Crop Insurance Corporation (FCIC). 

The Congress is considering legislation that would strengthen two of
the lending policies that expose the government to risk.  In February
1996, the Senate passed a bill that generally would prohibit USDA
from making farm program loans, including emergency loans, to
borrowers whose debts have been forgiven--a proposal similar to one
made by USDA.  This bill would also restrict a borrower's outstanding
emergency disaster loan principal to a maximum of $500,000.\9


--------------------
\8 In 1987, FSA attempted to strengthen its loan-making criteria but
withdrew its proposal in part because the Congress was concerned that
more stringent lending criteria might limit the availability of farm
loans.  Similarly, in 1989, we recommended that USDA work with the
Congress to improve the cash flow lending criteria (Farmers Home
Administration:  Sounder Loans Would Require Revised Loan-Making
Criteria (GAO/RCED-89-9, Feb.  14, 1989)).  However, the criteria
remain unchanged. 

\9 S.  1541, "Agricultural Reform and Improvement Act of 1996, 104th
Cong., 2d Sess., secs.  641, 622 (1996). 


      LENDING REQUIREMENTS ARE NOT
      FOLLOWED
---------------------------------------------------------- Letter :4.2

Reviews by FSA and USDA's Office of Inspector General (OIG) noted
weaknesses in FSA's emergency loan lending practices.  Among other
things, these reviews found that FSA field officials do not always
receive accurate information in determining applicants' loan
eligibility. 

To determine whether its field offices are complying with its lending
requirements, FSA conducts Coordinated Assessment Reviews (CAR).\10
For fiscal years 1992 through 1995, FSA completed CARs of 369
emergency loans.  As shown in table 4, for at least 14 percent of the
loans reviewed, FSA field offices had not verified information on the
level of an applicant's disaster loss, debt, or income before
approving the loan.  While FSA considers noncompliance rates of more
than 15 percent to be unacceptable, any noncompliance increases the
government's financial risk.  FSA officials noted, however, that in
times of certain natural disasters, such as the Midwest flooding in
1993, the need to assist people quickly sometimes takes priority over
following every detailed lending requirement. 



                                Table 4
                
                  Noncompliance With Selected Lending
                Requirements on Emergency Loans, Fiscal
                             Years 1992-95

                                             Number of         Rate of
Lending requirement                       loans tested   noncompliance
--------------------------------------  --------------  --------------
Disaster losses verified                           261            18.4
Debts verified                                     367            16.1
Operating expenses deemed realistic                369             7.0
Nonfarm and other farm income verified             305            14.4
 and included in farm plan
Financial plan feasible                            357             7.0
----------------------------------------------------------------------
Note:  Because the CARs of emergency loans covered only a relatively
small number of cases, the results apply only to the sampled cases
and cannot be projected to all emergency loans. 

Source:  FSA's CAR reports for fiscal years 1992-95. 

The OIG also found problems with lending practices.  According to
December 1994 and March 1995 reports, six of seven emergency loans
reviewed in Wisconsin and Illinois were overstated because they were
not based on the most current and accurate information available at
the time of the loan closings.  Specifically, six borrowers received
loans totaling about $100,000 more than they were entitled to receive
because FSA approved the loans on the basis of information about the
borrowers' eligibility that FSA believed to be accurate but later
found to be in error.  In commenting on a draft of this report, FSA
officials told us that they verified the information when they
approved the loans, but the information changed before they closed
the loans.  Furthermore, according to these officials, their current
standards do not require them to reverify information that is
provided by other USDA agencies, such as the Agricultural
Stabilization and Conservation Service. 


--------------------
\10 In the CARs, FSA reviews a random sample of loans in about 15
states annually.  Each state is reviewed at least once every 3 years. 


   FEW BORROWERS USE INSURANCE TO
   PROTECT THEIR PROPERTY
------------------------------------------------------------ Letter :5

We have previously reported that subsidized crop insurance, compared
to other forms of federal assistance such as loans and direct
payments, is an efficient and equitable method of providing disaster
assistance.\11 Although we did not perform a detailed analysis of why
borrowers did not obtain insurance, FSA county officials reported
that most borrowers who chose not to purchase insurance did so
because they did not consider coverage to be cost-effective.  Our
sample of loans made during fiscal years 1992 through 1994, before
recent crop insurance reform legislation provided coverage at minimal
cost, indicates that very few borrowers obtained insurance to protect
their crops against losses resulting from natural disasters, even
though insurance was frequently available. 

