SNAP participation program rules, targets, and the nonreporter adjustment form are discussed below.

SimulationMode:

This tells whether the run being performed is a baseline alignment run or a non-alignment run. During an alignment simulation, adjustments are made to the random variable that is used in determining whether a unit eligible for SNAP benefits chooses to participate. The adjusted random variable from the final alignment simulation is saved and used in non-alignment simulations. For the baseline and alignment runs, SimulationMode should be set to "Alignment Simulation." See the section below on Running an Alternative Simulation for instructions on non-alignment simulations.

ParticipationBasedOnPrevRun:

ParticipationBasedOnPrevRun specifies the methodology used for determining whether a SNAP unit that is eligible for benefits chooses to participate. There are four possible participation methodologies: 1) participation is determined by the FoodStamps participation function; 2) units that participated in a previous simulation continue to participate, if still eligible; 3) units eligible for benefits in a previous simulation make the same participation decision as in the previous simulation, if still eligible; and 4) all eligible units participate. For the baseline and alignment runs, ParticipationBasedOnPrevRun should be set to the first option.

If the participation decision is annual and the second or third options are selected, then BenefitsRcvdInPreviousSim and BenefitsEligibleForInPrevSim must be specified. BenefitsRcvdInPreviousSim lists the variable containing the annual SNAP benefits simulated in the previous simulation of interest, and BenefitsEligibleForInPrevSim lists the variable containing the annual SNAP benefits for which the unit was simulated to be eligible. The FoodStamps annual participation function is used to simulate the participation decision for all eligible units (under option 1), for eligible units that were ineligible or did not participate in the previous simulation (under option 2), and for eligible units that were ineligible in the previous simulation (under option 3).

If the participation decision is monthly and the second or third options are selected, then MonthlyBenRcvdInPrevSim and MonthlyBenEligForInPrevSim must be specified. MonthlyBenRcvdInPrevSim lists the variable containing the monthly SNAP benefits simulated in the previous simulation of interest, and MonthlyBenEligForInPrevSim lists the variable containing SNAP benefits for which the unit was simulated to be eligible. The FoodStamps monthly participation function is used to simulate the participation decision for all eligible units (under option 1), for eligible units that were ineligible or did not participate in the previous simulation (under option 2), and for eligible units that were ineligible in the previous simulation (under option 3).

Probabilities of Participation:

These program rules give the probability that a unit eligible for SNAP benefits will choose to participate. Probabilities of participation vary by monthly benefit level and unit type. If the participation decision is annual, the monthly benefit level is the average amount for the months in which the unit is eligible. If the participation decision is monthly, the simulated benefit level for that month is used.

To simulate the participation decision, units are classified into six different "unit types": units receiving TANF/AFDC, units receiving SSI, units consisting of child(ren) and one adult, units consisting of child(ren) and two or more adults, units containing a person aged 60 or more, and all other units. A unit is categorized according to the first "type" that describes it. For example, a unit classified as "receiving TANF/AFDC" might also contain an elderly person. Probabilities of participation for units receiving TANF/AFDC are specified through ProbPartAFDC1to50, ProbPartAFDC51To100, ProbPartAFDC101To150, ProbPartAFDC151To200, ProbPartAFDC201To300, and ProbPartAFDCOver300 (for units with monthly SNAP benefits less than or equal to $50, $51 to $100, $101 to $150, $151 to $200, $201 to $300, and greater than $300, respectively). There is a corresponding set of program rules for units receiving SSI (prefixed ProbPartSSI), units with child(ren) and one adult (prefixed ProbPart1Parent), units with child(ren) and two or more adults (prefixed ProbPart2Parent), units containing a person aged 60 or more (prefixed ProbPartElderly), and other units (prefixed ProbPartOther).

Probabilities of Correct Response:

These program rules, prefixed "ProbCorr" correspond to the same benefit levels and unit types as the program rules holding the probabilities of participation. The probability of correct response is the probability that a unit that is eligible for SNAP benefits under current law but does not report receiving SNAP benefits will be "believed" to be reporting correctly, and therefore not simulated to receive SNAP benefits in the baseline simulation.

Adjustments:

Two multiplicative state-level adjustment factors are used in the alignment process. These factors are applied to all units in a particular state, regardless of unit type or benefit level. ParticipationStateAdjustment adjusts the probability of participation and CorrectResponseStateAdjustment adjusts the probability of correct response. Additional multiplicative adjustment factors are used to help align to citizenship targets. PartAdjCitizenChildUnits, PartAdjNonCitizenChildUnits, and PartAdjNoChildrenUnits adjust the probability of participation for units with citizen children, units with only non-citizen children, and units with no children. CorrAdjCitizenChildUnits, CorrAdjNonCitizenChildUnits, and CorrAdjNoChildrenUnits adjust the probability of correct response. Each of these factors varies by state of residence (California versus balance of the nation) and the citizenship status of the head and spouse of the household (illegal alien, legal immigrant/refugee, naturalized citizen, and native citizen). The unit is placed into the first citizenship category that describes the citizenship status of either the head or the spouse.

State and citizenship status adjustments to probabilities of participation and correct response do not vary during the year. If a monthly participation decision is being simulated, the state and citizenship status adjustments can be thought of as fixed multiplicative adjustments that are applied to each month's probabilities.

Baseline Targets:

We try to hit three sets of targets in the SNAP alignment process. We first align to national targets, by unit type and benefit level. These national targets are not entered into TRIM3. Comparison to the national targets is performed outside of TRIM3 during the alignment process. The second set of targets consists of TargetNumberOfUnits, TargetNumberOfPersons, and TargetDollarAmount. These show the number of units and persons participating in the average month and total annual benefits, by state. The third set of targets is available for aligning by citizenship status. TargetCitizenChildUnits, TargetNonCitizenChildUnits, and TargetNoChildrenUnits correspond to the adjustment factors prefixed with PartAdj… and CorrAdj….

NonReporterAdjustment form:

The form NonReporterAdjustment provides the user with an additional adjustment to help in those last stages of the alignment when there are a few problem areas that cannot be made better without making something equally important substantially worse.

This adjustment takes place in a baseline simulation. It is made as one final adjustment after the participation decision has been made. It only affects eligible non-reporters, and it makes use of information input by the user through this simple form. The form indicates what percentage of particular types of eligible non-reporters should be changed from participants to non-participants, or from non-participants to participants. The form can be changed from year to year as necessary to target different problem areas in each year.

Let’s say we want 5 percent of eligible non-reporter units who live in California, are not currently participating, and are in a household with an illegal alien head or spouse to be made into participants. In addition, we want 10 percent of eligible non-reporter units who live in a state other than California, are currently participating, and are in a household with an illegal alien head or spouse to be made into nonparticipants. The NonreporterAdjustment form would look something like this:

Enter 0

If Part_HhHdAndSpCitizenship=1 then return .05

If Part_HhHdAndSpCitizenship=5 then return -.10