ATTACHMENT_I.pdf

Performance Work Statement (PWS) The Effects of Asset Limits on SNAP Participation and Financial Stability

OVERVIEW

The U.S. Department of Agriculture (USDA) Food and Nutrition Service (FNS) is seeking a contractor to conduct a study that will examine the effects of asset tests on low-income individuals’ financial stability and participation in the Supplemental Nutrition Assistance Program (SNAP)1. The study will be conducted in two phases:

• Phase I will provide an overview of available public datasets that collect information on SNAP receipt, asset levels, savings and other banking information, income sources, employment, housing and general demographic information, and assess whether there is sufficient data to address the research questions of interest to FNS.

• In Phase II, if sufficient data are available and analyses are deemed feasible by FNS, the contractor shall develop plans for and conduct proposed analyses of the effects of asset limits on low-income SNAP households.

BACKGROUND

SNAP is the largest of the domestic food and nutrition programs administered by FNS and its State partners. Benefits are deposited electronically in households’ accounts and accessed using an electronic benefit transfer (EBT) card, which can be used to purchase food in authorized food stores. The program is authorized by Congress, administered nationally by FNS, and operated by State agencies and their local offices. During fiscal year 2013, SNAP served an average of 47.6 million people each month and delivered a total of $76.1 billion in benefits.

SNAP benefits are funded entirely at the Federal level, however, the administrative costs of implementing the program are split between FNS and State agencies. States are also responsible for administering the program and are able to choose specific state options that impact the application process, eligibility rules, and other aspects of program participation. Currently, only a few States use an asset test to determine households’ eligibility because most States have Broad-Based Categorical Eligibility (BBCE) policies.

Under BBCE States can determine that a household is categorically, or automatically, eligible for SNAP based on receipt of benefits from the Temporary Aid for Needy Families program (TANF), Supplemental Security Income (SSI), or state funded General

1 In October 1, 2008, the Food Stamp Program changed its name to the Supplemental Nutrition Assistance Program (SNAP). We use SNAP throughout the document, with the exception of when citing past reports that contain the words “Food Stamp” in their titles.

Assistance programs. These households have already gone through eligibility determination for other programs serving low-income individuals and can therefore bypass the income and resource tests that would normally be part of the SNAP application process.

SNAP benefits for categorically eligible households are still determined under the same rules as other households and a household may be categorically eligible for the program but still have a net income that is too high to receive a SNAP benefit. Under BBCE, States can also determine eligibility for households below a certain gross income level based on receipt of non-cash TANF benefits, which can include access to pamphlets or a referral to an “800” number.

Many States that have implemented BBCE have eliminated the asset test as a SNAP eligibility requirement. Five States have raised the asset limits to higher than the minimum through BBCE.2 Asset limits are indexed to inflation and the minimum a state can implement in the current fiscal year is $2000 for most households and $3250 for households containing an elderly or disabled member. If they were not participating in BBCE, Idaho, Michigan and Texas would all have asset limits of $5000.

Research on Dynamics and Determinants of SNAP Participation

Research currently exists on the reasons for entry and exit into SNAP, as well as the duration of SNAP spells. While less research has been conducted on the specific impact of asset limits on SNAP spells, there is a body of research on how assets and savings impact the circumstances of low income families. Research on the dynamics of SNAP participation helps to inform the agency on the common reasons that participants leave the program, the reasons for entry, re-entry and gaps in participation, and the barriers that potential recipients see that may keep them from recertification. These studies include limited analysis on assets and savings, but do not delve into how savings and debt may affect SNAP participation.

The most recent FNS study of SNAP participation dynamics used data from the Survey of Income and Program Participation (SIPP) 2004-2006 and shows that the likelihood of SNAP entry depended on an individual’s past and current circumstances (Mabli, et. al, 2010). For example, individuals were more likely than those with the same income to enter the program if they had been on SNAP before. Additionally, between 2004 and 2006 the SNAP caseload grew even as the economy was improving. More people entered the program during this time period than during the early 2000’s, people participated for longer spells and fewer people experienced exit triggers—new job, marriage, or change in family composition—causing them to leave SNAP. While, more than half of the individuals who entered SNAP during the study period exited within one year, more than half of these participants re-entered the program within two years. Finally, the median spell length for the panel was 18 months, indicating that long term SNAP cases are prevalent.