More specifically, we estimate that about 96 percent of the emergency
loans made during fiscal years 1992 through 1994 covered crop losses,
4 percent covered real property losses, and 8 percent covered losses
of other property, including livestock.\12 In most cases, crop
insurance coverage was available to the borrowers either through FCIC
or other sources.  However, as shown in table 5, the borrowers
frequently did not purchase coverage, even when both options were
available.  The table also shows that a smaller percentage of the
borrowers rejected hazard insurance. 



                                Table 5
                
                Insurance Coverage on Emergency Disaster
                  Loans Made During Fiscal Years 1992
                              Through 1994


                                              Percentage    Percentage
                                                of loans  of loans for
                                  Source/      for which         which
                      Percentage  type          coverage     available
                        of loans  of                 was  coverage was
Type                    involved  coverage     available  not obtained
--------------------  ----------  ----------  ----------  ------------
Crop                        95.8  FCIC              96.7          51.6
                                  Other             69.9          77.1
                                  FCIC or           98.2          44.5
                                   other
                                  Both              68.4          91.6
Real property                4.1  Hazard            81.1          19.5
Other property\a             8.1  Hazard            64.2          36.7
----------------------------------------------------------------------
Note:  Estimates of sampling errors appear in app.  II. 

\a Includes items such as equipment and livestock. 

Source:  GAO's analysis of county supervisors' responses to GAO's
questionnaire. 

The following example illustrates a situation in which a borrower did
not obtain crop insurance: 

  An Iowa corn and soybean farmer, whose emergency loan application
     showed annual nonfarm income of $34,500 and about $8,000 in cash
     and certificates of deposits, did not obtain either FCIC or
     other crop insurance, even though both were available.  The
     farmer lost $64,090 in crops as a result of flooding in 1993 and
     received approximately $21,000 in USDA disaster assistance, as
     well as an emergency loan for $34,480 in 1994.  As of January
     1995, this borrower was delinquent on the emergency loan
     payments.  FCIC coverage would have cost $1,151 and paid the
     borrower about $15,200 for crop losses. 

According to FSA county officials, most borrowers did not buy
insurance because they did not consider coverage to be
cost-effective.\13 The Food Security Act of 1985 makes applicants
ineligible for emergency crop loss assistance if federal crop
insurance was available to them and they did not purchase it for
crops planted or harvested after December 31, 1986.  This eligibility
requirement has had little impact, however, because it has been
waived in most years by subsequent disaster legislation enacted to
minimize the economic hardships that some farmers might face in the
absence of federal assistance.  These waivers have not been targeted
to grant relief to selected types of borrowers.  Rather, the waivers
have been made available to all who were interested in obtaining
emergency loans in a particular year. 

The crop insurance reform legislation enacted in 1994 may increase
the use of crop insurance among those seeking USDA benefits, such as
emergency loans, and decrease the availability of ad hoc disaster
assistance for crop losses.  The Federal Crop Insurance Reform Act of
1994, which became effective in 1995, generally conditions the
receipt of USDA benefits, including emergency loans and price support
benefits, upon an applicant's having obtained at least the minimum
level of crop insurance available under the act, known as
catastrophic risk protection, at a cost ranging from $50 to $600 for
a borrower.\14 The 1994 act also made the passage of agricultural
disaster assistance legislation more difficult. 

Apart from crop loss insurance, the Congress is now considering
legislation that may increase the use of hazard insurance by farmers. 
The agricultural credit legislation passed by the Senate in 1996\15
would prohibit USDA from making emergency loans to farmers or
ranchers unless the applicants had hazard insurance that insured
their property at the time of the loss.  The level of insurance
needed to satisfy this requirement would be determined by the
Secretary of Agriculture. 


--------------------
\11 Disaster Assistance:  Crop Insurance Can Provide Assistance More
Effectively Than Other Programs (GAO/RCED-89-211, Sept.  20, 1989). 

\12 The total of these percentages exceeds 100 percent because some
loans covered more than one type of property loss. 

\13 In determining why insurance coverage was not obtained, we
limited our review to asking FSA field officials why coverage was not
obtained. 

