The food stamp application process office operations and client experiences

              m.ml'
United States
Department of
Agriculture
Food and
Nutrition
Service
Office of
Analysis and
Evaluation
The Food Stamp
Application Process:
Office Operations and
Client Experiences
COMPLETE!
tf-
THE FOOD STAMP APPLICATION PROCESS:
OFFICE OPERATIONS AND CLIENT EXPERIENCES
Authors:
Susan Bartlett
Nancy R. Burstein
Gary Silverstein
Dorothy Rosenbaum
April 1992
Prepared for: Prepared by:
Theodore F. Macaluso, Ph.D. Abt Associates Inc.
U.S. Department of Agriculture 55 Wheeler Street
Food and Nutrition Service Cambridge, MA 02138
Office of Analysis & Evaluation
3101 Park Center Drive, 2nd Floor
Alexandria, VA 22302
This study was sponsored by the Office of Analysis and Evaluation, Food and Nutrition Service, U.S. Department of
Agriculture as part of its ongoing research agenda (contract number S3-3198-8-34). Points of view or opinions stated
in this report are those of the authors and do not necessarily represent the official position of the Food and Nutrition
Service.
TABLE OF CONTENTS
Page
EXECUTIVE SUMMARY i
Chapter One INTRODUCTION 1
Chapter Two THE FOOD STAMP APPLICATION PROCESS 9
2.1 Obtaining Information and an Application 10
2.2 Filing a Food Stamp Application 14
2.3 Certification Interviews 21
2.4 Providing Necessary Verification 24
2.5 Eligibility Determination and
Benefit Calculation 29
Chapter Three COSTS OF THE APPLICATION PROCESS 31
3.1 Monetary Costs 31
3.2 Time Costs 58
3.3 Summary and Conclusions 69
Chapter Four COMPLETION AND DROPOUT RATES AND
REASONS FOR DROPPING OUT OF THE
APPLICATION PROCESS 71
4.1 Completion Rates by Site 71
4.2 Reported Reasons for Dropping Out 75
4.3 Reasons for Dropping Out: Perceived
Ineligibility Versus Process Dropouts 85
4.4 Relationship to Household Characteristics 90
4.5 Summary and Conclusions 105
Chapter Five POTENTIAL CHANGES TO THE FOOD STAMP
APPLICATION PROCESS Ill
5.1 Problems and Burdens Encountered in the
Application Process 113
5.2 Potential Modifications to the Food Stamp
Application Process 115
APPENDK A SUPPLEMENTARY EXHIBITS A-l
APPENDDC B CONSTRAINTS IMPOSED ON THE FOOD STAMP
APPLICATION PROCESS BY CURRENT
LEGISLATION B-l
BIBLIOGRAPHY
LIST OF EXHIBITS
Exhibit 1.1
Exhibit 3.1
Exhibit 3.2
*
Exhibit 3.3
Exhibit 3.4
Exhibit 3.5
Exhibit 3.6
Exhibit 3.7
Exhibit 3.8
Exhibit 3.9
Exhibit 3.10
Exhibit 3.11
Exhibit 3.12
Exhibit 3.13
Exhibit 3.14
Exhibit 3.15
Exhibit 3.16
Page
Applicant Survey: Number Interviewed by
Site and Stratum 7
Total Out-of-Pocket Costs 33
Components of Out-of-Pocket Costs of
Application 36
Components of Out-of-Pocket Costs of
Application: All Respondents 38
Components of Out-of-Pocket Costs of
Application: Approvals 39
Number of Contacts with the Food Stamp
Office 42
Number of In-Person Visits to the Food
Stamp Office 43
Number of Telephone Calls and Letters to
the Food Stamp Office 45
Transportation Costs for Office Visits 47
Babysitting Expenses for Office Visits 49
Forgone Wages for Office Visits 50
Costs of Telephone Calls 52
Costs of Documentation 53
Components of Projected Out-of-Pocket Costs
of Application: Dropouts 57
Total Time Costs 59
Components of Time Costs of Application 61
Components of Time Costs of Application:
All Respondents 62
w
LIST OF EXHIBITS
(continued)
Page
Exhibit 3.17
Exhibit 3.18
Exhibit 3.19
Exhibit 4.1
Exhibit 4.2
Exhibit 4.3
Exhibit 4.4
Exhibit 4.5
Exhibit 4.6
Exhibit 4.7
Exhibit 4.8
Exhibit 4.9
Exhibit 4.10
Components of Time Costs of Application:
Approvals 64
Time Costs of Substantive Activities 65
Time Costs of Travel and Waiting 67
Percent Completing Each Step of the
Application Process 73
Main Reason Non-Approved Respondents
Gave for Not Receiving Food Stamps 77
Factors Related to Discontinuance of
Food Stamp Application Process 79
Factors Related to Discontinuance of
Food Stamp Application Process, by County 82
Documentation Required and Provided:
Applicants Denied for Failing to Provide
All Documents Versus Approved Applicants 84
Classification of Reasons for Dropping Out
of the Application Process, by Stage of
the Process Completed 86
Outcome of the Application Process for
All Who Contacted the Food Stamp Office 89
Application Process Dropouts and
Completers: By Household Size 92
Application Process Dropouts and
Completers: For Households with Special
Needs 94
Application Process Dropouts and
Completers: For Applicants with
Particular Demographic Characteristics 96
Exhibit 4.11
Exhibit 4.12
Exhibit 4.13
Exhibit 4.14
Exhibit A. 1
Exhibit A.2
Exhibit A. 3
Exhibit A.4
Exhibit A. 5
Exhibit A. 6
Exhibit B.l
LIST OF EXHIBITS
(continued)
Page
Application Process Dropouts and
Completers: By Income Source 98
Application Process Dropouts and
Completers: By Cost of Application 100
Application Process Dropouts and
Completers: By Available Supports 102
Application Process Dropouts and
Completers: By Estimated Potential
Food Stamp Allotment 104
Main Reasons Applicants Gave for Not
Receiving Food Stamps: Southern
Urban County A-3
Main Reasons Applicants Gave for Not
Receiving Food Stamps: Southern
Rural County A-4
Main Reasons Applicants Gave for Not
Receiving Food Stamps: Midwestern
Large Urban County A-5
Main Reasons Applicants Gave for Not
Receiving Food Stamps: Midwestern
Mid-sized County A-6
Main Reasons Applicants Gave for Not
Receiving Food Stamps: Midwestern
Rural County A-7
Selected Household Characteristics of
Survey Respondents and the National
Food Stamp Caseload A-8
Title 7, United States Code: Requirements
for Obtaining Information About the Food
Stamp Program and/or an Application Form B-4
Vi
LEST OF EXHIBITS
(continued)
Page
Exhibit B.2
Exhibit B.3
Exhibit B.4
Exhibit B.5
Exhibit B.6
Title 7, United States Code: Requirements
for Filing A Food Stamp Application B-6
Title 7, United States Code: Requirements
for Conducting Food Stamp Eligibility
Determination Interviews B-8
Title 7, United States Code: Requirements
for Verifying Food Stamp Application
Information B-9
Food, Agriculture, Conservation and Trade
Act of 1990 (P.L. 101-624): Requirements for
Food Stamp Application Processing B 13
Food, Agriculture, Conservation and Trade
Act of 1990 (P.L. 101-624): Requirements for
Audits, Demonstrations and Studies Regarding
Food Stamp Application Processing B-15
VII
ACKNOWLEDGMENTS
This study would not have been possible without the cooperation and support of the state
and local food stamp office staff in the two states and five local offices that generously agreed
to participate in this study. We promised confidentiality to the participating sites and thus cannot
thank individuals directly. However, both state and local staff spent considerable time with our
researchers over the course of numerous site visits. They allowed us to observe all aspects of
their food stamp application operations, kept reception logs of all individuals who contacted them
about benefits, and answered endless questions about the structure of the application process in
their sites. Staff also provided valuable insights concerning the study's findings.
Staff of the Food and Nutrition Service provided guidance and support throughout the
entire study. As project officer, Theodore Macaluso was always available to answer design,
policy, and operational questions as well as provide valuable reviews as the work progressed,
and we thank him for all his help. Steven Carlson, Marilyn Carpenter, Barbara Fay, Arthur
Foley, Abigail Nichols, Thomas O'Connor, Carol Olander, Joseph Pinto, and Judith Seymour
attended briefings, reviewed ongoing work, and provided valuable insights at various points
during the project.
Many persons at Abt Associates contributed to the study. In particular, William
Hamilton provided guidance at every step along the way. Mary Beth Sullivan was a valuable
project member in the design and data collection phases of the project. Rick DeFriesse and
Peter Goodrich developed analysis files from the states' automated data bases. Finally, Eileen
Fahey led the production team that prepared the report.
V/it
EXECUTIVE SUMMARY
As part of its on-going efforts to increase the efficiency of the Food Stamp Program and
eliminate barriers to access, the Food and Nutrition Service initiated this study of the food stamp
application process. The study's primary goal is to learn about ways in which federal, state, and
local policies and procedures may encourage or inhibit individuals from completing the process,
or impose unnecessary costs and burdens as a condition of participation.
The study was designed to achieve four specific objectives:
• Describe the application process as it typically works;
• Develop hypotheses concerning how the application process may affect program
participation;
• Collect evidence on completion rates and the applicants' time and monetary costs
of the application process; and
• Identify potential changes in the application process that might facilitate applicants
completing the process or reduce the costs and burdens of the process.
Study Design
We examined the application process in two states in 1990, one in the Midwest and one
in the South. In the Midwestern state, two urban and one rural county participated in the study.
One urban and one rural county were studied in the Southern state.
The study is necessarily exploratory in nature, as no previous work has examined the
application process in detail or systematically studied the extent to which individuals interested
in the Food Stamp Program actually complete the application process. The design involved two
distinct data collection efforts. The process study involved in-depth interviews with state and
local food stamp officials and extensive observations conducted over a three-day period in each
of the five local offices in the study. The result of this component of the study was a detailed
description of the application process as it worked in our sites. The process is essentially
structured similarly in all the sites, but differences do exist that have a marked effect on
applicants.
iy
The quantitative component of the study provides preliminary data on the rates at which
applicants drop out of the process at different stages, and the reasons they do so. Data for this
part of the study came from four sources. First, logs were maintained to collect the names of
everyone who contacted the food stamp office during the study period to inquire about food
stamp benefits or Aid to Families with Dependent Children (AFDC), or to inquire in general
about assistance. Second, using the states' automated casefile data, we determined whether each
individual completed the food stamp application process and if not, where they dropped out of
the process. These data allowed us to calculate the rates at which applicants completed the
different stages of the food stamp application process. Third, an applicant survey ascertained
why those who dropped out of the process did so. It also collected extensive information on the
number of trips an applicant made to the food stamp office, the time required, and out-of-pocket
expenses involved. Finally, we obtained detailed information from applicants' casefiles
concerning verification requirements and the outcome of the application process.
The Food Stamp Application Process
The food stamp application process requires the applicant household and the food stamp
agency to complete a number of activities in order for the household to be considered eligible
for the Program. The applicant activities are:
• Obtaining information about the Program and an application form;
• Completing and filing the application;
• Completing the certification interview; and
• Providing verification of information.
Once an applicant completes these steps, the caseworker determines whether the
household is eligible to receive food stamps, and if so, the amount of the monthly benefit.
The study considers the application process to begin when a household calls or visits the
local office to inquire about obtaining food stamps. In the urban offices in this study, applicants
ii
<
generally visit the office. In the rural offices, interested individuals are more likely to begin the
application process by calling to request information and application materials.
The second step in the application process involves completing the application (or at a
minimum, filling in name, address, and signature) and filing it with the local office. Often
individuals do this at the same time they pick up the application, though some take the
application home to complete it. Those who do not complete the application in the office may
return it either in person or by mail. Benefits are calculated from the date the application is
filed, so office staff in the study sites often encourage people to return to the office with the
application right away rather than delay the start date because of a slow mail system.
All applicants must complete an interview with a caseworker in which details of their
circumstances are recorded. This interview is generally conducted in person, though applicants
may request a telephone interview or a home visit. The interview can take anywhere from 20
minutes to 1 1/2 hours. Persons interested in applying for AFDC and Medicaid may do so at
the same time they apply for food stamps and thereby complete only one certification interview.
The length of the interview is thus affected by the number of programs for which the household
is applying.
The final step of the application process involves documenting the information provided
in the certification interview. In general, applicants must provide documentation of their
household composition, income, and expenses. AFDC and Medicaid applicants also have to
verify information relating to their children and medical circumstances.
All people who are approved to receive food stamps complete these four activities,
though depending on the procedures in the local welfare office some activities may be done
differently and some may occur simultaneously. For example, in four of the five study sites,
most applicants visit the office to obtain information and office staff encourage them to do so.
The other office, which is a small rural office, considers that it is unfair to ask those interested
in benefits to visit the office. Staff encourage people to call for information and most do so.