A companion to the study on the dynamics of SNAP participation is Determinants of Supplemental Nutrition Assistance Program Entry and Exit (Mabli, et. al, 2010). This report specifically examines factors associated with entering and exiting SNAP, particularly

2 http://origin.www.fns.usda.gov/snap/rules/Memo/BBCE.pdf in terms of individual and family demographic and economic characteristics, state economic measures, and SNAP policies. The most recent Determinants report found that changes in employment, income and family composition were the strongest determinants of SNAP entry, duration, and re-entry during the sample period. Further, economic status and positive net worth were determinants of program entry and duration, as families with positive net worth or with a member who was currently employed were less likely to get on the program and if they did enter SNAP, were more likely to spend less time on the program. Individuals living in States that eliminated the asset test through BBCE had an increased probability of entering SNAP. Individuals living in families that experienced a documented job loss also had a monthly entry rate that was 94 percent higher than individuals in families with no job loss, suggesting that these families were unable to adjust to income shocks. Finally, individuals who lived in States with higher wages for low wage workers were less likely to re-enter SNAP after exiting the program.

Assets and Low Income Families

Research on asset building over an individual’s lifetime indicates that intergenerational transfer of assets is typically imperative when it comes to maintaining or improving one’s socioeconomic status. While some economic mobility occurs, socioeconomic status tends to be a perpetuating cycle and individuals with lower-income parents are likely to remain lower-income as they transfer into adulthood. This is generally due to the transfer of wealth from generation to generation, and the ability to transfer resources to children can have a large impact on future opportunities for children, particularly when it comes to increasing their likelihood of asset accumulations. This research also shows that family structure can have a large impact on asset accumulation. Single mothers in particular are at a disadvantage when it comes to savings when compared to married-couple families. The number of children in a family also has an impact on net worth as an adult, possibly due to distribution of family resources from parents to children (Rank, 2008). Although this research looks at asset building over the life cycle and the impact of savings on long term wealth and stability, it does not discuss the relationship between savings and SNAP.

Previous research on assets and liabilities generally focuses on how savings and debt can affect a low-income family’s ability to participate in formal banking systems and prepare for income shocks. Building assets and avoiding debt can help prepare families for dealing with disruptions in steady income and can contribute towards financial and economic independence (Carasso and McKernan, 2007). Carasso and McKernan’s tabulations from the 2004 Survey on Consumer Finances (SCF) show most bottom quintile families do not own a home, hold a retirement account, and have checking accounts valued around $600. Additionally, Social Security and Medicare comprise about 90% of expected income for families in the bottom quintile. Additional research using the SCF indicates that families making less than $20,000 typically have no retirement savings. These families are generally more likely to be eligible for means-tested programs, are less likely to save for future expenses because savings can be perceived as a tax on current earnings since they can lead to disqualification of federal and state benefit programs (McDonald, Orszag, and Russell, 2007 ).

Research on the impact of welfare program rules on asset holdings for low-income families found that more generous asset limits are associated with increased liquid asset holdings for low-education single mothers and families (McKernan and Ratcliffe, 2007). This research also found that more generous vehicle asset limits and expanded categorical eligibility in SNAP are generally associated with increased vehicle asset holdings for low-education single mothers and families. Finally, this study found that receiving a more generous State Earned Income Tax Credit (EITC) amount generally resulted in lower liquid asset holdings.

A recent analysis looked at the impact of privately funded conditional cash transfers (CCT) distributed to families in low-income neighborhoods in New York City. The program launched in 2010 by the Center for Economic Opportunity (CEO), and was evaluated in 2013. CEO offered cash assistance to low-income families to reduce immediate hardship, but required that the assistance go partially towards efforts to build up “human capital” in order to reduce the risk of long term and second generation poverty. The program tied cash awards to specified activities and outcomes in children’s education, families’ preventive health, and parents’ employment. An evaluation of the intervention used a randomized controlled trial and found that the CCT reduced current poverty and material hardship, including hunger and some housing hardships. The CCT also helped parents increase savings and reduce reliance on families and friends for cash loans. Although there was no significant impact on whether or not being in the study group reduced the likelihood of being on SNAP, participants in the study group were less likely to be at or below the poverty line, more likely to be over 130% of the poverty line, more likely to report earnings as an income source in the month prior when compared with the control group. Further, participants in the study group were more likely to have any type of banking account, less likely to have zero dollars in savings, less likely to owe back rent once the intervention ended when compared to the control group (Riccio, et. al, 2013).

