|
Evaluation of the Food Distribution Program on Indian Reservations
Volume
I: Final Report
EXECUTIVE SUMMARY Overview The
Food Distribution Program on Indian Reservations (FDPIR) provides
supplemental foods to low-income households living on or near Indian
reservations. While recognizing a need for assistance among American
Indians, Congress was concerned that this need may not be adequately
addressed by the Food Stamp Program, the largest and most widely available
food assistance program in the United States. The primary concern was that
the remote location of many reservations may make it difficult for many
American Indians to participate in the Food Stamp Program because they
live some distance from a food stamp office and food stores are scarce or
far away. Thus, FDPIR represents an alternative to the Food Stamp Program
for residents of Indian reservations. This study presents
the first nationally representative profiles of
FDPIR participant and program characteristics, and the food-assistance
needs and preferences among this particular target population. This
executive summary provides a brief historical perspective on the program,
describes the objectives and methodology of the evaluation, and summarizes
the major findings. Program
Background Household Eligibility and
Benefits. To be eligible to receive a commodity package, a
household must meet the income eligibility criteria established by Federal
legislation, and either reside on an Indian reservation or be a tribal
member who resides in the designated service area of a FDPIR program. The
income limits used to determine FDPIR eligibility are the same as Food
Stamp Program net monthly income limits plus the standard deduction used
in determining eligibility for that program. However, FDPIR differs from
the Food Stamp Program in that the amount of food an income-eligible
household receives is based solely on the number of members it contains,
regardless of the specific level of income it has. The
monthly food package consists of a variety of canned and
packaged commodities in such categories as meats, fruits, vegetables,
dairy products, grains, and cereals. The Food and Nutrition Service (FNS)
uses some surplus foods from the U.S. Department of Agriculture (USDA),
but purchases most food specifically for use in FDPIR. Although
supplemental in nature, the overall FDPIR food package is designed to
provide adequate levels of most nutrients and food energy to participating
households. Program Administration. Since its
inception in 1977, FDPIR has grown to 105 local programs serving
approximately 138,000 persons each month in Fiscal Year 1989. The vast
majority of programs are administered by Indian Tribal Organizations (ITOs)
under direct agreements with FNS, although some operate under the
supervision of an agency of State government. FNS oversees
FDPIR program operations through its network of
Regional, Field, and Satellite Offices. FNS also works with other USDA
agencies to procure the commodities, process the orders of local programs,
and arrange shipments of food to local FDPIR programs. Staff in FNS
Regional Offices assess the capability of ITOs to administer FDPIR, review
annual plans of operation and budget requests prepared by individual State
agencies and ITOs, and provide technical assistance and managerial
oversight to local programs. Objectives
of the Evaluation The specific objectives of the
evaluation relate to three sets of the issues: Program
Operations:
-
describe State agency or fro administration of FDPIR
in terms of written policy, reported practice, and estimated costs;
and
-
describe program practices that are intended to
maximize the efficiency and integrity of the program.
Participant Characteristics:
Comparison of the contributions of FDPIR and the Food
Stamp Program in providing food assistance to American Indians:
-
provide a preliminary comparison of the availability
and acceptability Of FDPIR commodities versus food stamps for American
Indians; and
-
explore the unique contributions each program makes in
meeting the food assistance needs of this population.
Research Design and Methods
To address the first set of objectives, we collected
information about the structure and operation of FDPIR programs from
a variety of sources, including staff interviews, plans of operation,
statistical reports that local FDPIR programs routinely submit to FNS, and
Management Evaluation (ME) reviews.
To meet the second and third set of objectives, we
collected household-level data from the following sources:
-
a national probability sample of 827 FDPIR case
records drawn from 30 programs;
-
interviews with 757 participants whose case records
were selected in the sample;
-
interviews with a sample of 107 American Indian
households participating in the Food Stamp Program in Arizona,
Montana, and Wisconsin;
-
a probability sample of American Indian households
whose food stamp cases were reviewed in the summer of 1986 under the
Integrated Quality Control System operated by State food stamp
agencies and FNS; and
-
a series of focus groups with FDPIR or Food Stamp
Program participants.
The sample of FDPIR households was drawn from lists of
participants from 30 FDPIR programs (five large, 10 medium-sized, and 15
small programs) that were representative of all 105 local programs. The
number of households studied in the evaluation was large enough to produce
precise and reliable statistics for the full FDPIR sample and major
subgroups such as households living in larger regions and those containing
an elderly member.
Summary of Findings The
following conclusions and summary of findings is organized
around the three sets of study objectives pertaining to program
operations, program participants, and a comparison of how FDPIR and the
Food Stamp Program meet the food assistance needs of American Indians.