Another difference among offices is that two of the offices screen applicants for gross eligibility
iii
either when they request an application or Hie it. In these offices most potential applicants who
are obviously ineligible choose not to proceed with the application process.
Major Quantitative Findings
Costs' incurred in the application process. Applicants and potential applicants spend,
on average, almost five hours performing application activities. Approximately two hours of
this time is spent filling out the application, meeting with food stamp office workers, and
obtaining required documents. The other three hours is spent in transit to the office and waiting
to meet with office staff. The further an applicant gets in the process, the more time it takes.
Those who are approved for benefits spend an average of six hours in total.
Applicants and potential applicants incur out-of-pocket expenses totalling an average of
$10.40. Approximately $7.40 of this is spent visiting the food stamp office one or more times
during the course of the application process. The costs of these visits result primarily from
transportation costs and forgone wages. Another $3 is spent obtaining the documents necessary
to verify the household's circumstances.
Some applicants spend considerably more money and time than the average during the
application process. Approximately 5 percent of all applicants and potential applicants said they
incurred $50 or more in out-of-pocket expenses and forgone wages and spent 12 or more hours
completing the activities required of the process.
Completion rates at each step of the application process. Overall, 58 percent of all those
who contact the food stamp office are approved to receive benefits. Another 9 percent complete
the application process, but are denied benefits. The remaining 34 percent who contact the
office do not complete the application process. Of these, 19 percent do not even file an
application, 11 percent file an application but do not complete the certification interview, and
4 percent attend the interview but do not submit all the documents requested to verify their
'Applicants are not charged for applying for food stamps (such charges are prohibited by
law). The "costs" examined here are the time and money spent to go through the steps in the
application process.
iv
All
household's circumstances. These numbers indicate that most who contact the food stamp office
are approved to receive benefits. Examining the figures from another perspective, those who
are approved represent 71 percent of those who actually file an application and 87 percent of
those who complete the application process.
Reasons for not completing the application process. Of the 34 percent who did not
complete the application process, approximately half reported that they were either told by office
*taff that they were ineligible or perceived themselves ineligible. The other half thought they
would probably be eligible but still did not fulfill some procedural requirement of the application
process. This means that between 9 and 18 percent of all those who contacted the food stamp
office to inquire about benefits thought they were potentially eligible for benefits, but either
found some aspect of the application process difficult or decided that the benefits they would
receive were not worth the effort it would take to complete the application process.
Factors most likely to lead to not completing the process. Applicants who thought they
were potentially eligible for benefits reported that several aspects of the application process were
particularly likely to cause them not to complete the process. These factors are:
• Time and "hassle" involved;
• Length of the process or particular aspects of the process;
• Problems getting to the office;
• Confusion about the process;
• Long waits at the office; and
• "Unobtainable" documents required to verify the household's circumstances.
Some demographic groups are somewhat more likely than others to be deterred by the
procedural requirements of the application process. Employed persons were less likely than
others to complete the process even when they thought they were potentially eligible, particularly
compared to those already receiving other governmental assistance. Those who failed to
complete the process were also more likely to be eligible for only small food stamp allotments.
XI"
The observed differences among these groups are relatively small and in all groups, more were
approved for benefits than dropped out of the application process while believing they were
potentially eligible for benefits.
These patterns suggest that people's ability or willingness to complete the application
process varies in ways that do not correlate highly with characteristics that are often mentioned
as "risk" factors, such as being elderly or having limited education. The structure of the
application process does seem to matter: the offices in which applicants were more likely not
to complete the process even though they perceived they were eligible for benefits were the ones
that our observers found to be the most confusing and inconvenient. Nonetheless, the data do
not identify specific aspects of the process that clearly inhibit participation, nor do they allow
forecasts of the changes in participation that might result from altering office procedures.
Potential Changes to the Food Stamp Application Process
The data from this study suggest some potential changes to the food stamp application
process that might improve access to the Program. Some of the suggestions would require
additional resources which are probably not readily available in today's economic environment.
States are facing large increases in Program participation along with severe budget cutbacks.
Nonetheless, whether they can be considered now or must be deferred, this study provides some
ideas for making the application process more accessible. Like most policy changes, the ideas
presented in this study would involve trade-offs~between protecting program integrity,
promoting administrative efficiency, and improving accessibility. These trade-offs must be
considered carefully before any changes are made.
The potential changes that seem most likely to have the greatest effect are within the
control of state and local policies; they do not require changes in federal laws or regulations.
Possibly useful changes would involve reducing the number of visits required, reducing waiting
time, cutting applicants' out-of-pocket expenses, and providing more information and assistance
to applicants at various points in the process. We cannot be sure that fixing these problems
would lead more people to complete the application process, but they do appear to be reasonable
areas in which to focus attention.
vi
yjv
Chapter One
INTRODUCTION
The Food and Nutrition Service initiated the current study as part of its on-going efforts
to facilitate participation in the Food Stamp Program and to improve the efficiency of Program
operations. The Food Stamp Program is administered by state or county agencies and operated
through local food stamp offices. The application process, like other aspects of the Program's
day-to-day operations, is defined by local policies and procedures but is broadly structured by
current legislation, Federal regulations, and state policies.
The Federal legislation and regulations implicitly reflect several, sometimes competing,
policy goals of the Food Stamp Program. The goals can be characterized as:
• Providing client services;
• Maintaining program integrity; and
• Achieving administrative efficiency.
Providing services to needy clients requires that benefits be easily accessible to all eligible
households, without undue delay or burdens on the household. Maintaining the integrity of the
Program necessitates ensuring that services are provided only to the intended beneficiaries.
Achieving administrative efficiency in processing applications requires minimizing administrative
costs of the program. These goals sometimes conflict, requiring tradeoffs that prevent fully
achieving any one goal.
This study is intended to provide information concerning federal, state, and local policies
and procedures that may encourage or inhibit clients from completing the application process or
impose unnecessary costs1 and burdens as a condition of participation. To the extent that the
policies and procedures have such an impact, the study considers ways to alleviate the problems.
'Applicants are not charged for applying for food stamps (such charges are prohibited by
law). Applicants may need to spend time and money to comply with the requirements of the
application process, however. Throughout this report, we use the term "costs" to refer to these
incidental time and money expenditures.
1
Previous studies sponsored by the Food and Nutrition Service, and others, including the
Congressional Budget Office, the General Accounting Office, and various state and local
agencies, have indicated that the Food Stamp Program is not serving all the eligible population.
Although some of these studies suggest that certain features of the application process may be
hindering participation, they fail to provide conclusive evidence. None of the studies are
systematic or detailed enough to provide a thorough understanding of the application process.
Nor do they provide reliable estimates on the number of applicants affected, whether any are
actually kept from participating in the Food Stamp Program, and what aspects of the process
present particular difficulty.
Study Objectives
The study has been designed to achieve four specific objectives:
• Describe the application process as it typically works;
• Develop hypotheses concerning how the application process may affect
Program participation;
• Collect evidence on the number of people completing each step of the
application process and the cost of the process; and
• Identify potential changes in the application process that might facilitate
applicants completing the process.
The study is exploratory in nature. No previous work provides even a systematic
description of the process and how it may affect participation. Thus, considerable effort was
spent assembling descriptive information and developing conceptual structures. It is hoped that
this work will provide direction for future studies designed to test hypotheses concerning factors
that encourage participation or future demonstrations to test potential changes in the application
process.
It is equally important to say what the study is net. First, it is not a compliance
investigation. That is, we attempted to learn about the general policies and practices in specific
food stamp offices, but not to determine whether they would be judged to be in compliance with
the applicable food stamp regulations. Second, the study does not attempt to test the validity
of criticisms of the application process that have been made by client advocates or other groups.
Study Design
Study Sites. The study was carried out in 1990 in five local offices in two states. A
Midwestern and a Southern state volunteered to participate in the study. Within the Midwestern
. tate, a large urban county, a mid-sized urban county and a rural county participated. One large
urban and one rural county participated in the Southern state. We guaranteed the sites
confidentiality as a condition of participating in the study. Therefore, we refer to them
throughout the report by descriptive labels, specifically - Midwestern Large Urban county,
Midwestern Mid-sized Urban county, Midwestern Rural county, Southern Urban county, and
Southern Rural county.
We selected the states and local offices based on several broad criteria. First, we did not
want unusual states and sites. The sample size for the study is small and, even though statistical
generalizability would be impossible, we wanted the information to be as broadly applicable as
possible. Second, we sought geographic diversity in the states and local offices to get a varied
picture. The two states are located in different Food Stamp Program regions and have
significantly different industrial mixes and populations, one being a Midwestern industrial state
and the other a Southern, more rural state. We selected three urban and two rural sites to
ensure additional diversity. Finally, the data collection strategy required that the local offices
be large enough to provide a sufficient number of cases for the study. Thus, we required that
the small offices have caseloads of approximately 1,000. This restriction only affected our
choice of rural sites. All the urban sites were substantially larger than this minimum.
Data Sources. As dictated by the exploratory nature of the study, the design involved
two distinct data collection efforts, a process study and a quantitative component.
The process study involved assembling descriptive information concerning the
organization and structure of the application process. Data for the study came from three
sources. At the beginning of the study, we interviewed state food stamp staff and local office
staff including the office director, food stamp and AFDC supervisors, and some caseworkers,
clerks, and receptionists. The purpose of these visits was to understand how the food stamp
application process worked in the individual study sites. We focused on examining the steps in
the process, the order in which they were accomplished, and the responsibilities of applicants
and office staff at each stage. We also discussed with staff their perceptions of the problems
associated with the process, points where they thought applicants might have particular difficulty,
and what if any points imposed unnecessary burdens on applicants.
The second source of data for the process study came from 3-day observations in each
of the local offices in the study. During these visits, we spent time observing all facets of the
application process. We sat in the waiting room and observed applicants as they came into the
office, filled out applications, and waited to ae seen by office staff. We also spent time with
the receptionists and switchboard operators, observing the initial interaction between potential
applicants and staff. Finally, we sat in on numerous certification interviews and observed the
types of data collected, the information presented to applicants, and the nature of the interaction
between the caseworker and applicant.
The final source of process data came from visits conducted with the state and local
offices after the preliminary data analysis was completed. We shared our findings with staff,
discussed their interpretation and implications, and solicited the staffs recommendations for
potential changes in the application process.
The quantitative component of the study was designed to provide preliminary data on the
rates at which applicants complete the different stages of the application process and the reasons
why some do not complete the process. Four sources of data contributed to this analysis. First,
the study sites maintained reception logs that recorded the names of all people who called or
visited the office to obtain information or an application for the Food Stamp Program or AFDC.
(AFDC applicants were recorded on the assumption that most would also apply for food stamps.)
Individuals making general inquiries about assistance were also recorded. The study sites began
maintaining the logs in May, 1990 and continued for a period of one to four and one-half
months, depending on how long it took to reach the targeted number of names. The goals were
based on our assumptions about the number of cases required to obtain reasonable estimates of
the desired variables. The goal was 1,700 names for each of the urban sites and 1,130 names
and for each of the rural sites.
Second, we used the state's automated casefile data to determine how far each applicant
got in the process. These data allowed us to calculate the rates at which applicants completed
the different stages of the application process.
In order to examine why individuals did not complete the application process, we
conducted an applicant survey. The survey asked those respondents who did not complete the
process the reasons they did not do so. The survey also collected extensive information on the
activities required of the application process including the time taken and expenses incurred.
The sample for the applicant survey was a stratified random sample, stratified according
to how far an individual got in the application process. The stratification was necessary because
we assumed that most people who contacted the food stamp office would be approved for
benefits. Given the study objectives, we wanted to get as reliable estimates as possible for those
who did not complete the application process. The strata that we used were:
• Those who contacted the office but never filed an application;
• Those who filed an application but did not complete the certification
interview;
• Those who completed the interview but did not furnish all required
documentation;
• Those approved for food stamp benefits; and
• All other denials (most of whom were circumstantial denials).1
The interviewing goal was to complete 30 interviews per stratum in each of the three
urban sites. In each of the two rural sites, the goal was to complete 20 interviews per stratum.
'Most households classified as circumstantial denials had income or resources (e.g., savings
accounts, vehicles) that exceeded Food Stamp Program eligibility requirements. Examples of
other types of circumstantial denials include households living outside the office's service area
or ineligible students.
We completed a total of 706 interviews. Exhibit 1.1 presents the distribution by site and stratum
and shows that the targeted objectives were met for most groups.1
Finally, caseworkers in each of the sites extracted casefile data for all the sampled cases
to provide additional information about the application process, including data on the verification
documents requested and provided, and more detailed tracking information.
.Man of the Report
Chapter 2 presents the results of the process study and describes how the application
process works in our study sites. The discussion is synthesized from our studies of the five
different sites. Most aspects of the process are similar across all sites, though the discussion
highlights the significant differences.