Some of these studies suggest that participation in the mainstream financial market increases economic stability, which could lead to shorter term participation in SNAP. Although there is little research linking asset accumulation to SNAP receipt, the goal of this study is to see if a link exists between savings and reduced participation. The above studies should inform the discussion of and analysis on the effects of asset accumulation on SNAP participation.

Available Datasets

Very few datasets have substantial and thorough asset information, particularly for low-income families. Several datasets collect some asset information, however, most datasets with any asset information have some response and reliability concerns. In 2007, the Office of the Assistant Secretary for Planning and Evaluation at the Department of Health and Human Services conducted a review titled “Primary Data Sources for Holdings of Low-Resource Households’ Assets”. The review identified 12 datasets that could potentially provide important information on low-income households’ assets and income liabilities. The datasets are:

1. American Dream Demonstration Account Monitoring (ADD-AM) Data

2. American Dream Demonstration Experiment (ADD-E) Data

3. Assets For Independence Act (AFIA) Evaluation Data

4. Consumer Expenditure Survey (CEX)

5. Current Population Survey (CPS)

6. Health and Retirement Study (HRS)

7. Home Mortgage Disclosure Act (HMDA) Data

8. National Longitudinal Study of Youth 1979 (NLSY79)

9. National Survey of Family and Households (NSFH)

10. Panel Study of Income Dynamics (PSID)

11. Survey of Consumer Finances (SCF)

12. Survey of Income and Program Participation (SIPP)

Based on their evaluation, ASPE identified three datasets that would have the greatest potential for asset research: the Survey of Consumer Finances (SCF), the Survey of Income and Program Participation (SIPP), and the Panel Study of Income Dynamics (PSID).

Survey of Income and Program Participation

The SIPP contains detailed information on financial assets, including checking and savings accounts, stocks and IRAs, and detailed information on nonfinancial, tangible assets such as personal residences and vehicles. Additionally, the SIPP aims to collect significant information on household liabilities such as mortgages and credit card debt. SIPP has extensive variables on SNAP receipt and reasons for SNAP participation, as well as extensive information on income and eligibility for low income benefit programs. The large sample size in the SIPP is an advantage as it allows researchers to analyze large sub-populations. SIPP panels can be matched with administrative data, particularly data that contains employer information and earning history. SIPP panels are longitudinal and typically last several years. This allows for researchers to observe changes in assets and debt over time, but the length of panels also contributes to a large amount of attrition. However, the SIPP has a relatively high rate of non-response when compared with other, similar datasets.

Panel Study of Income Dynamics

The PSID collects asset data that provides accounting on major components of net worth and includes variables on home values, financial assets, tangible assets, and debt. The PSID is a longitudinal panel, following the same families for several decades, which allows researchers to look at intergenerational wealth and trends of assets on family history. The PSID asks questions on SNAP receipt, as well as questions on participation in other low income benefit programs, and on income and poverty levels. Additionally, there is little missing data in the PSID on asset variables. PSID asks questions on asset accumulation and savings, however, there are fewer questions on this subject than in comparable data sources. Finally, the sample size on the PSID is much smaller than the SIPP, which reduces some of the likelihood of strong subgroup analysis.

Survey of Consumer Finances

Of the datasets available, the SCF is the most focused on asset ownership and provides the most in-depth questions on assets and liabilities. The SCF is a cross-sectional survey that provides data every three years. It provides an accurate estimate of net worth in the US by oversampling wealthy populations who own a large proportion of overall assets. The SCF asks about food purchasing patterns, including food purchased using SNAP benefits. It also asks about purchases made with debit and EBT cards. Additionally, the SCF asks for income information, so researchers can reasonably assess SNAP eligibility based on income levels. The small sample size, especially for low-income individuals, and lack of other non-asset related questions limits its usefulness when considering analysis of low-income sub populations.