Larger programs appear to achieve administrative
economies of scale.
Program Operations Half
of local FDPIR programs serve fewer than 250 households per month
("small programs"), and all but five of the remaining programs
serve between 250 and 1,200 households per month ("medium-sized
programs"). The five "large programs" serve more than 1,200
households per month. Average administrative
costs per household ranged from $614 among small programs to $287 among
large programs. Similarly, the number of participant households per
full-time equivalent (FrE) staff position is nearly 100 in large programs,
compared to approximately 66 in small programs. The
flat-rate administrative cost guideline established by FNS
does not factor in these relative levels of efficiency. All of the
programs whose administrative costs exceeded the guideline (30 percent of
the value of distributed commodities) serve fewer than 250 households per
month.
Local programs vary widely in the level and type of
resources used to meet administrative matching-fund requirements.
In regions other than the West, about half of local
programs meet the 25-percent administrative funds matching requirement.
FNS waives the requirement upon acceptance of justification submitted by
local programs. Among small and medium-sized programs in the study sample,
nearly two-thirds of the match was based on in-kind contributions, in most
cases the estimated market value of warehouse and office space. In
contrast, larger programs make relatively substantial cash contributions
to support the operation of the program.
Program staffing is not highly specialized in FDPIR,
especially in small programs.
Generally, local FDPIR programs operate with a narrow
range of staff functions in the areas of program administration and
supervision, certification, storage and distribution, and nutrition
education. In small programs, it is not uncommon for an individual to
serve in all four areas. In fact, one out of five of the sample programs
are two-person operations. The highest average salary for
any staff position in Fiscal Year 1989 was $21,185 for program directors.
Although staff in larger programs tend to be paid more than their
counterparts in smaller programs, the generally low level of salaries for
these administrative positions seems to reflect conditions that prevail in
the labor markets in areas served by FDPIR.
Local FDPIR programs use a variety of means to
enhance the accessibility of the program for participants and potential
applicants.
Most directors expressed the opinion that all potentially
eligible
households are aware of the availability of benefits and where to apply
for them. With a few specific exceptions, most programs' outreach efforts
rely on publicizing the distribution schedule each month. At the same
time, many programs operate tailgate certification and distribution
systems, and make home deliveries of commodities to elderly and disabled
participants, to make the program more accessible. Although
program directors do not perceive any language barriers to exist for more
than a small percentage of their clientele, ail programs make provision
for translators to be available (either a staff member or some other
person). A few programs require staff to be bilingual.
The certification process in FDPIR is less demanding
for both applicants and certification specialists compared to the Food
Stamp Program and some other assistance programs.
Less information is required of FDPIR applicants than
those in other food assistance programs, and fewer items have to be
verified and documented. Also, fewer factors must be considered in
calculating whether a household meets established FDPIR income limits
established for FDPIR. In contrast, the Food Stamp Program involves a
check on gross income, a calculation of net income that involves more
possible deductions from gross income than in FDPIR, and the calculation
of a precise allotment of food stamps for eligible households. Finally,
most FDPIR applicants are able to obtain food the day they apply, partly
because regulations permit local programs to grant a one-month
certification pending verification of information.
Nutrition education activities vary widely across
programs.
Federal regulations do not require local FDPIR programs to
offer extensive nutrition education services to program participants.
Rather, programs are encouraged to coordinate with local organizations
that can disseminate food and nutrition information to FDPIR households.
The 30 sample programs in this study allocated an average of five percent
of their administrative funds to this function, with program support
ranging from zero to almost 25 percent of their annual administrative
budgets. Over 25 percent of the programs reported no nutrition education
budget. About six out of ten programs reported
nutrition education personnel expenditures. With few exceptions, these
staff have little or no formal training in either health or nutrition. The
focus of nutrition education activities tends to be distributing commodity
recipes and cookbooks, and demonstrating how specific items can be
prepared. Almost one-half of the programs distribute other general food
and nutrition information. However, many programs are unable to maintain a
supply of their nutrition education materials, thus limiting the
effectiveness of their efforts. The effectiveness of FDPIR
in providing a nutritious diet to participants depends in large measure on
the participants' ability to: properly select and use commodity foods;
identify potential nutrition-related health problems; and make changes in
their households and the community to improve health and nutrition. The
development of these skills is particularly important among American
Indians because they experience high rates of diet related health
problems, such as diabetes, high blood pressure, and obesity than the
general population.
Local programs use a combination of commodity
distribution methods to meet recipient needs and local situations.