Chapter 3 describes the application process from another perspective, focusing on what
is required of potential applicants at each step of the process. The chapter presents estimates
of the costs of the application process, in terms of out-of-pocket expenses and time spent in the
various required activities. We use these estimates in Chapter 4 to examine whether the costs
of the process affect the likelihood that an individual will complete or drop out of the process.
Chapter 4 turns to the question of how many people who are potentially interested in food
stamps fail to get them and why. We first present estimates of the percent of individuals who
complete the different stages of the application process, and determine where the dropout rates
are highest. We then examine the reasons people give for not completing the process, with
particular attention to those who perceive themselves eligible for benefits but still do not
complete the application process. Finally, we attempt to explain why some complete the
process, while others do not. To do this, we examine the outcome of the process for households
with different characteristics.
'We used a quota approach, in which interviewers were provided with a substantial list of
potential respondent names and instructed to stop interviewing when they reached the quota.
Thus there is no meaningful "response rate" in the usual sense of that term. This means that
we cannot fully judge the representativeness of the sample or the likely importance of non-response
bias. No problems of non-representativeness are known to exist, however.
Exhibit 1.1
APPLICANT SURVEY: NUMBER INTERVIEWED BY SITE AND STRATUM
Southern State Midwestern State
Large
Urban
County
Rural
County
Large
Urban
County
Mid-Sized
Urban
County
Rural
County TOTAL
Strata
Did not file
application 31 22 30 30 18 131
Did not
attend
certification
interview 50' 3' 27 29 22 131
Did not
complete
documentation 39 22 32 34 21 148
Approved 32 20 33 31 23 139
Other denials 25. U 22 3.3 2£ 331
TOTAL 187 93 159 157 110 70G
1 We oversampled this group in the Urban county because few individuals did not complete
their interview in the Rural county during the study period. We deemed it desirable to
get as good estimates as possible for this stratum in the Southern state.
7
Chapter 5 suggests some potential changes in the application process that seem likely to
reduce the costs or burden of the application process. The potential changes may also increase
the chance that applicants will complete the entire food stamp application process, but this is
more difficult to say without further study.
Chapter Two
THE FOOD STAMP APPLICATION PROCESS
The food stamp application process requires that households complete four basic steps
in order to be considered eligible for the Program. These steps are:
• Request information about the Program and obtain an application if
interested;
• Complete and file an application;
• Complete an eligibility determination interview; and
• Document household circumstances.
Individuals interested in obtaining information about food stamp benefits may visit or call
the food stamp office. Applications may be filed in person, or mailed to the agency. In our
study sites, the majority of applicants obtain and file an application during a single office visit.
When they come into the office, applicants are greeted by a receptionist who determines whether
they are interested in applying for food stamps, Aid to Families With Dependent Children
(AFDC), or Medicaid. The receptionist provides an application and encourages persons to
complete the form before leaving the office.1 Once the application is returned to the front desk,
the receptionist (or other office staff) reviews the form to determine whether the household is
eligible for expedited food stamp benefits.2 Staff may also discuss intake procedures and
identify verification materials that will be requested during the intake interview. In some
offices, the eligibility determination interview is scheduled before the applicant leaves the office.
'Applicants in our study sites were not told that they only needed to fill out their name and
address and sign the application in order to apply for benefits. The staff argued that they needed
to have complete forms in order to process the application accurately. However, receptionists
helped applicants complete the form if necessary.
households meeting certain income and resource criteria and homeless households are
eligible for expedited services. This means they are entitled to receive food stamp benefits
within five days. We use the terms expedited services and emergency services interchangeably.
In others, applicants are told that they will receive an appointment letter in the mail within the
week. Eligibility interviews are typically scheduled for 1-2 weeks from the date of application.
Interviews for expedited applicants are scheduled for no later than the next business day.
During the eligibility interview, caseworkers obtain detailed information on each
household member's income, resources, assets and expenses. Caseworkers also answer
questions about the intake process and benefit structure. Interviews for applicants that are only
applying for food stamps average 30-45 minutes, though interviews for expedited households
may take as little as 20 minutes. Interviews for households applying for multiple benefits
require 60-90 minutes.
The application process is completed when the applicant furnishes all supporting
documentation requested by the caseworker during the interview. The caseworker then
determines the household's eligibility and, if applicable, benefit amount.
The remainder of this chapter discusses each step in the food stamp application process
in more detail. It should be kept in mind that the discussion pertains to our five study sites only.
Other food stamp offices may structure the application process differently. Issues that are
important in our sites may not be important elsewhere. In addition, other sites may experience
problems not experienced by the five sites discussed here. Information comes from interviews
conducted with state and county office staff in December 1989 and January 1990, from project
staff observations at each of the five food stamp agencies in June 1990, and from the applicant
survey.
2.1 Obtaining Information and an Application
Applicant Activities
Most individuals who contact the welfare office are interested in applying for benefits and
know exactly what programs they want to apply for. According to the applicant survey, two-thirds
of the respondents knew exactly which benefits they wanted to apply for during their
initial contact with the food stamp agency. An additional 29 percent were prepared to apply
when they first contacted the office, but were unsure which programs were appropriate. The
10
remaining three percent were only interested in getting information during their initial office
contact.
Even some of those respondents who knew which benefits they wanted to apply for
learned about additional programs during their initial contact with the office (16 percent). Most
of these individuals reported learning about programs not administered by the welfare office,
such as Supplemental Security Income (SSI), Women, Infants, and Children Program (WIC),
and the state's job training program. However, some learned about programs administered by
the welfare office other than the program(s) for which they originally intended to apply.
Persons wishing to learn about food stamp benefits and obtain an application may call
or visit the food stamp office. According to agency staff in four of the study offices, applicants
typically initiate the process by visiting the office, although some call for information. Workers
in one of the rural offices encourage individuals to call for information and reportedly most do
so.
Applicants visiting the food stamp office must generally wait in line for less than five
minutes to talk to a receptionist. The wait can be longer in large urban offices and at the
beginning of the month when the number of new applicants is the greatest. Persons who have
previously received food stamps are generally familiar with the process and, therefore, have few
questions about application procedures. First-time applicants are likely to have questions about
eligibility criteria and benefit amounts, however. Most applicants who visit the office request
an application.
Applicants calling the food stamp office generally wait several minutes before talking
with staff. In some offices, callers receive a busy signal until a worker is available to talk with
them. In others, callers are either put on hold, transferred to another worker, or asked to leave
their name and number. However, it is rare for callers to get through immediately to the
appropriate worker. Clients frequently complain that they have difficulty reaching the office
because the telephone is constantly busy. In addition, persons calling from public telephones
generally cannot leave messages and must therefore continually attempt to contact the office.
11
Callers generally ask about office hours, directions to the office, and application
procedures. In addition, first-time applicants are likely to ask about eligibility criteria and
benefit amounts.
Office Activities
Applicants who visit the office are greeted at the front desk by a receptionist who
determines which benefits the person is interested in applying for. We observed that
receptionists answered questions patiently and completely, but rarely offered unsolicited
information. In our study sites, the majority of persons inquiring about food stamps
simultaneously receive an application. In fact, receptionists routinely respond to programmatic
questions by dispensing an application, with applicant inquiries being deferred until after the
form has been filed. Eighty percent of respondents to the applicant survey reported that a
worker encouraged them to file an application during their first contact with the food stamp
office. The remaining respondents did not report that they were encouraged to do so.
Even if the household expresses interest only in one of the programs administered by the
office, receptionists usually inquire about their interest in other benefits. Persons unfamiliar
with other programs are asked a series of questions about their household composition and the
receptionist provides guidance on which benefits to apply for and explains what parts of the
application must be filled out.
Applicants are instructed to complete the form and return it to the front desk. Applicants
in our sample sites are not informed that they only need to fill out their name and address and
sign the form in order to apply for benefits. Receptionists accept incomplete forms and work
with applicants to complete them. Receptionists do not feel that they can accurately process
applications, particularly determining whether a household is eligible for expedited benefits,
unless the forms are complete.
Telephone inquiries can be handled by a receptionist, clerk, caseworker, or supervisor.
Some offices have a phone line dedicated to food stamps. In other offices, the phone line also
handles AFDC and Medicaid inquiries. Receptionists are generally able to answer basic
12
questions about food stamp eligibility and program rules. More complicated questions are
referred to a case worker or supervisor. Telephone conversations generally last 3-5 minutes.
Staff in one study site offer to mail applications to all callers. However, in the other four
sites, applications are only mailed to callers that insist or complain. Although this policy may
deter some households from obtaining an application, staff are reluctant to mail out applications
for several reasons. First, persons filing via mail often furnish incomplete or inaccurate
information, thereby delaying or nullifying the application process. Second, clients are likely
to write "I don't know" for critical income or resource questions that are used to screen for
expedited services. Thus, applicants entitled to expedited processing may inadvertently delay
their receipt of food stamp assistance. Finally, staff indicate that mailing applications delays
benefits and reduces the household's allotment in the month of application.
Differences By Applicant Characteristics
The process of requesting information and obtaining an application is the same for all
applicants. However, each of the study sites attempts to facilitate access to the Food Stamp
Program for specific subgroups.
Handicapped and Elderly Applicants. Any individual may call the food stamp office and
request that an application be mailed, making it easier for elderly and disabled applicants to
obtain an application. However, four of the study sites will only mail applications on request.
(The fifth encourages aU callers to file via mail.) Thus, it may not be clear to all callers that
this option is available to them.
Homeless Applicants. In four of the study sites, applications are available in locations
other than the welfare and SSA offices, including homeless shelters, hospitals, local mental
health centers, service centers, and community advocate groups.
Expedited Services. Workers in the Southern Rural county encourage all callers to file
via mail. However, callers that appear eligible for expedited food stamp benefits are encouraged
to visit the office within a day or two to initiate an application without delay. Staff recommend
that these callers not wait to receive an application by mail. In an effort to identify expedited
13
applicants, all uUers in the other study sites are encouraged to visit the office to obtain an
application.
Employed Applicant. All offices in the study opened between 7:00 and 8:00 a.m. and
closed at 4:30 p.m. Thus, employed persons in each of the study sites can visit the food stamp
office early in the morning and during their lunch hour to request an application. They also
have the option of calling the office during business hours. None of the offices offer evening
hours which might present difficulties for some, particularly those working far away.
First-Time Applicants. Receptionists indicate that first-time applicants may not know
how to ask for the information they want. Thus, "welfare savvy" persons who are familiar with
the Program may be better able than first-time applicants to identify and obtain the information
they need to get through the process.
Differences By Office
Access. All three urban study sites are served by fairly extensive public transit systems.
The Midwestern Rural county has bus transportation within the city where the office is located.
The Southern Rural county is not served by any public transit, which, according to local
advocates, creates problems for some households.. This lack of public transportation and the fact
that the office is located several miles outside the county seat are the reasons the office
encourages interested individuals to call for information and an application.
Congestion in the office. Two of the Midwestern offices issue food stamp coupons in
the welfare office. Thus, the offices are quite congested during the beginning of the month.
This is particularly a problem in the Midwestern Large Urban office because of the extensive
caseload. Households coming in to apply for benefits at the beginning of the month are faced
with a crowded office, which might inhibit some from completing the application process.
2.2 Filing a Food Stamp Application
Persons may file an application via mail, or by visiting the office during normal working
hours. Offices also receive applications from the local Social Security Administration office for
14
SSI applicants. The majority of applicants obtain and file a food stamp application during a
single office visit. Even persons who call the office and receive an application by mail are likely
to bring the application to the office in order to file it.
Food stamp agencies are required to offer alternative filing methods for applicants who
cannot visit the office during regular business hours. Thus, all of the study sites permit clients
to mail applications, though as discussed above, staff generally discourage filing via mail.
Applicant Activities
Applicants use tables or counter space in the waiting area to complete the form. On
average, it takes 15 minutes to complete the application and return it to the front desk. The
application seeks basic identifying information such as name, address, phone number, social
security number, and birth dates of all household members. In addition, the form asks for
information on income, resources, and expenses. Applicants are asked to complete the income
and resource information if they want to be considered for emergency or expedited benefits.
Almost all applicants take the time to provide this information.
In all of the study offices, a single application form is used to apply for food stamps,
AFDC, and Medicaid, making it relatively easy to apply for multiple programs. According to
the applicant survey, 61 percent of the respondents in the study sites applied for food stamps
only. Another IS percent said they simultaneously applied for food stamps, AFDC and
Medicaid, while 12 percent mentioned food stamps and Medicaid, and 11 percent said food
stamps and AFDC.1 Less than one percent said they applied for AFDC or Medicaid but not
food stamps.
'Households applying only for food stamps are referred to as non-assistance (NPA)
households. Those applying for AFDC and/or Medicaid as well as food stamps are referred to
as public assistance (PA) households.
15
Staff Activities
Office staff are frequently called upon to help individuals complete the application form.