\14 Since the enactment of the 1994 crop insurance reform
legislation, both Houses of Congress have passed proposed legislation
which would give a person an alternative to obtaining the minimum
level of crop insurance in order to be eligible for certain benefits,
including emergency loans, with respect to spring planted 1996 and
subsequent crops.  Under this proposed legislation, a person could
also remain eligible for these benefits by waiving, in writing, any
eligibility for emergency crop loss assistance in connection with the
crop. 

\15 The February 1996 Senate bill would strike the 1985 law denying
eligibility for emergency loans to those who had federal crop
insurance available to them, apparently because the linkage between
obtaining crop insurance and eligibility for emergency loans was
addressed in the 1994 crop insurance reform legislation. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

FSA's emergency loan program has lost billions of dollars in debt
that has not been repaid, and it stands to lose billions more, given
the characteristics of the borrowers currently holding emergency
loans.  The Congress is considering legislative changes whose
implementation would help reduce the program's losses.  However,
these changes would not correct the weaknesses stemming from FSA's
cash flow lending policy. 

The 1994 insurance reform legislation strengthens the requirement
that farmers have insurance in order to receive federal assistance,
including emergency loans.  In most years, the Congress has waived
similar requirements for obtaining emergency loans, reflecting its
desire to assist farmers suffering from the economic consequences of
natural disasters.  The Congress has, historically, granted waivers
to all farmers who sought loans within a given year.  Continued use
of this blanket type of waiver may contribute to concerns about
equity.  For example, borrowers who, on a one-time basis, neglected
to obtain insurance, would be treated in exactly the same way as
borrowers who have repeatedly chosen not to obtain insurance and have
relied, instead, on federal assistance to cover their losses. 


   RECOMMENDATION TO THE SECRETARY
   OF AGRICULTURE
------------------------------------------------------------ Letter :7

We continue to believe that our 1989 recommendation to USDA to
strengthen its cash flow lending policy has merit.  More
specifically, we recommend that the Secretary of Agriculture direct
the FSA Administrator to develop regulations that improve the cash
flow analyses used in loan-making decisions by incorporating an
allowance to cover contingencies and the costs of replacing
equipment. 


   MATTERS FOR CONGRESSIONAL
   CONSIDERATION
------------------------------------------------------------ Letter :8

We recognize that recent legislation creates added incentives for
borrowers to purchase crop insurance and that the Congress may
consider many factors when deciding whether to waive the existing
requirement for crop insurance.  However, if the Congress decides to
waive this requirement, it may wish to consider options that would
more selectively target the applicants who would be eligible for the
waiver and limit the amount of the loan that they could receive. 
These options could include (1) prohibiting borrowers who have
previously been granted insurance waivers from receiving additional
waivers and/or (2) reducing the amount of an emergency loan to
exclude the value of the proceeds that would have been available if
the borrower had chosen to purchase the required insurance. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :9

We provided copies of a draft of this report to FSA for its review
and comment.  In a meeting to discuss FSA's comments, the Deputy
Director for Farm Credit Programs and Farm Credit Program staff
generally agreed with the report's conclusions and recommendations. 
However, they believed that several additional factors should be
better recognized in discussions of the loan portfolio's risk. 

FSA stated that its emergency loan obligations have decreased
significantly in recent years; therefore, the government's exposure
to risk has also decreased.  We agree and show the decline in
emergency loan obligations in table 1 of our report. 

FSA also noted that the past losses and much of the risk associated
with the current portfolio are due to policies that are no longer in
effect.  We agree that the current portfolio's problems can be linked
to past policies; however, as noted in our report, not all of the
policies that have contributed to these problems have been corrected. 
Consequently, there are still significant risks associated with the
emergency loan program that could be reduced by further congressional
or agency actions. 

Finally, FSA stated that loan repayment statistics that we developed
on recent loan making indicate acceptable performance, given the
agency's role as a lender of last resort.  In our view, the Congress
is ultimately responsible for determining what constitutes acceptable
levels of performance for these loans.  Our report provides
information that can help the Congress make this determination,
noting, among other things, that these relatively recent loans have
already shown signs of repayment problems. 

We incorporated other technical corrections suggested by FSA
officials as appropriate. 

We performed our work between November 1994 and March 1996 in
accordance with generally accepted government auditing standards. 
Our objectives, scope, and methodology are discussed in appendix I. 
Our methodology for sampling and analyzing data is discussed in
appendix II.  The emergency disaster loan survey is presented in
appendix III. 