In an effort to reduce participants' difficulty in
obtaining commodities, about half of the programs in the study sample use
the tailgate distribution method in addition to central warehouse pickup.
Another third of the programs also deliver commodities to the homes of a
relatively small number of elderly and disabled participants. Very few
programs, primarily those serving small caseloads, rely solely on
commodity distribution from a central warehouse location. While
the use of tailgate distribution systems reduces the distance participants
must travel to obtain their food package, the selection of items within
FDPIR food groups is somewhat reduced, and participants are able to obtain
some commodity items only every two or three months. Given the tendency of
smaller programs to distribute from a warehouse (or to make home
deliveries to elderly and disabled participants), participants who obtain
food from small programs may have a greater selection in any given month. The
availability of particular food items also may be affected by
market conditions, local ordering practices, and shipping schedules to
local programs. Significant variations exist across regions in terms of
the availability of specific food items. Notably, in any given month,
households in the West were not able to select from as wide a range of
items as households in other regions. This may be related to the fact that
programs in the FNS Western Region follow different ordering procedures
that can cause delays in shipments. Or, it may be an unintended
consequence of local programs' effort to enhance the accessibility of the
program by extensively using tailgate distribution systems.
A variety of administrative controls are used to
maintain program integrity.
To maintain the integrity of FDPIR operations, local
programs have instituted controls related to the eligibility of
participants. First, consistent with program regulations, each household's
reported income is routinely verified. However, recognizing the low level
of income in areas served by the program, certification specialists
generally accept applicants' statements of financial resources. Second,
all sample programs make an effort to identify dual participation in FDPIR
and the Food Stamp Program, usually through an exchange of participation
lists with local food stamp offices. Third, even though a small number of
households reportedly received food for which they were not eligible, more
than half the sample programs have pursued claims against such households. Inventory
controls are maintained by following inventory procedures prescribed by
FNS. Sometimes programs use microcomputers and software provided by FNS.
While rates of inventory discrepancies are low across all programs, large
programs are most effective in controlling inventory discrepancies. Twenty-five
of the 30 programs included in the study were visited in Fiscal Year 1989
by either FNS or, in the case of programs administered under the
supervision of an agency of State government, by State personnel. Most
programs underwent an ME review during the two years prior to this
evaluation. Profile of Program
Participants
Many FDPIR households include elderly persons, and
single-parent households constitute a relatively small proportion of the
caseload.
More than one-third (38.9 percent) of all households
include an elderly person (that is, someone aged 60 or older), and nearly
one in five of the elderly live in an extended family household. The
elderly account for 62 percent of the one-person households. Half of the
households served by FDPIR contain children, but only one in ten is headed
by a single parent with one or more children, and one-fourth are single
adults living alone or together. The average FDPIR household contains 3.2
members.
Adult FDPIR participants have completed an average of
ten years of education. About half of FDPIR adults
were working; were looking for work, or were laid off and looking for
work. FDPIR households are poor by any
conventional standard.
Income levels for FDPIR households are very low. According
to
their case records, nearly one in ten households, do not have any income.
More than one-third of the households have gross income less than or equal
to 50 percent of the poverty level established for 1989. Only one in five
households have gross income that placed them above the poverty level. About
31 percent of the participating households receive income
from AFDC, Supplemental Security Income (SSI), or General Assistance
through the Bureau of Indian Affairs or State welfare
agencies. In sharp contrast to the Food Stamp Program, in which
approximately 40 percent of the participating households receive AFDC,
only five percent or so of FDPIR households receive payments from this
program. This latter point reflects an important difference in the pattern
of participation for American Indian households that receive this form of
public assistance.
The impoverishment of FDPIR households is reflected in
their level of liquid assets as well as their income. More than three fourths
of the household case records indicated no cash on hand and nearly as many
showed no financial assets of any kind. Among the households that had
liquid assets, the average value of theft' assets was $221.
Even though average distances to key destinations are
not great, many FDPIR participants experience transportation difficulties.
Although there is some regional variation in travel
distances, the nearest food store is usually within four to five miles of
participants' homes, whereas fresh meat and vegetables require driving to
a store four to eight miles away. Commodities usually could be obtained at
a site located six to nine miles from the participant's home.
More than two-thirds of participant households own a car
or truck. Nearly three-fourths of the respondents either travel in their
family's car to the store or to the FDPIR office, or get a ride with a
friend. About one-tenth of the households have to pay a friend to drive
them for shopping or for recertification.
