Barriers to Participation in the Food Stamp Program Among Food Pantry Clients in Los Angeles

Abstract

Substantial numbers of food pantry clients are eligible for food stamps but do not receive them. Background characteristics of 14317 food pantry users in Los Angeles were analyzed to provide information helpful in food stamp outreach programs. Ninety percent of food pantry users were living well below poverty level, 59% were Hispanic, and 44% were homeless. Only 15% of the food pantry clients received food stamps, with homelessness and limited English language skills acting as barriers to food stamp program participation.

Individuals are considered to be food insecure when they are unable to obtain the quality and quantity of food needed for basic health and well-being.1 Food pantries are a type of private emergency food assistance service that provides low-income households with packages of food items usually requiring additional preparation. Food pantry users represent a group at highest risk for being food insecure who can benefit from participation in the USDA’s food stamp program.2^–7 The majority of food pantry clients are eligible to receive food stamps because of low income, but a substantial number do not receive them.8^–10

In 2001, 1.46 million adults in California experienced food insecurity and had incomes below 130% of the federal poverty level, yet 1.21 million were not getting food stamps.11 In both California and Los Angeles County, more recent statistics indicate that about half of the eligible participants do not receive food stamps.12^–15 Barriers to food stamp participation in California and Los Angeles include a finger imaging requirement, a lengthy and complex application process, and lack of knowledge about eligibility.16^,17 Nationally, changing eligibility restrictions and the stigma associated with participation also act as barriers to increasing food stamp enrollment.18 This research project compares sociodemographic characteristics of food pantry clients who are food stamp recipients versus nonrecipients to provide information helpful in improving food stamp outreach and enrollment.

METHODS

Data were collected on 14317 clients attending 2 different food pantries in Pomona and Ontario (inland cities in the greater Los Angeles area) during 2003. Bilingual food pantry workers interviewed clients to gather information on eligibility for emergency food assistance, and the data were later entered into the Access software program (Cisco Systems, San Jose, Calif). Data were pooled, as sociodemographic profiles were similar for both communities, with a greater proportion of Hispanics (65% for Pomona and 60% for Ontario) and a higher percentage having less than a high school education (45% for Pomona and 38% for Ontario) than Los Angeles County.19 The number of people living in poverty was higher than the national average (12%) for residents in Pomona (22%), Ontario (16%), and Los Angeles (18%). Food stamp participation in the sample population was measured by self-reported food stamp income.

Income, housing, ethnic background, and homelessness were the sociodemographic characteristics analyzed for frequency in the sample population. The following hypotheses were tested by regressing the binary outcome “food stamp participation” on food pantry client sociodemographic variables: (1) single-parent families with children would be more likely to receive food stamps; (2) English language ability would encourage food stamp program participation; and (3) homeless clients would be less likely to receive food stamps.

RESULTS

Table 1 ▶ provides background information on the food pantry clients in the study. Ninety percent of client head of households reported incomes less than $1000 per month, including 44% who reported no income at all. The number of food stamp recipients in this study reporting no income at all was 52%, which is twice the national average.20

TABLE 1—

Characteristics of Food Pantry Clients

	% of Total Sample
Ethnicity (n = 14 317)
African American	14.20
American Indian	0.30
Asian/Pacific Islander	0.70
Non-Hispanic White	25.30
Hispanic	58.80
Other	0.70
Age groups, y (n = 14 011)
17–34	30.57
35–49	46.52
50–64	17.38
65–84	5.52
Income from food stamps (n = 14 273)
No food stamp income	86.75
$1–$100	1.94
$101–$150	3.36
$151–$200	2.04
$201–$250	2.14
$251–$300	1.63
$301–$350	0.92
$351–$450	0.82
> $450	0.41
Language (n = 14 317)
Asian	0.00
English	64.10
English and Spanish	7.90
Other	0.10
Spanish	27.90
Education (n = 13 135)
No school	3.03
Kindergarten–6th grade completed	16.91
7th–11th grade completed	36.00
High school graduate/GED	42.00
College graduate	2.06
Housing cost as a % of income (n = 14 135)
No housing cost	20.60
1–30	7.50
31–50	17.20
51–75	21.60
76–100	14.30
More than > 100%	18.80
Binary variables
Shared housing	33.18
Single parent, father	0.45
Single parent, mother	2.83
Homeless	44.20
Receive food stamps	19.35
Total monthly income, $ (n = 14 202)
No income	43.90
1–500	17.40
501–1000	28.80
1001–1500	7.30
1501–2000	2.00
> 2000	0.60

A total of 25% of the food pantry clients reported spending at least 75% of their income on housing, and 22% of food pantry clients shared housing to make it more affordable. Clients reporting housing costs greater than 100% of their income were sharing housing.

Just under 60% of the pantry users were Hispanic. Non-Hispanic Whites represented 25.3% of the food pantry users, African Americans another 14.2%, and others 11%. Fewer (20.7%) of the Hispanic pantry clients were homeless compared with non-Hispanic Whites (46%), due partly to the fact that Hispanics in Los Angeles form multifamily households, including relatives, friends, and lodgers, to reduce housing costs.21

Results of the binary outcome model are found in Table 2 ▶. The results confirm the hypotheses that single parents (P < .05) and clients with better English language ability were more likely to receive food stamps (P < .021), and that being homeless made it less likely that a pantry user would also be a food stamp recipient (P < .001).

TABLE 2—

Covariates of the Probability That Food Pantry Clients Receive Food Stamps

	Mean (SEM)	% Receiving Food Stamps	Bivariate Odds Ratio	Multivariate OR (95% CI)	P
Overall		14.13
Constant				1.766	.004
Continuous variables
Age, y	42.38 (0.12)			0.971 (0.966, 0.976)	.001
Total income/mo	$584.70 ($102.07)			1.000 (1.0, 1.0)	.358
Housing cost/mo	$514.69 ($1.90)			0.999 (0.999, 0.999)	.001
Education, y	9.85 (0.031)			0.972 (0.949, 0.995)	.019
Binary variables
Single parent, father		15.2	1.18	2.672 (1.068, 6.681)	.036
Single parent, mother		29.3	2.80	2.488 (1.670, 3.706)	.001
English ability		13.8	0.799	1.257 (1.035, 1.527)	.021
Homeless		5.7	0.236	0.217 (0.188, 0.251)	.001

Note. CI = confidence interval; OR = odds ratio. Model χ² = 723.606. Model significance Nagelkerke R² = 0.139, 0.139.

The results also indicate that food pantry clients that were better educated (P<.019) and older clients (P<.001) were less likely to receive food stamps. Shared housing did not have a significant impact on whether a food pantry user would be likely to receive food stamps.

DISCUSSION

These data are consistent with California and national statistics indicating that food stamp recipients tend to be younger, less educated, single-parent households struggling with high housing costs.20^–22 What is unique to this group of food pantry clients is that homelessness and limited English language ability appear to act as barriers to food stamp participation.

Our findings support our hypotheses that food pantry users in the study population who are homeless and those with limited English language skills are less likely to receive food stamps. Lack of permanent address prohibits many homeless individuals from being able to sign up for and receive food stamp benefits. Undocumented immigrants are ineligible for food stamps and comprise many of the limited English language food pantry clients who are food stamp nonrecipients in Los Angeles.12^,23 However, children of undocumented immigrants born in the United States are eligible to receive food stamps and can be enrolled through additional outreach efforts.12 Information from this study can be used by health care professionals and policy-makers to design food stamp outreach efforts to reach underserved populations such as the homeless and individuals for whom English is a second language.