Assistance is rarely offered when the form is first provided to the applicant, however. More
commonly, staff either respond to specific questions or ask individuals about missing information
once the form has been filed. In particular, staff are frequently required to ask applicants for
more precise information about income, resources and expenses.
Applications returned to the front desk are reviewed by staff (e.g., receptionist, clerk,
caseworker) to ensure that all necessary information has been provided. For example, staff
make sure that all household members are listed on the application form and that individuals
have applied for the appropriate benefits. Once they have established the household's
composition, staff review income and resource information to identify households eligible for
expedited food stamp benefits. As mentioned in the previous section, applicants are only required
to provide their name, address and signature to file an application. Nonetheless, staff in the five
study sites do not apprise applicants of this option, but insist that they provide income and
resource information to identify those potentially eligible for expedited benefits. Persons
reporting no income are questioned about expenses (e.g., "How do you pay rent?" or, "How do
you put gas into your car?"). Staff may also probe to determine whether applicants have
provided net or gross income and whether the information reflects weekly or monthly totals.
They may also probe to ensure that all forms of unearned income, such as support payments and
assistance from friends are included on the form. Once an applicant's status regarding expedited
services has been determined, clerks schedule intake appointments for expedited applicants, may
schedule other interviews, and identify documents the applicant must supply during the interview
to verify the household's circumstances.
The interval between submitting an application and attending the intake interview is a
function of the volume of applications the office receives. For example, in June 1990,
applicants were given appointments one to two weeks in the future, except those eligible for
expedited services. Expedited food stamp cases were being seen on the following day. By mid-
July 1990, all regular applicants were being scheduled one to two days from the date of
16
application. However, a surge of applications in January 1991 pushed the interview date to as
long as three weeks after the date of application.
These sudden influxes of applications place considerable pressure on clerical staff and
caseworkers to process expedited and regular applicants within the required timeframes (five and
30 days, respectively). For example, one study site was temporarily required to schedule three
interviews for every appointment slot in January 1991 (under the assumption that a significant
number would fail to attend the interview). In February 1991, as the number of applications
diminished, the office scaled back to scheduling two interviews per appointment slot.
In three of the study sites, applicants receive an appointment for their intake interview
at the time they file an application. Applicants can request a later date if the appointment is not
satisfactory, though they cannot request an earlier date. In the Midwestern Large Urban county,
applicants are informed that they will receive an appointment notice in the mail within two
weeks.1 In the Southern Rural county, applicants who request an application via telephone
receive an appointment notice and an application form in the mail.
Differences By Applicant Characteristics
Handicapped and Elderly Applicants. Office staff report that they frequently assist
elderly applicants complete the form because these applicants often have difficulty deciphering
the small print on the application. Office staff are generally willing to spend extra time with
people if the office is not busy and the applicant is obviously in need of assistance or expresses
confusion. However, receptionists do not appear to look for people that might need their help.
A local hunger advocacy group in one study site provides assistance to elderly food stamp
applicants living in selected apartment buildings. The advocate believes that without this
assistance, 90 percent would not complete or file the form.
'During our observations in June 1990, office staff checked state and county computer files
to determine if household members already received public assistance, food stamps and/or
Medicaid before scheduling the interview. At this writing, the computer checks are being done
after scheduling the interview, and staff report that applicants now receive notices within a few
days.
17
Illiterate Applicants. Staff will help individuals who have difficulty reading if the
applicant requests help. However, as with the elderly and disabled, staff do not routinely offer
to assist applicants.
Non-English Speaking Applicants. The number of non-English speaking persons in our
study sites is small. None of the offices is required to provide applications or informational
materials in any language other than English. Staff report that few people need interpretation,
and those that do bring relatives or friends to act as translators.
Homeless. The two large urban offices station caseworkers in hospitals and homeless
shelters to assist residents with the application form.
Emergency Assistance. Agency staff make an effort to refer applicants to other local
social services. For example, people who appear distraught by the length of time they will have
to wait for food stamp benefits are referred to local food pantries. In some sites, clients stating
that they require medical attention are referred to local health clinics. In the Midwestern state,
applicants ineligible for AFDC may be sent to the township for county assistance with rent or
utilities. Non-expedited applicants may also receive immediate food assistance from the
township.
SSI Applicants. In all the study sites, the local Social Security (SSA) office provides
food stamp applications for at least some SSI applicants. All of the study sites report that they
get fewer than five applications per week from SSA. In the Southern Rural county, staff report
only one food stamp application per year from SSA. They suspect that the SSA office simply
tells those interested in food stamps to visit the welfare office.
Public Assistance (PA) Applicants. Those applying for AFDC and Medicaid as well as
food stamps are instructed about additional verification requirements they will have to satisfy
during the intake interview at the time they file the application.
In three of the study sites, PA and NPA applications are processed by the same office
staff. In the Midwestern Rural county, NPA and PA applications are reviewed by clerks in
different units, who may transfer an application to the other clerk.
18
The Midwestern Mid-sized county has separate front desks for NPA and PA applicants.
A receptionist directs those only interested in food stamps to the NPA desk and those inquiring
about food stamps and AFDC to the PA desk. If individuals make a general inquiry about
assistance, the receptionist asks basic questions (e.g., "Are you single?" "Do you have
children?") and directs them to the appropriate program. Applications are reviewed and
processed by separate units. Once they have reviewed an applicant's household information,
clerks may refer the individual to the other unit. For example, single parent households
erroneously referred to the NPA unit are told to take their application to the PA unit.
Supervisors felt that this office structure caused confusion for some clients. In addition, some
clients were not informed about all the benefits for which they were eligible and some were
required to submit multiple applications. In an effort to facilitate application processing, the
Food Stamp Director recently assumed responsibility for all NPA and PA certification activities.
The office may eventually integrate front desk duties for NPA and PA clerical staff, thereby
simplifying the routing of applicants and maximizing the effectiveness of increasingly scarce staff
resources.
Differences By Office
Alternative Filing Methods. In the Southern Rural county, persons who call the office
are mailed an application and informed that they can either return the form in the mail or file
at the time they appear for the certification interview (which delays the effective date of their
food stamp eligibility). However, callers potentially eligible for expedited benefits are
encouraged to visit the office within a day or two to file an application in person.
In the Midwestern Mid-sized county, persons who are unable to visit during regular
business hours may deposit their completed application in a drop box outside the office. In
addition, three sites allow applicants to file their application at other social service providers
(e.g. homeless shelters, hospitals, service centers).
Pre-Screening. Applicants in the Southern state are "pre-screened" at the time they file
their food stamp application. The purpose of this exercise is to determine whether the
application should be expedited and whether the household is clearly ineligible for food stamps.
19
The pre-screening interview is conducted by caseworkers or supervisors and lasts between five
and ten minutes. Workers begin by saying: "I'm going to pre-screen you to see if you are
eligible." This is accomplished by determining whether applicants satisfy the Program's gross
income and resource requirements.
In the Southern Urban site, the pre-screening interview occurs after the application has
been filed at the front desk. The wait to see a screener is between 30 minutes and 3 1/2 hours,
depending on the time of day. This long delay causes some applicants to file their application
without waiting for the pre-screening interview. The receptionist keeps these applications in a
pre-screening box to see if the applicant returns. After two days, she sets up a certification
interview and sends out an appointment notice. This letter also informs applicants about the
documents they should bring to the interview.
Applicants in the Southern Rural county generally request an application via telephone
and formally file for benefits on the day of their intake interview. Thus, most applicants are
pre-screened over the telephone at the time they request their application. Obviously ineligible
applicants often choose not to get an application. Persons visiting the office to apply are pre-screened
after they file their application and rarely wait more than 30 minutes to meet with a
screener.
Receptionists in the Midwestern Mid-sized county conduct an informal "pre-screening"
when an individual files a signed application. If an applicant's gross income exceeds Program
allowances by hundreds of dollars, the receptionist sometimes informs the applicant that he or
she is ineligible. Staff estimate that this affects one or two applicants per week. Some of these
applicants (one or two per month) choose to voluntarily withdraw their applications, and they
are denied food stamps on this basis. Most, however, continue the application process. This
action to deny occurs before the interview is scheduled and is taken on the basis of clients'
statements regarding their circumstances. Staff indicate; that such a denial saves time for both
the applicant and the agency in cases where the applicant is clearly ineligible.
20
2.3 Certification Interviews
Before conducting the certification interview, caseworkers review the application and any
information generated by clerical staff from computer searches. At the designated time, the
caseworker enters the waiting room and calls out the name of the applicant. (In some sites,
applicants sign in at the front desk, so caseworkers know whether or not applicants are on time.)
Caseworkers are free to use their discretion in structuring eligibility interviews, so long
as the information obtained from applicants is complete and properly verified. Caseworkers use
an interview guide to ensure that they obtain all necessary information on the following topics:
• Are all household members eligible for food stamps? (Includes
consideration of age, social security numbers, citizenship, residence,
whether any members are students, and definition of the food stamp
household.)
• What are the household's resources? (e.g., vehicles, bank accounts, cash
on hand, retirement accounts, lump sum payments, etc.)
• What is the household's earned and unearned income?
• What are the household's shelter and utility costs?
• Do any household members qualify for medical deductions?
• Does the household have any dependent care expenses?
• Are any household members mandatory work registrants?
During the interview, applicants are instructed to read.a list of their rights and
responsibilities including their right to a fair hearing, the need to report changes in a timely
manner, the penalty for providing false information during the certification process, and proper
and improper uses of food stamp coupons. Once applicants have read these instructions, they
are told to sign a statement that they understand the obligations. Many applicants have been on
the Program before, and understand these rules. Nonetheless, caseworkers devote several
minutes to this activity, especially if they perceive that an applicant is not fully cognizant of
these policies.
21
Caseworkers conclude the interview by providing and reviewing with the applicant a form
that lists any verification items that the client must supply in order for the application to be
processed. The caseworker tells the client to mail or bring in the documents to the office.
Applicants are generally told that the sooner they provide the verification, the sooner the
paperwork will be processed.
Differences By Applicant Characteristics
PA Applicants. In addition to the information required for the Food Stamp Program, PA
applicants must provide detailed information on the absent parent(s) including race, age, height,
nicknames, hair and eye color, and criminal record. Caseworkers also ask or orally review
whether paternity has been established. Regarding Medicaid, applicants receive information
about medical services available for their children and the types of expenses covered by the
program.
Because PA interviews obtain information for multiple programs, agencies allot more
time than for NPA interviews. Staff generally allow 45 minutes for food-stamp-only interviews
and one and one-half to two hours for PA interviews. On average, approximately one-third of
the joint Interview is devoted to food stamps. PA staff indicate that it is difficult to provide
information about all programs and satisfy the data gathering requirements. As a result, PA
caseworkers are frequently unable to complete the entire interview in the allotted time.
Handicapped and Elderly Applicants. These applicants are not required to come into the
office for a certification interview. They may have an authorized representative attend the
interview for them, which some sites encourage. The office will arrange for a home visit or
telephone interview if necessary. In our observations staff did not offer the option of home
visits or telephone interviews, but applicants who requested them received them.
Some sites attempt to assist those elderly and handicapped individuals who come into the
office to file by scheduling a certification interview at the time they file.
Expedited Services. Applicants eligible for expedited food stamps must receive their
benefits within five calendar days of the date the application is filed. To meet this deadline
22
offices interview expedited applicants on the day or the day after they file their application. If
a client that should have been expedited is not given an expedited appointment (i.e., the
caseworker determines during the certification interview that the client should have been
expedited) the caseworker has five days from the date of the interview to process the application
and provide the client with food stamps. On the other hand, staff report that many applicants
who are given expedited interviews rum out to be ineligible for expedited benefits (though most
still qualify for food stamps).
Interviews for expedited applicants generally take less time than other interviews because
by definition, these applicants have limited resources and income.
In the Southern Urban county, staff report that approximately 40 percent of food stamp
applicants are eligible for expedited benefits. In some months, the office conducts group
interviews for expedited clients to meet the five day processing deadline. These interviews,
which are conducted by three caseworkers, last 90 minutes and accommodate as many as 14
clients. During especially busy periods, up to three group interviews can occur in a single day.
During the group interview workers explain the application process following a scripted
presentation on a flip chart. Workers help applicants sign and hand in all the necessary forms
including the application workbook which covers all the details of the household's circumstances
necessary to determine eligibility. After the group interview, caseworkers hold five-to-ten •
minute private interviews with all applicants to complete the personal sections of the application
workbook including the income and resource questions.
Non-English Speaking. Because it is an infrequent occurrence, none of the study sites
has a formal policy for conducting certification interviews with non-English speaking applicants.
Staff report that most such applicants bring their own translators. When this is not possible,
staff deal with individual circumstances as they arise. For example, one office hires interpreters
from a local university if staff know in advance that they will not be able to communicate with
an applicant.