Owning a vehicle does not necessarily eliminate
transportation problems. About one out of six (one-tenth of all households
in the sample) reported that they very often had difficulty getting where
they needed to go because of problems with their cars or trucks. More than
half of all households that owned a vehicle reported that they sometimes
could not travel because they lacked money to buy gas.
Four out of ten households without vehicles (one out of
eight sample households) reported that they very often had problems
getting where they needed to go because transportation was not available.
Nearly two-thirds said that at least sometimes they lacked money to pay
someone to drive them.
Dietary Needs of Program
Participants
In addition to program commodities, FDPIR households
rely upon food purchases, home food production, and other USDA programs to
meet their dietary needs.
FDPIR households included in this study spent an average
of $31 per month per household member for food to supplement the
commodities they received. About $24 of this amount was spent at food
stores, and 43 percent of the sample households reported that they only
bought food at grocery stores. Households that ate at restaurants or
bought take-out food generally had higher average incomes, indicating that
spending at restaurants and for take-out foods did not detract from
purchases at grocery stores.
About one-half of all FDPIR households produce some of
their food themselves. These food production activities include growing
fruits and vegetables, maintaining livestock for dairy and meat, raising
poultry for eggs, and hunting and fishing. These activities varied by
region.
Nearly half of the FDPIR households reported participating
in other food assistance programs, mostly in other USDA programs. About 70
percent of the households with school-aged children participated in the
School Lunch Program and 44 percent of these households participated in
the School Breakfast Program. Nearly one in six FDPIR households received
benefits under the WIC Program, representing 52 percent of the households
with a child aged five or less. Over one-fourth of all households with
elderly participated in one or more senior citizen assistance programs.
Seven out of eight respondents report that their
households had enough to eat during the survey reference month, while one
out of eight say that they sometimes or often did not have enough to eat.
Almost one-half of the study population reported that they
had enough to eat, but not always the kinds of food they wanted. Another
nearly 40 percent responded that they had enough of the kinds of foods
that they wanted. About one out of eight respondents reported that they
sometimes or often did not have
enough food to eat. Four out of five of these households reportedly were
without food or money to buy food five or six days per month. Two-thirds
of these households also skipped an average of more than four days of
meals per month.
Self-reports of inadequate food supplies varied greatly by
region. One-quarter of all FDPIR households in the West reported they
sometimes or often did not have enough to eat, and they represented three
out of five of all FDPIR households reporting this. Nearly three-fourths
of the FDPIR households who reported that they had to skip meals were from
the West.
Most FDPIR households report adequate food preparation
and storage facilities, with many of the households lacking basic
facilities living in the West.
FDPIR households generally reported having adequate
storage and cooking facilities. However, a significant minority lacked at
least one of five basic household facilities. One-fifth of the sample
reported not having hot running water in their home, 15 percent reported
no indoor running water and 7.3 percent of sample reported they had no
electricity. About one in ten of the FDPIR households reported having no
refrigerator, while 6.3 percent reported that they did not have either an
oven or cooktop stove.
The availability of basic housing facilities and food
preparation and storage resources also varied by region. Three-fourths of
those reporting no indoor running water lived in the West (more
than one-third of all western FDPIR households did not have indoor running
water). Ninety percent of the FDPIR households who reported having no
refrigerator or no electricity were located in the West, representing over
one-fifth of all FDPIR households located in that region. Finally, of
those who reported they did not have either an oven or cooktop stove,
two-thirds lived in the West.
Over half of all FDPIR households have at least one
adult with one or more nutrition-related health problems, and more than
one out of four households have at least one member who is supposed to be
on a special diet.
Over half of all FDPIR households reported that at least
one adult (a person 16 years old or older) had one or more nutrition
related health problems. More than one out of four households had at least
one member who was supposed to be on a special diet. Almost one-third of
all households reported at least one person with diagnosed high blood
pressure, about one-quarter with a member having diagnosed diabetes and
over one-fifth with at least one overweight household member. For diabetes
and obesity, these self-report rates fall below published estimates among
American Indians.
While these same nutrition-related problems were
identified by focus group participants as health issues of significant
concern to their reservations, there also were misconceptions and a lack
of information related to improving dietary habits. Participants expressed
frustration in changing dietary practices in view of family and community
preferences. They also perceived the need for more health and nutrition
education. These expressed needs go beyond the scope of nutrition
education services required by current Federal regulations.
Food Preferences of Program
Participants
Program participants express strong positive
preferences for almost all commodity food items.
Within each of 15 commodity food groups (juices, fruits,
vegetables, and so on), respondents indicated their household food
preferences. In all, preferences were given for 69 food items. In some
cases (among dried beans, for example) these preferences varied by region.