FNS-managed datasets

The MATH Microsimulation models allow users to simulate changes to program rules and provide estimates of the potential effects of changes on program eligibility, participation, and benefits (Filion, 2013). There are two models within the platform: the MATH SIPP+ model and the Quality Control (QC) Minimodel. The MATH SIPP+ model uses SNAP administrative data along with SIPP data and is used for expansive reforms that affect individuals not currently participating in SNAP. The current version of the QC Minimodel is based on the 2012 SNAP Quality Control Data and can be used to assess the impact of program changes on current SNAP beneficiaries only. The QC data is collected using a nationally representative sample of SNAP participants in order to measure the accuracy of state eligibility and benefit determinations. QC data can be used to establish terms and circumstances of current SNAP participants, as well as benefit levels and the impact of changes in program rules on current participants. The basic functions of both models can be run on a web-based platform, so no special software is needed and all users have their own accounts and the ability to run their own reforms. The MATH SIPP+ model can be used to assess the impact of changing asset limits that currently exist, along with the impact of changing any current eligibility or income rule. The QC Minimodel does not have asset data for households in States without asset limits, therefore it cannot be used to model changes to asset limits.

SCOPE

FNS is seeking a valid and reliable Contractor to examine the effects of asset tests on low-income individuals’ financial stability and participation in the Supplemental Nutrition Assistance Program (SNAP). This Performance Work Statement describes the objectives and requirements for the required study.

The contractor shall provide all management, supervision, labor and equipment, and shall plan, schedule, coordinate and assure effective performance of all services/tasks as described in Solicitation AG-3198-S-14-0040, inclusive of this Performance Work Statement.

STUDY OBJECTIVES AND RESEARCH QUESTIONS

The purpose of this study is to advance knowledge of the effects of asset limits on SNAP participation and the ability of low income families to participate in traditional financial markets, weather financial shocks, and generally improve household circumstances such that they can rely less on SNAP and other programs over time. FNS is interested in the availability of data to answer the following research questions:

1. What are the characteristics and circumstances of households with high assets and low income?

2. What are the characteristics and demographics of States that have high SNAP asset limits versus States with low SNAP asset limits.

a. Do States with higher or no asset limits have caseloads with shorter SNAP spell lengths, and can those shorter spells be attributed to the accumulations of assets that allow them to weather financial shocks?

b. Conversely, might States with low asset limits have shorter and/or fewer number of spells per household because households that accumulate assets can more easily exceed the asset limit and lose eligibility?

c. What percentage of people in States with asset limits are income eligible but asset ineligible?

3. What types of asset holdings are typical for SNAP households, and how do those vary between households in States with higher and lower asset limits?

4. Do asset tests contribute to SNAP caseload churn?

a. Are SNAP spells longer for households with no assets versus households that have some assets but still fall within

SNAP asset eligibility?

5. Do asset tests limit the ability of low income families to adequately save or prepare for income shocks?

a. Do asset limits prevent SNAP households from participating in mainstream financial markets?

b. Do asset limits prevent SNAP households from participating in programs that provide large lump sum cash transfers such as the Earned Income Tax Credit due to the perception that participation in the EITC will result in elimination from other means tested programs?

c. Do asset limits hinder the ability of low-income households to respond to trigger events that affect household income?

6. What role does debt play in a households ability to gain assets or participate in the formal banking system?

7. Are there differences in SNAP spells for households with debt compared to households that do not have substantial debt?

Objective 1: Review data sources and assess the feasibility of addressing the research questions

The contractor shall conduct a review of all available public datasets that collect information on SNAP receipt, levels of assets, savings and other banking information, income sources, employment, and housing, as well as general demographic information that would be necessary to fully address the research questions of this study. Following the review, the contractor shall produce a feasibility assessment that includes the available options for addressing each research question and recommends an approach for moving forward.

Based on the feasibility analysis conducted to meet Objective 1, FNS will determine whether to conduct the work under Objective 2, which will provide the actual analysis of stated research questions.