Employed. Certification interviews impose a burden on persons with jobs who are unable
to visit the welfare office in the early morning or during their lunch hour. Approximately 8
23
percent of the survey respondents reported losing an average of $24 in wages each time they
were required to visit the welfare office. Eligibility interviews can also take longer if there are
several sources of earned income to document, for example, for an applicant who has worked
at several fast food franchises in the previous 30 days.
Previously Received Food Stamps. Applicants who have previously applied for food
stamps are generally well prepared for the certification interview. They tend not to be surprised
or disturbed by the substance of the interview. In addition, they generally know what types of
verification items to bring to the interview. Applicants who have previously received food
stamps tend to ask practical questions such as "When do you think I will receive my food
stamps? I need to know how much of my next paycheck to spend on food."
No-Shows. Federal regulations require that the food stamp office schedule a second
interview for those applicants who fail to show up for the fir* interview, no matter what the
reason. In the Southern state, local offices will do this. The Midwestern state has received a
waiver from FNS and thus offices are not required to reschedule missed eligibility interviews.
Persons failing to attend their initial interview have their applications held for 30 days from the
date it was filed. During this period, individuals may call to reschedule the appointments. At
the end of 30 days, the application is denied. Applicants are sent a notice indicating that their
case was denied for failure to attend the interview.
2.4 Providing Necessary Verification
At the time they file an application, persons are informed of generic (and in some
instances, specific) items they must bring to the certification interview. During the interview,
caseworkers review these verification items and instruct applicants about any additional
documents that are needed to determine eligibility and benefit amounts. After the interview,
applicants assume primary responsibility for obtaining these items, although for some items
caseworkers will use methods such as collateral contacts and computer searches to corroborate
a household's circumstances.
24
Applicant Activities
The items that applicants must verify in order to complete their food stamp application
are discussed below.
Identity of the Applicant. Applicants must furnish a current proof of their identity.
Examples of acceptable items include a driver's license, work or school identification card, voter
registration, birth certificate, military records, passport, or U.S. immigration papers. If
authorized representatives are interviewed, the identity of both the applicant and the authorized
representative must be provided.
Citizenship. The citizenship of all household members must be verified. Applicants with
alien status must have papers from the U.S. Immigration and Naturalization Service.
County Residency. Applicants must provide proof that they reside within the welfare
department's jurisdiction. Examples of acceptable items include rent receipts, mortgage payment
books, utility bills (or installation orders), landlord statements, library cards and driver's
licenses. Caseworkers in one site report that some applicants will mail themselves a letter as
proof of their home address.
Social Security Number. During the intake interview, applicants must provide a Social
Security number for all household members. Examples of acceptable items include a Social
Security card, Railroad Retirement number, or Veteran's Claim number. In some sites,
caseworkers can locate an individual's Social Security number on an automated data base.
In the Southern state, caseworkers will initiate and forward to the SSA office applications
for SSNs for those household members that do not have numbers. Applicants in the Midwestern
state who do not have an SSN are given a referral form by the welfare office. The applicant
must take the referral form to the SSA office and apply for an SSN. The SSA office sends a
copy of the referral form back to the welfare office as proof that the individual has applied for
an SSN.
Resources. Applicants must document the financial resources of all household members.
This includes cash on hand, savings or credit union accounts, checking accounts, stocks, bonds,
25
etc. In some sites, they must furnish registration or title documents that verify ownership of
vehicles. In the Midwestern state, applicants also sign the Authorization for Release of Financial
Information Form. This document authorizes the release of bank records to verify personal
financial data.
Unearned Income. Applicants are required to provide records of all unearned income,
e.g., Social Security, Supplemental Security Income (SSI), Railroad Retirement, Veteran's
benefits, military allotment, Unemployment Insurance (UI), child support and alimony.
Acceptable forms of proof vary. For example, applicants receiving direct support or alimony
must furnish copies of support checks, or a statement from the payor (persons receiving
payments through the court must provide the name of payor and a court clerk's statement).
Applicants receiving income from roomers must supply a signed statement from the renter
stating the amount paid to the household for the entire month.
Earned Income- Applicants must provide pay stubs for any earned income received in
the previous 30 days. (Applicants with fluctuating income may be required to provide
documentation beyond the previous 30 days.) If no pay stubs are available, applicants must
furnish a signed employer's statement of earnings for the entire month (or statements from all
individuals for whom work was done). Self-employed applicants must furnish current business
records and receipts or the previous year's tax return if income is not received on a monthly
basis.
Applicants who have recently lost their job must describe the circumstances surrounding
the termination, specifically whether they quit or were fired from the job. The applicant's
statement is generally sufficient. However, in one of the study sites, the caseworker always
calls the former employer to verify the circumstances.
In the Midwestern state, household members over the age of 18 are required to sign the
state's Request for Information Regarding Earnings Form. This form grants the office authority
to contact employers to verify information on earned income, and to match applicants' Social
Security numbers against state employment and Unemployment Insurance records.
26
Contributions. Applicants receiving contributions must supply a note from the person
providing the assistance documenting the monetary amount and date assistance was provided.
Expenses. Applicants claiming deductions must furnish documentation of the frequency,
billing dates and amounts of the allowable expenses (e.g., rent/mortgage, taxes, insurance,
assessments, utilities, telephone bills, home repairs, dependent care, and medical bills).
Household Composition. Applicants may have to document that an individual should (or
'should not) be included in the household for the purpose of food stamp eligibility and benefit
calculations. In most cases, workers accept the applicant's word about household composition.
After the intake interview, applicants have until the end of the processing period (i.e.,
30 days from the date of application) to furnish the required items. In the Midwestern state,
applicants failing to provide adequate documentation receive a notice that their application has
been held for an additional 30 days, pending receipt of these items (these applicants lose benefits
for the first month). At the end of 60 days, applications are automatically denied if the applicant
fails to furnish necessary documentation. All applicants in the Southern state failing to provide
adequate documentation are denied after 30 days. If they provide verification within the
succeeding 30 days, their application is re-opened.
Staff Activities
Unit clerks and caseworkers assist applicants to identify acceptable verification items by
providing a generic list of items applicants need to bring to the certification interview at the time
they file their application. Staff may also prepare applicants for the interview by probing for
unique verification requirements that they will be required to fulfill.
Prior to the certification interview, clerks run computer checks on state and county data
bases to determine whether household members already receive food stamps or AFDC.
Computer matching can also be performed to determine benefit history for Medicaid, social
security and SSI, employment history and earnings, and unemployment insurance benefits.
During the interview, caseworkers review any documentation furnished by the applicant.
Items that satisfy the verification requirement are copied and returned during the interview.
27
Applicants failing to provide complete documentation at the interview are provided a checklist
of outstanding items. Caseworkers instruct applicants to mail outstanding items to the office or
to bring documentation to the front desk.
The burden of obtaining documentation rests with applicants. Caseworkers will,
however, assume responsibility for certain items such as:
• Contacting government agencies to verify receipt of Social Security, SSI
or Unemployment Insurance;
• Initiating and forwarding an application *.o the Social Security
Administration for household members that do not have a Social Security
number (only in the Southern state); and
• Contacting employers to verify the remount of earned income claimed by
applicants. This generally happens after applicants have demonstrated a
willingness to exhaust all means of verification available to them.
None of the offices in our study reported using other welfare workers (e.g., social service or
protective service workers) to help obtain documentation.
According to federal regulations as well as local office policy, workers must assist
applicants who have difficulty obtaining required documents. Approximately 17 percent of
applicant survey respondents completing the certification interview indicated that their
caseworkers offered to help them verify their household circumstances without being asked. An
additional ten percent of respondents reported that they requested assistance from then-caseworker.
After the intake interview, caseworkers await outstanding verification items before
processing the application. Once all documentation has been received, caseworkers complete
the necessary paperwork to either approve or deny the application. Workers have 30 days from
the date the food stamp application was filed to make an eligibility determination.
Differences By Applicant Characteristics
AFDC/FS Applicants. Persons applying for food stamp and AFDC benefits must satisfy
the documentation requirements of both programs. For example, AFDC rules require that
28
applicants document the ages of all household members by providing a birth certificate,
baptismal record, driver's license, insurance policy, or records from a clinic, doctor or hospital.
(In the Southern sites caseworkers can check computer files for birth certificates for those born
in the state.) In addition, AFDC applicants are required to certify that their children actually
live in the household. Examples of acceptable proof include a child care provider's records,
landlord statement, statement from the school, and a signed statement from a non-related person
having knowledge of child's location. AFDC applicants must also provide detailed information
about absent parents (for enforcement of child support conditions). Failure to fulfill AFDC
verification requirements does not affect food stamp eligibility, and vice versa. However,
workers do not clarify which documents are required for which programs.
Workers must meet the 30 day processing requirement for the food stamp application.
Eligibility determination for the AFDC application must be complete in 45 days.
Homeless and Expedited Applicants. Expedited applicants may receive their initial food
stamp allotment before verifying their circumstances. However, documentation must be received
within the 30-day deadline if they are to continue receiving benefits. In addition, homeless
applicants are not required to verify their residence.
Non-English Speaking and Elderly Applicants. In some specific cases, community or
advocate groups help these applicants complete the application process. This may involve help
obtaining necessary documentation.
Categorically Eligible. Food stamp applicants who already receive (or are authorized to
receive) AFDC and/or SSI are only required to provide items that exceed the verification
requirements of these two programs.
2.5 Eligibility Determination and Benefit Calculation
In order to be approved for food stamp benefits, an individual must complete all the steps
described above. After all the verification is submitted, the caseworker determines whether the
household is eligible for benefits and if so, the amount of the monthly allotment. The worker
then sends the applicant the appropriate notice.
29
Those individuals who file an application but do not complete all the required steps of
the application process are automatically denied benefits. They receive a notice informing them
which requirements they failed to fulfill.
The food stamp office does not consider a household a food stamp applicant until it files
an application. Therefore, those who obtain an application but do not return it to the office
receive no communication from the food stamp office.
The next chapter provides additional information of what the food stamp application
process requires of applicants and potential applicants. All individuals who contact the food
stamp office for information about benefits spend at least some time and probably incur some
expenses, and the chapter presents estimates of these costs.
30
Chapter Three
COSTS OF THE APPLICATION PROCESS
The activities that must be completed in order to apply for food stamp benefits all impose
some cost and burdens on individuals. No matter how much of the process a household
completes, it must spend at least some time on required activities, even if this just involves
..making a telephone call. Most people will also incur expenses during the application process
in the course of visits to the food stamp office. In this chapter, we attempt to measure these
costs and burdens in order to provide additional information on what the process requires of
applicants. We also use these estimates in Chapter 4 when we examine the reasons some
individuals do not complete the application process.
In this chapter we calculate the costs of the application process based on data from the
client survey. In the first section, estimates are derived for the total monetary costs of the
application process, based on the numbers of in-person, telephone, and mail contacts that clients
had with the welfare office and the cost of each one. We also include the cost to clients of
obtaining necessary documentation. In the second section time costs are calculated, including
time spent actively, for meeting and talking with caseworkers, filling out the application form,
and collecting documentation. Time spent on travel to the office and waiting to meet with staff
is also included. Throughout the chapter we use qualitative data obtained from case studies of
the sites to interpret and explain the observed findings.
3.1 Monetary Costs
The mean out-of-pocket cost of the application process for all individuals who enter the
food stamp office, regardless of how far they went in the application process, is $10.40 for all
31
5 counties combined (Exhibit 3.1).1 The mean ranges from $4 in the Southern Rural county
to $12 in the Midwestern Large Urban and Rural counties. Although the mean amount is
modest, the distribution is quite skewed. The 95th percentile value for the sample as a whole
was $46.50-four and a half times the mean. The median values were therefore substantially
lower than the means, $3.61 for the sample as a whole, and ranging from only about $1 in the
Southern Rural county to $4 in the Midwestern Large Urban and Rural counties. Thus, most
applicants and potential applicants spend $4 or less applying for benefits, though some spend
substantially more.
Standard errors for the estimates of the means are also displayed. These can be used to
calculate the approximate statistical significance of differences between pairs of estimates.2
'The sampling design for the applicant survey was a stratified design, stratified according
to how far the individual got in the application process. Thus, the number of (unweighted) cases
in each stratum is roughly equal. The data in this and all subsequent exhibits were weighted to
obtain representativeness with respect to the population entering the food stamp offices in the
five counties. This was done as follows. The average number of people that enter each office
per day was estimated based on the counts of people that were logged in each office during the
observation period, and then allocated among the five strata according to automated data records
on the logged individuals. The resulting population number, divided by the number of such
cases that were included in the.survey sample, was then used as the weight for sample members
in the corresponding county and stratum.
2 The formula for calculating statistical significance between pairs of estimates is:
t=(mrni2) / Vs? + Sj2,
where m, and m2 are the estimated means, and
s, and s2 are the corresponding estimates of the standard error.