---------------------------------------------------------- Letter :9.1

We are sending copies of this report to the appropriate congressional
committees; interested Members of Congress; the Secretary of
Agriculture; the FSA Administrator; the Director, Office of
Management and Budget; and other interested parties.  We will also
make copies available on request. 

Please call me at (202) 512-5138 if you or your staff have any
questions.  Major contributors to this report are listed in appendix
IV. 

Robert A.  Robinson
Director, Food and
 Agriculture Issues


OBJECTIVES, SCOPE AND METHODOLOGY
=========================================================== Appendix I

This review is part of a special GAO effort to address federal
programs that pose a high risk of waste, abuse, and mismanagement. 
To gain a complete understanding of the Farm Service Agency's (FSA)
emergency loan program, we reviewed FSA's regulations, operating
instructions, and other guidance to field offices.  We also
interviewed officials at the agency's headquarters in Washington,
D.C., and at state and county field offices.  We analyzed
computerized databases on the status of loans and loan obligations
provided by FSA's Finance Office in St.  Louis, Missouri. 

Additionally, we reviewed and analyzed our prior reports addressing
emergency loans, reports issued by the U.S.  Department of
Agriculture's (USDA) Office of Inspector General in December 1994 and
March 1995, and the results of FSA's internal control reviews. 

To obtain information on the characteristics of FSA's emergency loan
borrowers and the planned use of loan funds, we mailed a survey to
FSA county officials requesting information about a randomly selected
stratified sample of the loans made from fiscal years 1992 through
1994.  Appendix II discusses our survey's methodology and contains
our estimates and sampling errors.  Appendix III contains a copy of
the survey used. 

We started our field work in November 1994 and used September 30,
1995, as a cutoff date for the financial information about FSA's farm
loan portfolio.  This date allowed us to have relatively recent and
comparable data on the financial status, including the losses, of
FSA's emergency disaster loan portfolio.  We present the loss
information in nominal (versus constant) dollars.  We performed our
work in accordance with generally accepted government auditing
standards. 


METHODOLOGY FOR SURVEY OF
EMERGENCY DISASTER LOANS
========================================================== Appendix II

To obtain data on the emergency disaster loans that FSA made from
fiscal years 1992 through 1994, we obtained computerized records from
FSA of the 6,302 loans obligated during this period.  The emergency
loans obligated during this period totaled $279 million.  FSA
provided the automated data from its obligations database for each
year.  We divided this universe into two strata:  one stratum
consisted of loans whose obligation amount was less than $75,000, and
the other stratum consisted of loans whose obligation amount was
greater than or equal to $75,000. 

We conducted a nationwide mail survey to obtain detailed data on the
emergency disaster loans.  The survey questionnaires were mailed on
February 9, 1995, to the USDA county officials through whom the
emergency loans were made.  Of the 600 questionnaires mailed, 589
were returned with valid responses.  These 589 valid responses
represented an overall response rate of 100 percent because the
remaining 11 files were either not available or the loans were never
closed.  The initial and adjusted universe and the number of
responses by stratum are shown in table II.1.  Our questionnaire
appears in appendix III. 



                               Table II.1
                
                         GAO's Sample of Loans


                                             I          II       Total
----------------------------------  ----------  ----------  ----------
Size of stratum                          5,364         938       6,302
Size of initial sample                     425         175         600
Size of adjusted sample\a                  421         168         589
Estimated size of stratum                5,314         900       6,214
Number of valid responses                  421         168         589
----------------------------------------------------------------------
\a The sample size was adjusted to exclude 11 loans that either were
never closed or had no available record. 

We used the responses to the survey to project estimates for the
universe of 6,302 loans.  In addition, respondents supplied
documentation supporting some of the critical facts in their
responses to the survey.  We used this documentation to verify the
consistency of certain responses.  When inconsistencies occurred and
data were not available to determine the correct answer, we
telephoned the county officials to obtain the correct information. 