Objective 2: Analyze the effects of asset requirements on SNAP participation and spell length.

If the contractor determines that an analysis of asset limits can be completed using public datasets and FNS approves the suggested approach and analysis techniques, the contractor shall develop a detailed study plan that describes the data set(s) to be used, methods for preparing the datasets for use, the analysis approach for each question and the research questions to be addressed, and plans for reporting of findings and delivering the data files and documentation to FNS. Upon approval of the study plan by FNS, the contractor shall execute the analysis.

TASKS and SUBTASKS

CLIN 1.0 –Feasibility Study (Phase I)-Firm Fixed Price Work Portion

Task 1.1- Data overview

After the post-award orientation meeting and discussions with FNS, the contractor shall do a comprehensive evaluation of publicly available datasets that collect information on SNAP receipt, levels of assets, savings and other banking information, income sources, employment, and housing, as well as general demographic information that would be necessary to conduct a comprehensive analysis to achieve the research objectives of this study. The contractor will also review data based on the availability of cross-sectional and longitudinal analysis. As part of this review, the contractor shall assess the data sources in terms of comparability, completeness, accuracy and appropriateness for the study.

The feasibility analysis shall include an assessment of each research question and whether or not it can be answered using available data. The assessment should also include recommendations for changes to questions that would allow them to be answered if the original question cannot be completely addressed using available data.

If multiple data sources are needed to conduct the analyses, the contractor shall discuss how these data sources can be combined. The feasibility determination should include recommendations for the scope of analyses that can be completed within the available resources, as well as pros and cons to each approach and potential limitations of suggested approaches.

Task 1.2- Feasibility Assessment and Recommendations to FNS

The contractor shall submit a report to FNS with findings and recommendations on next steps. The report shall include but not be limited to:

For each dataset

• Concise but complete summary of scope of data and a list of relevant variables, including but not limited to:

o Demographic data o SNAP receipt o Receipt of other food assistance, cash assistance, or income maintenance benefits.

o Banking and savings information o Income sources o Employment status o Housing information o Other variables as necessary

• Discussion of availability and reliability of asset information

• Representativeness of the data at the state and/or national level

• Availability of cross sectional vs. longitudinal data

• Enumeration of the research questions that can be addressed using the dataset.

The feasibility analysis shall at a minimum, address each of the following:

• Given information provided for each dataset, what is the feasibility of an analysis of the effects of asset limits on SNAP participation and the ability of low income families to participate in traditional financial markets, weather financial shocks, and generally improve household circumstances such that they can rely less on SNAP and other programs over time.

• Recommendations for proceeding with one or more analyses, if feasible, which shall include:

o Specification of the research questions that will be addressed by each approach o Pros and cons of each analytical approach o Limitations of the results from each approach

The feasibility report shall contain an executive summary of not more than 5 pages; introduction, with background on the project and discussion of relevant literature; narrative chapters addressing, at a minimum, all items outlined in this task description; tables, charts, and other visual presentations of material to enhance, simplify, or streamline presentation; references; and technical appendices, as necessary.

The draft report shall be in near-final form, with only revisions to address FNS comments needed to finalize the report. If no revisions are requested by FNS, the draft report may be approved as the final deliverable. The draft shall not be submitted for COR review without implementation of the contractor’s quality control plan as described in the contractor’s proposal. Deliverables that have not undergone editing for clarity, format, spelling, and grammar, or do not adhere to requirements of this PWS may be rejected without further review. A draft and final report will be submitted, with the final report incorporating all FNS feedback and answering any questions from FNS reviewers.

Upon submission of the feasibility analysis, the contractor shall conduct a briefing for FNS to discuss the available options and facilitate the decision-making process for moving forward with Objective 2.

CLIN 2- Analysis and Final Report (Phase II)-IDIQ Work Portion

Task 2.1: Study Plan

Upon approval from FNS, the contractor shall prepare a final study plan that further elaborates on the recommendation(s) selected by FNS in Phase I, incorporating information from the feasibility report delivered under Task 1.1 and FNS feedback on the report.

The final study plan shall include:

Detailed plans for creation of the full set of data and analysis files and accompanying documentation.