When a hypothesis about the direction of the difference is being tested, significance levels based
on one-tailed test may be used~e.g., a value of I greater than 1.65 is considered statistically
significant at the 5 percent level. When there is no prior hypothesis about the direction of the
effect, a two tailed test is used~e.g., t must exceed 1.96 in absolute value to be considered
statistically significant at the 5 percent level. In any event, these simple tests are only
approximate, given the skewed nature of the data.
Tests show that the difference between the mean for the Southern Rural county and the
mean for each of the other sites is statistically significant at the 1 percent level (two-tailed tests).
The other differences are not statistically significant.
32
Exhibit 3.1
TOTAL OUT-OF-POCKET COSTS
Southern State
Large
Urban
County
Rural
County
Midwestern State
Large
Urban
County
Mid-sized
Urban
County
Rural
County TOTAL"
All Respondents:
Mean $9.54 $3.70 $11.57 $10.33 $10.71 $10.40
(Standard error) (1.60) (0.54) (1.44) (1.66) (2.54) (0.77)
Median $3.13 $1.20 $4.00 $3.84 $4.00 $3.61
95th percentile $42.00 $14.80 $52.88 $40.00 $41.48 $46.50
Respondents who:
Did not file
Mean $12.66 $1.49 $10.05 $1.57 $2.37 $8.91
(Standard error) (10.02) (0.47) (4.92) (0.86) (0.97) (2.51)
Median $3.36 $0.00 $0.00 $0.00 $0.50 $0.00
95th percentile $25.20 $4.80 $64.00 $11.68 $11.88 $52.88
Filed, but were
not interviewed
Mean $7.98 $2.41 $7.14 $5.98 $11.84 $7.36
(Standard error) (2.98) (0.68) (2.44) (1.78) (3.54) (1.33)
Median $0.50 $2.00 $3.00 $1.20 $4.96 $2.40
95th percentile $85.00 $8.20 $15.00 $31.97 $32.68 $31.97
Were interviewed,
but were not
approved
Mean $11.61 $6.92 $17.57 $23.48 $5.13 $16.28
(Standard error) (2.14) (1.43) (2.79) (5.69) (0.80) (1.61)
Median $4.09 $3.56 $10.00 $4.40 $3.60 $5.92
95th percentile $47.80 $16.00 $46.50 $139.00 $17.24 $49.68
Were approved
Mean $9.26 $3.67 $11.29 $8.77 $13.63 $9.80
(Standard error) (2.66) (0.92) (2.41) (1.49) (5.66) (1.26)
Median $3.84 $0.25 $4.00 $5.04 $4.80 $4.00
95th percentile $42.00 $12.80 $35.00 $40.00 $41.48 $42.00
Unweighted N 187 93 159 157 110 706
* The total statistics in this and subsequent tables are weighted by strata and/or county
as described on the previous page. As a result, the approved strata and the urban counties
have the most impact on the overall statistics. See Exhibit 4.1 for the overall
distributions by county and strata.
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
33
The remainder of the exhibit shows the costs that were incurred by respondents in four
strata:
• those who did not file an application;
• those who filed an application, but did not have a certification interview;
• those who had a certification interview, but were not approved to receive
food stamps;1 and
• those who were approved to receive food stamps.
For all counties combined, average costs were $7 to $10 for the first, second, and fourth strata,
and over $16 for those who were interviewed but were not approved. This latter value differed
significantly from each of the other three at the 5 percent level. The pattern is not consistent
across the individual counties, however. Instead, we see that respondents in particular strata
incurred relatively high or low costs depending on their location. Most of the pairwise
comparisons between sites within strata are not statistically significant. Exceptions are noted
where they occur.
Nonfilers in the Large Urban counties incurred average costs of $10 to $13, compared
with only $1 to $2 in the other three sites.2 This difference is related in part to whether
nonfilers visit the office in person. In the Southern Rural county, individuals generally made
their initial contact with the food stamp office by telephone. For those who filed, but were not
'The survey respondents are divided into strata in various ways for different parts of the
analysis. The third stratum here, which consists of individuals who completed their certification
interview but were not approved, contains people who performed varying amounts of documenta-tion
activity. While it would have been possible to break out those who reported completing
their documentation from those who did not, this distinction would not be useful for
understanding differences in costs. The group who report completing their documentation
includes applicants who were circumstantially ineligible but who went through the same
procedures as those who were eventually approved. It also includes individuals who were not
asked to provide additional verification because they were determined to be ineligible at the
interview. These latter individuals are therefore not necessarily any different, with respect to
documentation costs, than those who fail to complete their documentation.
2The comparison between the Large Urban Midwestern county and both the Mid-sized Urban
Midwestern county and the Southern Rural county are significant at the 10 percent level. All
other differences are not statistically significant.
34
interviewed, costs range from $6 to $12 for all sites except the Southern Rural county. The
much lower value of $2 for this site reflects the fact that many people obtain the application
form by mail and file without visiting the office. The observed differences between this county
and all other counties is significant at the 10 percent level, or better.
For those who were interviewed, but were not approved, the average costs ranged from
$12 to $23 in the Urban sites, compared with only $5 to $7 in the two Rural sites. Each of the
Urban sites differs from each of the Rural sites at the 10 percent level, or better.
Finally, for those persons who were approved for food stamp benefits, mean costs range
from only $4 in the Southern Rural county to $14 in the Midwestern Rural county. (The
Southern Rural county differs significantly from the others.) The overall means largely reflect
the results for this stratum, as it comprises over half of the population.
In the remainder of this section, we look at the individual components of out-of-pocket
cost. The relationships among these are depicted in Exhibit 3.2. Total costs are broken down
into the costs associated with contacts with the food stamp office, such as transportation and
postage, and those incurred outside the food stamp office, for obtaining documents.
Contacts with the food stamp office are of three types: in-person visits, telephone calls,
and letters.1 The costs of contacts may therefore be calculated as follows:
OUT-OF-POCKET COSTS OF CONTACTS =
<Nv * Cv) + (NT * Cr) + <Nt * CJ
where
Nv = number of office visits;
Cv = cost per office visit;
NT = number of telephone calls;
CT = cost per telephone call;
'Mail contacts include only items sent by the respondent to the food stamp office-not items
received from the food stamp office.
35
0\
COMPONENTS OF 0
EXHIBIT 3.2
UT-OF-POCKET COSTS OF APPIJCATION
.
/
TRANSPOR-TATION
NUMBER
OF VISITS /
VISITS PERCENT
PAYING
_ -
^ COST PER
VISIT
CHILD
CARE
AVERAGE
/ COST
——
/ NUMBER
OF CALLS
\
V
^—-•— \ PERCENT
INCURRING
CONTACTS
PHONE
CALLS FORGONE
COST PER WAGES
CALL AVERAGE
COST
/
\
\
NUMBER
OF LETTERS z::\ TOTAL OUT-OF- LETTERS
POCKET COSTS COST PER
LETTER
PERCENT
INCURRING
A
TRANSPOR-TATION
DOCUMENTATION /
/
AVERAGE
COST DUPLICATING
.
CHARGES FOR
DOCUMENTS
5(*
NL = number of letters sent; and
CL = cost per letter (assumed to be $0.25).
The cost per office visit may be further disaggregated into costs associated with transportation,
child care, and forgone wages. The latter two costs are a function of the proportion of people
who incur the cost and the mean expense incurred by those who do. The costs incurred for
providing documents are also of three types: travel to obtain the documents, duplication, and
charges for the documents. These total documentation costs are also a function of the proportion
of people incurring them.
The total out-of-pocket cost of about $10 is comprised of around $7 for office visits and
$3 for documentation, as Exhibit 3.3 shows. Phone calls and letters entail negligible expense:.
Examining the costs of visits, we find that nearly half of the costs of each office visit is
due to forgone wages. However, both forgone wages and child care expenses are heavily
concentrated among a small proportion of the population. This means that most people making
an office visit incur only transportation costs, though a few people incur very heavy costs for
the other items.
Only 40 percent of respondents incurred any documentation costs. Some potential
applicants drop out of the application process before it is necessary to provide documents, while
others succeed in obtaining the necessary documents at no cost. For those who do incur costs,
transportation to obtain the documents is clearly the bulk of the expense.
Exhibit 3.4 shows the same numbers for orJy those individuals who were approved to
receive benefits. Comparing these values with their counterparts in Exhibit 3.3, we note several
interesting differences. Most strikingly, those who were approved had lower costs per visit
($3.33 on average, versus $4.56) than all individuals. The difference in cost per visit is attrib-utable
mainly to a lower probability of losing pay. This is due in part to the fact that approved
individuals are less likely to have a job. More importantly, however, approved individuals who
had a job were less likely to lose pay during a visit to the welfare office than those who were
not approved and had a job (20 percent versus 78 percent).
37
CO
EXHIBIT 3.:
COMPONENTS OF OUT-OF-POCKET COSTS OF
>
APPLICAT1ON: ALL RESPONDENTS
TRANSPORTATION
$1.75
^ NUMBER OP VISITS
1.6
/ mM / VISITS
$7.30
PERCENT PAYING
/ 7.6%
COST PER VISIT
$456
CHILD CARE
$0.71
/ AVERAGE COST
\ "~-~ $928
NUMBER OF CALLS
05 \
PERCENT INCURRING
CONTACTS 9.1%
$734
PHONE CALLS
$0.01 \ FORGONE WACES
COST PER CALL $Z09
$0.02
AVERAGE COST
/ \ $23.06 // \\
/ \
NUMBER OF LETTERS
0.1
TOTAL OUT-OF-POCKET
COSTS
$10.40
\ LETTERS
$0.03 ■:■
COST PER LETTER
$025
>-Vr
-■"A- PERCENT INCURRING
405% TRANSPORTATION
DOCUMENTATION $4.45
$3.06
/
/
AVERAGE COST
$755
DUPLICATING
$0.69
: ;..■■ ':,;■■ 7: .•-■ ' ..•_': V
\
CHARGES FOR
DOCUMENTS
$2.41
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
3?
NO
EXHIBIT 3.-
COMPONENTS OF OUT-OF-POCKET COSTS
1
OF APPLIC:ATIO N: APPROVALS
/
TRANSPORTATION
$1.83
NUMBER OF VISITS
1.8 /
/
VISITS
$5.99
1 '
___ PERCENT PAYINC
5.4%
COST PER VISIT
$3.33
CHILD CARE
$0.48
/ AVERAGE COST
$8.90
NUMBER OF CALLS
05
\
\
\ PERCENT INCURRING
CONTACTS 3.4%
$6.04
PHONE CALLS
$0.01
_.,-
FORGONE WAGES
COST PER CALL $1.00
$0.03
AVERAGE COST
/ \ $29.68
// \\
/ \
NUMBER OF LETTERS
0.1
TOTAL OUT-OF-POCKET
COSTS
$9.80
\ LETTERS
$0.03
COST PER LETTER
"-"-"- $025
\
\ PERCENT INCURRINC
51.8% TRANSPORTATION
DOCUMENTATION $452
$376
/
/
AVERAGE COST
$726
DUPLICATING
$0.71
\
\
CHARGES FOR
DOCUMENTS
$2.03
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
3s-
Several factors might explain this pattern. One possibility is that the employed individu-als
who did not lose wages were working fewer hours, presumably with lower earnings and
hence with a higher probability of being approved. It could also be that those who did not lose
pay had relatively flexible working conditions. Finally, food stamp staff in the study sites report
that they will interview employed persons at a convenient time so that they do not lose wages,
if applicants request. Applicants who were not approved may have been less likely than others
to be aware of this option. In fact, losing wages may have deterred them from completing the
application process. The data do not allow us to distinguish among these explanations.1
Another difference between Exhibit 3.3 and Exhibit 3.4 is that approved individuals, not
surprisingly, made somewhat more visits to the welfare office on average than the general
respondent population. A greater percentage of those approved also incurred documentation
costs. However, among those who had documentation costs, the mean cost was slightly less for
approved applicants than for all respondents who incurred such costs.
Comparison of these two exhibits suggests that if those who did not complete the
application process had done so, it would have cost them more out-of-pocket than it cost those
who were approved, because of the higher probability of forgone wages and higher
documentation costs. An estimate of what it would have cost them depends on certain
assumptions about the cost components. At the conclusion of this section, we present estimates
of these hypothetical costs.
In the subsections that follow, the individual components of out-of-pocket costs are
analyzed, and we show how they fit together to comprise total monetary costs for individuals
in the various counties and strata.
'According to the applicant survey, approximately 2 percent of those who did not receive
food stamp benefits reported that they could not get time off from work (Exhibit 4.2).
However, the survey does not provide any other data to allow us to determine why the approved
were less likely to lose pay than those who were not approved.
40
Number of Contacts
The mean number of contacts (including visits, calls, and letters) across all five sites was
2.2, as Exhibit 3.5 shows. The figure varies very little, ranging only from 2.1 to 2.3 among
the five sites. For all five counties combined, about a quarter of sample members had only one
contact, and a third of the sample had three or more contacts.