It was most notable that the number of respondents indicating a positive
preference for any given item almost always exceeded the number expressing
dislike for that item. Also, in the vast majority of cases, expressions of
dislike represented personal taste (for example, perceiving an item as too
sour or too sweet) rather than perceptions of poor food quality. No
particular concern other than taste was mentioned by five percent or more
of the respondents.
Comparison of FDPIR and the Food
Stamp Program
There are distinct differences in the household
characteristics of FDPIR participants versus American Indian food stamp
recipients.
Small households, two-thirds of which include an elderly
person,
constitute a much larger segment of the FDPIR caseload. Also, households
with earnings seem to be more likely to participate in FDPIR. In contrast,
due largely to joint application procedures, categorical eligibility for
food stamps, and the co-location of administrative offices, the majority
of American Indian families that receive AFDC are more likely to
participate in the Food Stamp Program than in FDPIR. Based
on a small exploratory survey of American Indian households who
participated in the Food Stamp Program in September 1989, a larger
proportion of this group had received benefits for the previous twelve
months, compared to FDPIR households interviewed for this study. This is
consistent with the fact that AFDC households would tend to have longer
spells of participation due to factors related to their need for
assistance (specifically, deprivation of parental support), whereas
households with earned income would experience shorter (though perhaps
repeated) spells of participation. Nearly half of FDPIR and
food stamp households in this study indicated that they participated in
both programs at different times. However, American Indian households that
leave the Food Stamp Program were more likely to apply for and receive
commodities under FDPIR than former FDPIR participants were to apply for
and receive food stamps. In fact, this may be due to more lenient
eligibility standards in FDPIR, such as a lack of a gross income
eligibility standard and the treatment of household resources,
particularly vehicles.
Providing commodities through FDPIR appears to be
less expensive than providing food stamps to households who would be
eligible.
Given that American Indian households that receive
commodities tend to be smaller and more likely to have earnings, they
would be more likely to receive smaller food stamp allotments than the
average food stamp household (either Indian or non-Indian), if they were
to apply for food stamps. A simulation of food stamp eligibility for FDPIR
households indicates that approximately 13 percent would not be eligible
because of the gross income limitation and other factors, such as the
treatment of vehicles as financial assets. Yet, our simulation suggests
that it was less expensive to provide commodities to all the households
that participated in September 1989 than it would have been to provide
food stamp allotments to the 87 percent estimated to have been eligible
for food stamps.
Travel distances are usually ten miles or less to
FDPIR distribution sites, local food stamp offices and food stores.
Most participants travel ten miles or less to purchase
food, or apply for commodities or food stamps. However, as many as
one-fourth of the participants in some regions have to travel more than 20
miles each way. The distance most participants must travel to purchase
food is not significantly greater than the distance to the commodity
distribution point. Also, while the distance to the local food stamp
office is usually greater than the distance to the FDPIR distribution
site, the difference is not great.
The combined availability of FDPIR and the Food Stamp
Program provides more flexibility and a better level of service for
American Indians than either program would individually.
The relative acceptability of FDPIR and the Food Stamp
Program was addressed by survey and focus group participants. The choice
of food stamps versus commodities is largely due to the flexibility and
wider selection of foods, including fresh produce. On the other hand, a
large segment of FDPIR participants interviewed for this study indicated
that they applied for commodities rather than food stamps because they
perceived the value of the benefit to be greater. A smaller group of FDPIR
respondents felt that the effort to apply for and participate in the
commodity program was less than that required in the Food Stamp Program.
Thus, each program seems to offer participants distinct and readily
identifiable advantages that they deem to be important. The
distinct patterns of participation suggest that the combination of FDPIR
and the Food Stamp Program accommodates a diverse set of food assistance
needs among different types of American Indian households. For example,
regulations that require welfare agencies to allow AFDC applicants to
submit a single application for AFDC and food stamps, and that establish
categorical food stamp eligibility for many AFDC families, make it easier
for American Indians who receive AFDC to also obtain food stamps. However,
the availability of FDPIR provides other types of households, such as the
elderly, with a relatively simple application process, and therefore,
easier access to food benefits. The simulation of food
stamp eligibility suggests that more households - especially small
households with elderly members and households with earned income - can be
served at less cost by FDPIR than by the Food Stamp Program. On the other
hand, the Food Stamp Program appears to be effective in reaching
particular types of households (especially AFDC families) for whom the
relative costs of obtaining food assistance are lower and their desire for
flexibility in selecting foods is greater. Thus, the combination of FDPIR
and the Food Stamp Program tends to provide a better level of service for
this population than either program would individually.
Last modified: 12/04/2008
|