• Enumeration of all research questions to be addressed.

• Detailed plans for initial analyses to be conducted to assess the integrity and quality of the final data files.

• Description of analyses to be conducted to thoroughly address the research questions.

• Table shells and plans for presentation of findings in the final report.

The contractor shall prepare a draft of the final study plan for FNS review. The study plan shall incorporate and expand on the contractor’s technical proposal for Phase II and adhere to the requirements of this PWS. The contractor shall prepare the study plan as a complete stand-alone document that describes the objectives, research questions, and specific methods for preparing the data files, accomplishing the analysis and reporting for Phase II of the study. After FNS review, the contractor shall make any necessary changes to address FNS comments and submit the final study plan to FNS.

Task 2.2: Data Assessment Memorandum

The contractor shall produce a memorandum detailing the data compilation process and discussing any issues with the data, actions taken to address these issues, and implications on the research. The memorandum shall address any concerns with the data and make recommendations for how to proceed if issues arise with data manipulation and dataset development.

Task 2.3 - Perform Data Analysis Preliminary Results Memo

The contractor shall analyze the data in accordance with the final study plan approved by FNS in Task 2.1.

The contractor shall submit a memorandum and initial set of tables on preliminary analysis to FNS. The memorandum of preliminary findings should function as a standalone document and provide insight into initial findings, problems encountered, solutions for any data issues, and discussion of how these findings will be further elaborated in the final report.

Task 2.4 - Prepare Final Report

The contractor shall prepare a report that addresses study objective 2 and addresses all research questions as described in the final study plan approved by FNS. The report shall focus only on the analyses that were deemed feasible and requested by FNS. The report shall be written in plain language and geared toward an audience of policymakers and program officials, with use of technical appendices as appropriate. At a minimum, the following sections shall be included in the final report:

• An executive summary highlighting the report's major findings.

• An introduction and background section describing relevant issues, literature, program and policy context, and research questions.

• A description of the study methods, including data sources, analytic approach and limitations.

• A detailed discussion of the results by research question, with appropriate summary sections interpreting and discussing these findings in the context of program policy and available literature.

• A concluding chapter, presenting the contractor's thoughts on particularly relevant, useful, or surprising findings, which should provide an integrated discussion of these important findings from throughout the report (rather than a sequential re-statement of findings), be grounded in fact, address policy relevance, and, discuss issues for future research.

• Any additional technical appendixes necessary to fully document the research methods and results.

The contractor shall provide FNS with a detailed outline for the final report. The contractor shall incorporate FNS comments when revising the outline. Based on the final outline, the contractor shall prepare the draft version of the final report.

The contractor shall prepare and submit a report to FNS, in draft, revised draft, and final versions. The report format shall be consistent with the requirements of the most recent version of the Government Printing Office Style Manual. The contractor shall submit all versions of the report in electronic form in Adobe PDF and MS WORD format to FNS for review and comment. The report shall be written for a broad, non-technical audience with more detailed technical appendices as necessary. The focus of the report shall be to address the research questions and objectives as described in the PWS and further elaborated in the contractor’s final Revised Study Plan as approved by FNS.

The draft version of the final report shall be in “near final” form and of a quality that can be circulated within FNS and, at the discretion of FNS, sent to outside stakeholders. It shall include a draft Executive Summary that shall contain a concise description of the project background, the objectives, and the findings. The report shall include background information for this study, the research questions, methods, findings, and conclusions. The report shall include visual representations such as tables and graphs as identified by the contractor as helpful in presentation of information or as requested by FNS.

The contractor shall provide a revised draft version of the final report to FNS, incorporating all comments and revisions suggested by FNS on the first draft. The revised draft version of the final report shall be fully formatted and edited so that there are no grammatical or typographical errors.

Ten hard copies and an electronic version of the revised draft report and a stand-alone copy of the Executive Summary shall be submitted to FNS for clearance. Upon clearance of the report, the contractor shall submit 15 copies of the final report using book binding with the title of the report printed on the spine, or another binding format approved in writing by the COR. The contractor shall deliver electronic files of the full final report, stand-alone Executive Summary document, and the 508-compliant version of the full report in Microsoft Word and .PDF formats.