Individuals who got further through the process naturally had more contacts, as shown
m the lower part of the exhibit. Thus, the mean number of contacts was 1.4 for those who did
not file, contrasted with 2.4 for those who were interviewed (a statistically significant difference
at the 1 percent level). Overall, the mean number of contacts is the same for those who were
interviewed but denied as for those who were approved. However, the measured difference can
go in either direction in the individual sites.
In contrast with the findings for contacts of all types, we see substantial variations among
the sites in mean number of in-person contacts, which are likely to be the most burdensome part
of the process (Exhibit 3.6). The overall mean number of visits for the five counties ranged
from 1.3 in the Southern Rural county to 1.9 in the Midwestern Rural county.
For all sites combined, 84 percent of the sample reported one or two in-person visits, and
another 10 percent reported three or more visits. Two sites exhibited substantially different
patterns, reporting that 9 to 11 percent of individuals made no in-person visits. Office
procedures probably explain these differences. The Southern Rural county encourages those
interested in food stamps to call for information and an application. During the telephone
conversation, potential applicants are screened to ascertain whether they are within the
Program's gross income and resource limits. Most individuals who are told they are ineligible
do not continue the application process and thus do not make any visits to the office. Two
factors could explain the pattern in the Midwestern Large Urban county, though we do not have
the data to test either hypothesis. First, individuals can apply for food stamps at several service
centers in the city and thus not visit the main welfare office. Second, some individuals who call
the office for information may choose not to continue the application process.
At the other extreme, 10 percent of the sample in the Midwestern Rural county reported
4 or more in-person visits. Staff report that it is not uncommon for applicants to make two trips
41
Exhibit 3.5
NUMBER OP CONTACTS WITH THE POOD STAMP OFFICE
Southern
Large
Urban
County
State
Rural
County
Midwestern State
Large
Urban
County
Mid-Sized
Urban
County
Rural
County TOTAL
OVERALL MEAN
(Standard error)
2.3
(0.1)
2.1
(0.1)
2.1
(0.1)
2.3
(0.1)
2.3
(0.1)
2.2
(<0.1)
By Strata:
Did not file
Mean
(Standard error)
1.7
(0.3)
1.8
(0.3)
1.3
(0.1)
1.1
(0.1)
1.3
(0.2)
1.4
(0.1)
Filed, but
were not
interviewed
Mean
(Standard error)
2.0
(0.1)
1.5
(0.2)
1.7
(0.2)
1.5
(0.1)
1.9
(0.3)
1.7
(0.1)
Were
interviewed,
but were not
approved
Mean
(Standard error)
2.1
(0.1)
2.3
(0.2)
2.5
(0.1)
2.6
(0.2)
2.1
(0.2)
2.4
(0.1)
Ware
approved
Mean
(Standard error)
2.5
(0.1)
2.2
(0.2)
2.3
(0.2)
2.5
(0.1)
2.7
(0.2)
2.4
(0.1)
Percent of
respondentB with:
1 contact
6 or more
19.1%
42.4
31.4
4.4
2.4
0.4
30.5%
43.8
14.8
8.1
2.8
0.0
34.3%
38.8
17.8
3.9
5.2
0.2
19.2%
44.4
28.2
5.1
2.3
0.9
25.7%
40.7
17.0
13.7
0.4
2.5
25.4%
41.3
24.8
4.7
3.4
0.5
Unweighted N 187 93 159 157 110 706
Source: Data from 1990 aurvey of 706 food stamp applicanta in five countiea.
42
Exhibit 3.6
NUMBER OP IN-PERSON VISITS
TO THE POOD STAMP OPPICE
Southern State Midwestern State
Large
Urban
County
Rural
County
Large
Urban
County
Mid-sized
Urban
County
Rural
County TOTAL
OVERALL MEAN
(Standard error)
1.6
(0.1)
1.3
(0.1)
1.5
(0.1)
1.8
(0.1)
1.9
(0.1)
1.6
(<0.1)
By Stratat
Did not file
Mean
(Standard error)
1.4
(0.2)
0.6
(0.2)
0.6
(0.2)
0.7
(0.1)
0.8
(0.2)
0.8
(0.1)
Piled, but
were not
interviewed
Mean
(Standard error)
1.4
(0.1)
0.9
(0.1)
1.1
(0.1)
1.0
(0.1)
1.7
(0.3)
1.2
(0.1)
Were
interviewed,
but were not
approved
Mean
(Standard error)
1.7
(0.1)
1.4
(0.1)
1.9
(0.1)
1.9
(0.1)
1.8
(0.1)
1.8
(<0.1)
Here
approved
Mean
(Standard error)
1.7
(0.1)
1.5
(0.1)
1.9
(0.1)
2.0
(0.1)
2.2
(0.2)
1.8
(0.1)
Percent of respondents
with:
no viaite
1
2
3
4 or more
0.9% 9.0%
44.7 55.6
45.9 31.6
8.2 3.9
0.4 0.0
11.3% 2.9%
40.0 34.9
37.9 50.9
9.4 7.5
1.4 3.8
3.3% 5.5%
39.6 41.3
34.1 42.9
12.9 8.6
10.1 1.6
Unweighted N 187 93 159 157 110 706
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
43
to bring in required documents. In addition, applicants who are more than five minutes late for
their interview are not seen that day and must reschedule for a different day; these applicants
make two visits just to complete the certification interview.
As was the case for contacts in general, those who got further through the process made
more in-person visits on average. The mean number of such visits was 0.8 for those who did
not file, compared with 1.8 for both strata that got as far as the certification interview-a
i'atistically significant difference at the 1 percent level. (Observed differences of 0.2 visits or
more between pairs of counties are statistically significant at the 5 percent level, or better.)
Applicants and potential applicants report making relatively few telephone calls and
sending even fewer letters to the food stamp office, as Exhibit 3.7 shows. Overall, only 39
percent of respondents reported calling the food stamp office in the course of their application
process, and the great majority of these did so once only. Thus, the average number of calls
per respondent was 0.5. Calls were substantially more frequent in the Southern Rural county,
where 57 percent of the respondents made at least one call, and the mean number of calls was
0.7. The mean for this site differs significantly from the means for all the Midwestern sites, at
the 5 percent level or better. As discussed above, potential applicants at this site, in contrast
with other sites, were encouraged to call to request an application form.
Overall, 12.1 percent of respondents sent a letter to the food stamp office. Only one
respondent in the sample sent more than one letter. The proportion of respondents using the
mail ranged from 0.2 percent in the Midwestern Rural county to 14 percent in the Midwestern
Large Urban county.
It has been assumed for purposes of calculation that each mail contact cost $0.25. This
may be an overestimate of the true cost because postage-paid envelopes were supplied in the
offices for applicants to return required documents. The estimated total cost of mail contacts
was trivial, however, at only $0.03 per person.
44
Exhibit 3.7
NUMBER OF TELEPHONE CALLS AND LETTERS
TO THE FOOD STAMP OFriCE
Southern State Mic western State
Large
Urban
County
Rural
County
Large
Urban
County
Mid-sized
Urban
County
Rural
County TOTAL
TELEPHONE CAWS:
OVERALL MEAN
(Standard error)
0.6
(0.1)
0.7
(0.1)
0.5
(0.1)
0.4
(<0.1)
0.4
(0.1)
0.5
(<o.i)
By Strata:
Did not file
Mean
(Standard error)
0.3
(0.1)
1.0
(0.2)
0.7
(0.1)
0.4
(0.1)
0.5
(0.1)
0.6
(0.1)
Filed but were not
interviewed
Mean
(Standard error)
0.5
(0.1)
0.5
(0.2)
0.4
(0.1)
0.4
(0.1)
0.2
(0.1)
0.5
(0.1)
Were interviewed, but
were not approved
Mean
(Standard error)
0.2
(0.1)
0.8
(0.1)
0.5
(0.1)
0.5
(0.1)
0.3
(0.1)
0.4
(<0.1)
Here approved
Mean
(Standard error)
0.6
(0.1)
0.6
(0.1)
0.4
(0.1)
0.4
(0.1)
0.5
(0.1)
0.5
(<o.i)
Percent of respondents
with:
0 calls
1 call
2 or more calls
58.8%
29.1
12.1
43.2%
45.4
11.4
64.0%
25.7
10.2
61.1%
37.3
1.6
61.7%
34.2
4.1
61.0%
29.5
9.5
LETTERS:
OVERALL MEAN
(Standard error)
0.1
(0.02)
0.1
(0.03)
0.1
(0.03)
0.1
(0.03)
0.0
(0.00)
0.1
(0.01)
By Strata:
Did not file
Mean
(Standard error)
0.0
(0.00)
0.2
(0.09)
0.0
(0.02)
0.0
(0.00)
0.0
(0.00)
0.0
(0.01)
Filed, but were not
interviewed
Ml
(Standard error)
0.0
(0.03)
0.1
(0.09)
0.1
(0.06)
0.1
(0.06)
0.0
(0.00)
0.1
(0.02)
Here interviewed, but
were not approved
Mean
(Standard error)
0.1
(0.04)
0.0
(0.03)
0.2
(0.05)
0.1
(0.06)
0.0
(0.02)
0.1
(0.02)
Here approved
Mean
(Standard error)
0.1
(0.04)
0.1
(0.04)
0.1
(0.05)
0.1
(0.05)
0.0
(0.00)
0.1
(0.02)
Unweighted N 187 93 159 157 110 706
Source: Data from 1990 ilurvey of 706 food atamp applicaiits in five counties.
45
Cost Per Visit
Information is available on three types of out-of-pocket costs for in-person visits:
transportation, babysitter fees, and lost wages.1
The most common modes of transportation used to get to the food stamp office are using
one's own car and being driven, as Exhibit 3.8 shows. On average, approximately one-third of
individuals who make office visits use each of these modes. In all but the Midwestern Large
Urban site, driving one's own car is more common than being driven.
Across all sites, most of the remaining respondents used public transportation, although
a few borrowed a car, walked, or used some other means. The bus system in the Midwestern
Rural county serves only the town in which the office is located, and thus relatively few
applicants travel by bus. No public transportation system exists in the Southern Rural county,
where over 5 percent of respondents reported walking to the office. Staff report that a concen-tration
of clients in this site lives in an area about two miles from the office and they assume that
some of these clients walk to the office.
The mean transportation cost per visit for all five sites was $1.75, ranging from $1.10
in the Midwestern Rural county to $2.15 in the Midwestern Large Urban county.2 Maximum
'The next three exhibits are based on survey questions which pertained to the respondent's
usual experience when visiting the food stamp office. The data for each individual were
weighted by the number of visits made. The statistical significance of observed differences in
these tables is not examined. The sample sizes are often small because the tables pertain only
to those who made a visit. In addition, the data presented in these tables are used to calculate
the total costs discussed in the previous section, for which tests of statistical significance are
calculated.
2Costs of $0.24 per mile were ascribed to trips made in one's own car. For clients who
were driven or who borrowed a car, as well as for those who used public transportation, the
reported charges were used.
All offices except the Midwestern Large Urban office bad adequate free parking lots
attached to the building. Applicants in the Midwestern Large Urban office are required to pay
$.75 per hour for on-street parking or $3 to $6 for garage parking. Staff report that many
applicants park on the street, allow the meter to expire, and end up with a parking ticket. We
did not collect data on parking costs in the applicant survey and thus could not incorporate these
costs into our estimates.
46
Exhibit 3.8
TRANSPORTATION COSTS POR OPPICE VISITS
Southern State
Large
Urban
County
Rural
County
Midwestern State
Large
Urban
County
Mid-Sized
Urban
County
Rural
County TOTAL
Percent of
respondents
going by:
Own car
Driven
Borrowed Car
Bus
Taxi
Poot
Other
Cost per
roundtrip
Mean
Median
95th
percentile
Unweighted N
40.7% 53.2% 21.7% 46.0% 60.1% 35.6%
28.6 31.5 43.7 30.8 29.2 34.6
5.6 8.8 4.5 5.8 6.0 5.3
16.4 0.0 26.8 15.1 2.6 19.2
0.0 0.0 0.1 0.0 0.0 0.0
4.7 5.4 3.3 2.0 1.8 3.6
4.1 1.2 0.0 0.3 0.3 1.7
$1.62
$1.50
$1.47
$0.96
$2.15
$1.68
$1.39
$1.44
$1.10
$0.72
$1.75
$1.44
$5.00 $5.00 $9.60 $4.80 $3.12 $5.00
182 82 143 151 107 665
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
47
costs were as high as $20.00 for an individual who took a taxi and $13.00 for an individual who
paid for the use of a car.
Approximately 8 percent of respondents who visited the food stamp office reported that
they paid babysitters to look after their children or other dependents while they visited the food
stamp office (Exhibit 3.9).' The proportion of sample members incurring these expenses ranged
from 6 percent in the Southern Urban office to 9 percent in the Midwestern Large Urban and
-Southern Rural offices. The mean averaged over all individuals who made office visits was less
than a dollar for the sample as a whole. For those who incurred babysitting costs, however, the
mean was about $9. These data as well as our office observations suggest that many applicants
bring their children along with them.