Since we used a sample of emergency disaster loans to develop our
estimates, each estimate has a measurable precision, or sampling
error, which may be expressed as a plus/minus figure.  A sampling
error indicates how closely we can reproduce, from a sample, the
results that we would obtain if we used the same measurement methods
to take a complete count of the universe.  By adding the sampling
error to and subtracting it from the estimate, we can develop upper
and lower bounds for the estimate.  This range is called a confidence
interval.  Sampling errors and confidence intervals are stated at a
certain confidence level--in this case, 95 percent.  For example, a
confidence interval of 95 percent means that in 95 out of 100
instances, the sampling procedures we used would produce a confidence
interval containing the universe value that we are estimating.  As
table II.2 indicates, most of our estimates have relatively small
sampling errors of fewer than 5 percentage points. 



                               Table II.2
                
                   Sampling Errors at the 95-Percent
                 Confidence Level for Estimates of Loan
                            Characteristics

                                                            95-percent
                                    Estimate  Sampling      confidence
Description of estimate                   \a     error        interval
----------------------------------  --------  --------  --------------
Status of payment on loan
Behind on payments                      25.2       3.5    21.7 to 28.7
Crop
Type of property loss covered           95.8       1.5    94.3 to 97.3
FCIC coverage available                 96.7       1.4    95.3 to 98.1
FCIC coverage not obtained              51.6       4.4    47.2 to 56.0
Other coverage available                69.9       4.0    65.9 to 73.9
Other coverage not obtained             77.1       4.4    72.7 to 81.5
Either FCIC or other coverage
 available                              98.2       1.0    97.2 to 99.2
Neither FCIC nor other coverage
 obtained                               44.5       4.3    40.2 to 48.8
Both FCIC and other coverage
 available                              68.4       4.0    64.4 to 72.4
Both FCIC and other coverage
 not obtained                           91.6       2.5    89.1 to 94.1
Real property
Type of property loss covered            4.1       1.4      2.7 to 5.5
Hazard coverage available               81.1    15.6\b    65.5 to 96.7
Hazard coverage not obtained            19.5    16.8\b     2.7 to 36.3
Livestock and other property
 (e.g., equipment)
Type of property loss covered            8.1       2.1     6.0 to 10.2
Hazard coverage available               64.2    14.7\b    49.5 to 78.9
Hazard coverage not obtained            36.7    18.7\b    18.0 to 55.4
----------------------------------------------------------------------
\a The estimates presented in this table are based on the percentage
of loans. 

\b The sampling errors are large in these cases because the sample
was small. 




(See figure in printed edition.)Appendix III
EMERGENCY DISASTER LOAN SURVEY
========================================================== Appendix II



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV

RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION

Linda J.  Libician, Assistant Director
Robert E.  Robertson, Adviser
Reid H.  Jones, Project Leader
Syrene D.  Mitchell, Evaluator
Olin S.  Thummel, Evaluator
LuAnn Moy, Social Science Analyst
Robert C.  Sommer, Computer Analyst
Donna B.  Svoboda, Computer Analyst

RELATED GAO PRODUCTS

Consolidated Farm Service Agency:  Update on the Farm Loan Portfolio
(GAO/RCED-95-223FS, July 14, 1995). 

High-Risk Series:  Farm Loan Programs (GAO/HR-95-9, Feb.  1995). 

Farmers Home Administration:  The Guaranteed Farm Loan Program Could
Be Managed More Effectively (GAO/RCED-95-9, Nov.  16, 1994). 

Debt Settlements:  FmHA Can Do More to Collect on Loans and Avoid
Losses (GAO/RCED-95-11, Oct.  18, 1994). 

Farmers Home Administration:  Farm Loans to Delinquent Borrowers
(GAO/RCED-94-94FS, Feb.  8, 1994). 

High-Risk Series:  Farmers Home Administration's Farm Loan Programs
(GAO/HR-93-1, Dec.  1992). 

Farmers Home Administration:  Billions of Dollars in Farm Loans Are
at Risk (GAO/RCED-92-86, Apr.  3, 1992). 

Disaster Assistance:  Crop Insurance Can Provide Assistance More
Effectively Than Other Programs (GAO/RCED-89-211, Sept.  20, 1989). 

Farmers Home Administration:  Sounder Loans Would Require Revised
Loan-Making Criteria (GAO/RCED-89-9, Feb.  14, 1989). 

Farmers Home Administration:  Problems and Issues Facing the
Emergency Loan Program (GAO/RCED-88-4, Nov.  30, 1987). 

Federal Disaster Assistance:  What Should the Policy Be?  (PAD-80-39,
June 16, 1980). 


*** End of document. ***