Task 2.5- Present Findings to FNS

The contractor shall conduct a briefing for FNS to discuss the findings of the study following submission of the revised draft version of the final report. The briefing shall be conducted in a manner that is appropriate for a non-technical or policy audience. The contractor shall provide a draft PowerPoint slide deck for FNS review and a final PowerPoint slide deck that incorporates FNS comments. The final slides shall be provided to FNS in a fully-editable electronic format. Twenty-five paper copies of all presentation materials, with handouts containing no more than 4 slides per page, per side, shall be provided for attendees at the briefing. Graphics not easily read in slide or handout format shall be distributed as stand-alone materials.

Task 2.6- Deliver Dataset and Accompanying Documentation

The contractor shall supply FNS with the final dataset and the analysis files used to produce the findings. Concurrently, the contractor shall submit a codebook that describes the data submitted. Documentation shall include: file structure (data file name, record format, record length, block size, and number of records); code book (record layout including variable names, variable format, variable labels and missing values); variable definitions; imputation procedures; editing procedures; structure of any composite variables; and Stata or SAS code to allow full replication of analyses in the final report by FNS.

Offerors shall prepare and submit to FNS a set of CDs that include the dataset and data analytic files used to produce all results that appear in each version of the final report. An alternate format (for example, flash drive) for these electronic files may be used if agreeable to FNS. With each version of the final report, the contractor shall submit all data files that were used to create the deliverable including the peer review version, the clearance version, and the publication version. These files shall be provided in micro-computer format for use with PC SAS, STATA, or other suitable analytic software agreeable to FNS. In addition, offerors shall provide full documentation in hard copy and on CD for these data files. An alternate format (for example, flash drive) for these electronic files may be used if agreeable to FNS.

PERFORMANCE REQUIREMENTS SUMMARY

PERFORMANCE REQUIREMENTS SUMMARY (PRS)

Performance Measure

Standard

Performance Deliverable Monitoring

Surveillance Methodology

Rating

Surveillance Frequency

Monitoring Performed

Quality of Products or Services

Acceptance or rejection of deliverables is based on the following performance standards and will be rated per the performance ratings in column 5:

Performance Rating.

• Complete per the approved study plan

• Accurate per the approved study plan

• Incorporates FNS’s feedback on prior drafts

• Well-written with proper grammar

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100% Inspection Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Acceptance or Rejection of each deliverable is noted via a Government Transmittal Sheet attached to the deliverable

Schedule On time All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100% Inspection Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Task Orders via the Contractor Performance and Rating System (CPARS)

Cost Control According to the FFP listed in the approved budget per task in the schedule of deliverables

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100% Inspection Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables and each deliverable’s accompanying invoice

Task Orders via the Contractor Performance and Rating System (CPARS)

Business Relations

Professional and respectful of all parties

Regular and ad-hoc meetings between

Semi-Annual Performance Evaluation

Exceptional Very Good Satisfactory

Semi-Annually Contractor Performance and Rating System (CPARS) contractor and Government

Marginal Unsatisfactory

Management of Key Personnel

• Appropriate experience and expertise for the work involved

• Attends scheduled meetings

• Available for ad-hoc meetings as necessary for contractual concerns

• Time is scheduled per the staff loading chart in the contractor’s proposal

Regular and ad-hoc meetings between contractor and Government

Semi-Annual Performance Evaluation

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Semi-Annually Contractor Performance and Rating System (CPARS)

SCHEDULE OF DELIVERABLES

Offerors should carefully review and revise the draft timeline contained in the performance work statement. FNS expects to receive a proposal with firm timelines that have been customized to reflect the contractor’s proposal while incorporating the required timeframes for FNS review of deliverables: FNS requires a minimum of two weeks to review major study deliverables.