Just over 9 percent of the sample indicated that they lost pay in order to visit the food
stamp office (Exhibit 3.10). This proportion ranged from none in the Southern Rural county,
to 11 percent in the Midwestern Large Urban county. Averaged over all office visits, the mean
amount of lost wages was about $2 for the sample as a whole. For those who incurred these
costs, the mean was $23. According to staff in all the local offices and the state offices,
employed persons should not hive to lose pay to apply for food stamps. Special interview times
will be arranged for individuals who request them. However, from our observations, we
hypothesize that many employed people do not know that this option is available.
Summary. The total out-of-pocket cost per visit was $4.56 for all five sites combined,
consisting of $1.75 for transportation, $0.71 for child care, and $2.09 for forgone wages (adding
the components from Exhibits 3.8-3.10). The mean was lowest in the Southern Rural county-only
$1.79-and ranged from $3.49 to $5.20 in the other counties. The 95th percentile values
show the same pattern: $6 in the Southern Rural county and between $22 and $27 in the other
four counties.
The primary source of this variation is forgone wages. While mean transportation costs
vary by only a dollar among the five counties, and mean child care costs are under a dollar in
'Maximum values rather than 95th percentiles are presented in Exhibits 3.9-3.11, because
relatively few individuals in each county incurred costs.
48
Exhibit 3.9
BABYSITTING EXPENSES FOR OFFICE VISITS
Southern state
Large
Urban Rural
Midwestern State
Large
Urban
Mid-Sized
Urban Rural
County County County County County TOTAL
Pe.cent of
respondents who
paid a sitter 6.2% 8.9% 9.1% 7.7% 8.4% 7.6%
For those who paid:
Mean amount
Median amount
Maximum
amount
Unweighted N
$9.70
$7.00
$20.00
13
$3.67
$2.00
$12.00
8
$10.14
$10.00
$50.00
21
$8.12
$5.00
$31.00
16
$4.17
$5.00
$5.00
9
$9.28
$7.00
$50.00
67
Overall Mean
Per Visit $0.60 $0.32 $0.92 $0.62 $0.35 $0.71
Unweighted N 182 82 143 151 107 665
Source: Data from 1990 survey of 706 food stamp applicants in five counties,
49
Exhibit 3.10
FORGONE WAGES FOR OFFICE VISITS
Southern State
Large
Urban Rural
County County
Midwestern state
Large
Urban
County
Mid-Sized
Urban
County
Rural
County TOTAL
Percent of
respondents who
lost wages
For those who
lost wages:
Mean amount
Median amount
Maximum
amount
Unweighted N
Overall Mean
Per Visit
Unweighted N
9.3% 0.0% 10.6% 5.9% 8.6% 9.1%
$25.08
$18.00
$0.00
$0.00
$19.95
$20.00
$27.09
$32.00
$23.64
$25.00
$23.06
$25.00
$100.00
15
$0.00
0
$64.00
17
$32.00
10
$50.00
6
$100.00
48
$2.34 $0.00 $2.12 $1.61 $2.04 $2.09
182 82 143 151 107 665
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
50
each county, forgone wages impose no costs in the Southern Rural county but impose an average
cost of about $2 in each the other four sites.
Cost Per Telephone Call
Only 10.5 percent of those who made telephone calls reported incurring costs for those
calls. The percent ranged from none in the Midwestern Rural to 17 percent in the Southern
Urban county, as Exhibit 3.11 shows. The maximum cost for a call was $2.50 in one rural site.
The average cost over all calls was only $0.02.
Costs of Documentation
The final out-of-pocket expense of the application process involves the cost of obtaining
documents necessary to verify the household's circumstances. The survey queried respondents
on costs incurred for travel to find documents, duplication of documents, and charges for the
documents themselves.1 In interpreting the data, it should be borne in mind that respondents
are likely to intermingle documentation costs for AFDC with those for the Food Stamp Program,
if they applied for both. In fact, the survey showed that documentation costs for PA applicants
exceeded those of NPA applicants. However, there is no way to disentangle the costs for the
different programs. Applicants do not generally know which documents are necessary for the
AFDC program and which are necessary for the Food Stamp Program and thus approach
verification as a single issue.
Statistics are presented in Exhibit 3.12 for two subgroups of particular interest: those
whose applications were approved, and those who attended the certification interview but did not
complete the documentation process. The former group is comprised of individuals who fulfilled
all the necessary requirements. The latter group, in contrast, consists of individuals who
dropped out during this stage of the process, and hence may have had difficulty with the
'Travel costs to take the documents to the office are excluded here, as they are already
included in costs of contacts with the food stamp office.
51
Exhibit 3.11
C08T8 OF TELEPHONE CALLS
Southern State
Large
Urban Rural
County County
Midwtitern sut?
Large
Urban
County
Mid-sized
Urban
County
Rural
County TOTAL
Percent of
callers who
incurred costs
Mean cost for
those who
incurred costs
Maximum cost
Unweighted N
OVERALL MEAN
COST PER CALL
Unweighted N
17.0% 7.4% 5.3% 7.7% 0.0%
Source: Data from 1990 survey of 70S food stamp applicants in five counties.
10.5%
$0.18 $0.70 $0.23 $0.32 $0.00 $0.21
$0.25 $2.50 $0.25 $1.10 $0.00 $2.50
6 3 3 7 0 19
$0.03 $0.05 $0.01 $0.02 $0.00 $0.02
63 51 59 67 36 276
52
Exhibit 3.12
COSTS OF DOCUMENTATION
Southern State Midwestern State
Large Large Mid-Sized
Urban Rural Urban Urban Rural
County County County County County TOTAL
ApPF9v«lp:
Percent incurring
documentation
costs 52.8% 18.9% 48.7% 55.6% 60.0% 51.8%
Mean cost $2.49 $1.28 $5.35 $4.76 $5.35 $3.76
(Standard srror) (0.51) (0.65) (1.32) (1.15) (1.45) (0.45)
Median cost $0.10 $0.00 $0.00 $1.00 $3.53 $0.13
95th percentile $11.82 $11.00 $25.20 $20.00 $28.00 $20.00
Unweighted N 66 29 44 53 35 227
For thoss who
incurred costs:
Mean travel cost
Msan duplicating cost
Mean charge for
documents
Total mean cost
Unweighted N
Interviewees who did
n°t Complete documentation:
Percent incurring
documentation costs
Mean cost
(Standard srror)
Median cost
95th percentile
Unweighted N
For those who incurred
costs:
$3.45
$0.64
$3.47
$0.67
$4.66
$0.89
$7.00
$0.75
$5.88
$0.24
$4.52
$0.71
$0.63
$4.72
33
$2.67
$6.80
8
$5.44
$10.99
23
$0.80
$8.55
28
$2.79
$8.92
20
$2.03
$7.26
112
36.7% 40.8% 51.7% 46.9% 51.1% 47.6%
$1.59 $2.33 $8.17 $5.16 $1.93 $6.01
(0.67) (1.34) (2.60) (1.75) (0.54) (1.00)
$0.00 $0.00 $1.00 $0.00 $1.00 $0.00
$9.00 $20.00 $38.15 $30.00 $5.00 $38.15
31 13 25 34 21 124
Mean travel cost $2.86 $4.14 $8.10 $5.72 $3.08 $6.64
Mean duplicating cost $0.56 $0.02 $0.26 $1.02 $0.47 $0.45
Mean charge for
documents $0.91 $1.54 $7.45 $4.26 $0.23 $5.54
Total mean cost $4.33 $5.70 $15.81 $11.01 $3.79 $12.63
Unweighted N 10 4 16 15 11 56
TOTAL COST OF
DOCUMENTATION FOR
ALL RESPONDENTS
Mean
(Standard error)
Unweighted N
$2.04 $1.31 $3.76 $3.91 $4.02 $3.06
(0.30) (0.35) (0.61) (0.62) (0.67) (0.24
187 93 159 157 110 706
Source: Data from 1990 survey of 706 food stamp applicants in five counties.
53
documentation requirements.1 The final line of the exhibit presents summary statistics for all
respondents (including those who dropped out before the certification interview and those who
completed verification requirements but were denied benefits).
The mean cost of documentation across the entire sample was $3.06. This mean varied
from $1.31 in the Southern Rural county, to $4.02 in the Midwestern Rural county (the means
for each of the two Southern sites differ from the means for each of the three Midwestern sites
at the 5 percent level, or better). One reason costs are higher in the Midwestern Large Urban
and Rural counties may be that a higher proportion of respondents applied for AFDC in these
sites than in the Southern sites. Documentation costs for those applying for AFDC and food
stamps exceed the costs for those only applying for food stamps. In addition, the Midwestern
Rural county places a great emphasis on avoiding quality control errors and thus caseworkers
are quite strict about acceptable forms of verification, which might explain why the mean cost
is highest in this site. Substantial documentation costs were reported by some individuals who
did not attend the certification interview, and even by individuals who did not file an application.
However, for 60 percent of all the respondents the cost was zero.
When we focus on the respondents who were approved, we find means ranging from
$1.28 in the Southern Rural county to $5.35 in two of the Midwestern counties. (The means
in each of the two Southern sites differ from the means in each of the three Midwestern sites at
the 10 percent level, or better.) The 95th percentile values were also substantially higher in the
Midwestern sites than in the Southern sites, $20 to $28 versus $11 to $12, suggesting that some
especially costly documents are sometimes required in the Midwestern sites. We do not know
why some people spend substantial amounts of money on verification requirements. State and
local staff report that they do not require any costly documents, particularly in the Food Stamp
Program. However, some applicants may not understand the requirements and thus incur
unnecessary costs. For example, some applicants may visit a doctor for verification of a
disability to exempt them from work registration requirements. Or, AFDC applicants may incur
'This group includes people who reported in the survey that they did not complete their
documentation. It also includes some who said in the survey that they provided all their
documents, but whom the state's automated casefile data classified as denied for failure to
provide all documents.
54
significant costs obtaining out-of-state birth certificates or copies of court records of divorce
decrees. Caseworkers can and will obtain these documents without charge though some
applicants may not be aware of this. It should be noted that the median value is quite low in
all sites but the Midwestern Rural county, indicating that half of all individuals who are
approved incur virtually no out-of-pocket costs for verification.
Detail is also presented on the three components of documentation cost for those
ipproved individuals who incurred any costs. For respondents in four of the counties,
transportation costs are the dominant component. In the Midwestern Large Urban county,
however, charges for the documents themselves are of roughly equal importance.
Among interviewees who did not complete documentation, nearly the same percentage
incurred some documentation costs as among those who were approved. It is likely that this
reflects two different phenomena. Some individuals in this subgroup did not attempt to obtain
any documents, and hence incurred no costs, while others eventually gave up on completing their
documentation after having incurred significant costs.
The mean cost of documentation for all counties combined is substantially higher for
these dropouts than for those who were approved, $6.01 versus $3.76 respectively, a statistically
significant difference at the 10 percent level. The difference comes primarily from higher
charges for documents, although transportation costs were also somewhat higher.1
Total Monetary Costs
For the entire sample of respondents, the mean total out of pocket cost is $10.40.
Seventy percent of this is attributable to office visits, and virtually all of the remainder is costs
of documentation.
Mean total costs were lowest in the Southern Rural county-under $4 per person. Mean
costs in the remaining sites ranged from $9 to $12. Costs were lower in the Southern Rural
'The appropriate comparison to make is between costs averaged over all individuals i*\ the
two subgroups. For charges for documents the value is $1.05 (0.518 * $2.03) for appn ,ais and
$2.64 (0.476 * $5.54) for the dropouts. For transportation costs, this value is ,2.34 for
approvals (0.518 * $4.52) and $3.16 for the dropouts (0.476 * $6.64).
55
county because applicants make fewer visits to the welfare office than those in other sites,
because no applicant reported losing wages, and because documentation costs are also low in this
site.
One reason some applicants and potential applicants may not have completed the
application process is because they felt the costs of the process were too high. As a rough test
of this hypothesis, we project what the total out-of-pocket costs would have been for dropouts
if they had completed the process (Exhibit 3.13). To generate these numbers, we assumed that
in each site, the number of visits and costs of telephone calls, letters, and documentation would
have been the same for the dropouts as they were, on average, for those who were approved in
that site.1
The cost per visit used in the calculations is each respondent's reported value, where that
is available. However, some individuals who dropped out early in the process made no in-person
visits, so that the cost per visit is unknown. For these individuals, cost per visit was
imputed, based on a site-specific regression model which included such household characteristics
as presence of an earner, presence of young children, and possession of a car. Because of this
imputation process, mean values are not shown for the components of cost per visit. Similarly,
mean values are not shown for the components of other cost elements for which site-level means
were assumed.
We see that the projected costs that would have been incurred by dr