Task Deliverable Format/#* Due** (Weeks from award)

Task 0: Project Orientation Meeting

0.0: Project Orientation Meeting

Project orientation meeting agenda n/a 1 Project orientation meeting n/a 2 Project orientation meeting summary memorandum 1 3

CLIN1: Feasibility Assessment Task 1: Data Assessment and Recommendations

1.2: Feasibility Assessment Report

Draft memo 3 12 Revised memo 3 15 Final Memo 3 18

CLIN 2: Analysis and Final Report Task 2.1: Study Plan

2.1 Prepare Study plan

Draft Study Plan 3 21

Revised Study Plan 3 24 Final Study Plan 3 27

Task 2.2: Data memorandum

2.2 Data Assessment

Memo

Draft memo 3 31 Revised memo 3 34 Final memo 3 37

Task 2.3: Perform Data Analysis

Task Deliverable Format/#* Due** (Weeks from award)

2.3 Preliminary Data

Analysis Memo

Draft memorandum of preliminary analysis 3 41 Revised memorandum of preliminary analysis 3 44 Final memorandum of preliminary analysis 3 47

Task 2.4: Prepare Final Report

2.4 Prepare Final

Report

Draft Outline 3 50 Final Outline 3 52 Draft Final Report 3 56 Revised Final Report 3 59 Final Report 25 62

Task 2.5: Present Findings to FNS

2.5 Presentation of

findings

Draft Presentation Materials 1 62 Final Presentation Materials 1 64 Final Presentation at FNS n/a 64

Task 2.6: Codebook and data set documentation

2.6: Codebook and data set documentation

Draft Codebook 1 64 Final Codebook 1 64 Final dataset CD-ROM 64 Final dataset code and documentation CD-ROM 64 n/a Monthly progress reports electronic 10th of each month

*Electronic copies shall be provided on all deliverables in addition to the number of hard copies specified. For documents exceeding 70 pages bound hard copies should be provided. All final reports should be bound.

** Weeks are from date of contract award. Minor changes to due dates may be made by mutual agreement, in writing, of the COR and contractor. Significant changes, including changes that affect the CLIN or contract Period of Performance require approval of the contracting officer and a modification to the contract. Time between deliverables shall be assumed to be split evenly between FNS for review and the contractor for revision.

References

Mabli, James et al., Dynamics of Supplemental Nutrition Assistance Program Participation in the Mid-2000’s, Prepared by Mathematica Policy Research, U.S. Department of Agriculture, Food and Nutrition Service, Office of Research and Analysis Project Officer: Jenny Laster Genser, Alexandria, VA: 2011.

Mabli, James et al., Determinants of Supplemental Nutrition Assistance Program Participation in the Mid-2000s, Prepared by Mathematica Policy Research, U.S. Department of Agriculture, Food and Nutrition Service, Office of Research and Analysis, Project Officer: Jenny Laster Genser, Alexandria VA: 2011

Ratcliffe, Caroline et al., Assessing Asset Data on Low-Income Households: Current Availability and Options for Improvement, U.S.

Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Washington, D.C.: 2007

Rank, Mark, Asset Building Over the Life Course, U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation,, Washington D.C.: 2008

U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, The Effects of Welfare and IDA Program Rules on the Asset Holdings of Low-Income Families, by Signe-Mary McKernan, et al., Washington, D.C.: 2007

Carasso, Adam, et al., The Balance Sheets of Low-Income Households: What We Know about Their Assets and Liabilities, U.S.

Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Washington, D.C.: 2007

McDonald, Gordon, Peter Orszag, and Gina Russell, The Effect of Asset Tests on Saving, The Retirement Savings Project.

Washington, D.C.: 2007

Filion, Kai, et al., Technical Documentation for the Fiscal Year 2012 Supplemental Nutrition Assistance Program Quality Control Database and the QC Minimodel, U.S. Department of Agriculture, Food and Nutrition Service, Office of Policy Support,, Project Officer: Jenny Laster Genser, Alexandria, VA: 2013

Schecter, Bruce and Mark Brinkley, MATHWEB (Web-Based MATH Microsimulation Model) User Documentation, U.S. Department of Agriculture, Food and Nutrition Service, Office of Policy Support, Project Officer: Jenny Laster Genser, Alexandria, VA: 2013

Riccio, James, et. al, Conditional Cash Transfers in New York City: The Continuing Story of the Opportunity NYC-Family Rewards Demonstration, Prepared by MDRC New York, New York: September 2013