Revised: December 7, 2000


View PDF version

If you cannot access the file through the link, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

Under USDA guidance and proposed USDA regulations, states have the opportunity to wipe clean their "clocks" on the food stamp three-month cut-off. In states electing this approach, affected individuals will qualify for a new three months of benefits without regard to any food stamps they may have received since the three-month cut-off provision first took effect. (Individuals who become employed for a month or comply for a month with workfare or certain other employment and training programs after this new three months of eligibility ends will qualify at that time for an additional three months of benefits.)

Exercising this option is likely to simplify program administration and reduce states' quality control (QC) error rates. Although several states have taken this option, others may not be aware of its availability.

Background. The welfare law disqualifies certain childless unemployed persons if, during periods when they were not exempt or covered by a waiver, they received food stamps for "not less than three months" within a "36-month period." The statute does not define which "36-month period" is to be used. USDA's guidance and its proposed rules effectively give states three options:(1)

A "rolling clock." Under a rolling clock, a state must analyze each individual's food stamp eligibility for a given month based on his or her receipt of benefits during the 36 months immediately preceding that month. This requires a new calculation for each month for which the person seeks food stamps. If an eligibility worker errs in this calculation, or if an earlier eligibility worker failed properly to record an individual's non-exempt status for a prior month, the state may be liable for a QC error.

An "individualized fixed clock." This method applies a separate 36-month clock to each recipient, with the first month of the 36-month clock being the first month of food stamps the individual receives that counts toward the time limit. Each person becomes eligible for a new three months of benefits on the three-year anniversary of the first month in which he or she first received time-limited food stamps, regardless of his or her benefit history during that 36-month period. Thus, individuals who were receiving food stamps when the time limit took effect in December 1996 could be given a new three months as early as December 1, 1999, but those who did not enter the program until later would have to wait until a later date to have their eligibility restored. Here again, eligibility workers would be required to look backwards up to three years to determine the starting date for each individual's "clock," and flaws in the records made in the early months of the time limit could result in subsequent QC errors (or in improper denials of benefits to individuals who still have months remaining on their "clocks").

A "statewide fixed clock." The administratively simplest approach, and the one that is most helpful for recipients, is to determine everyone's eligibility with reference to the same 36-month period. Thus, each individual could receive three months that count against the time limit between December 1, 1996, and November 30, 1999, and another three months in the next 36-month period, which a state could decide to start any time after December 1, 1999. It would not matter when in the first 36-month period the individual received food stamps; he or she would automatically qualify for a new three months of benefits on December 1, 1999. If a state has been using a rolling clock or an individualized fixed clock, it can convert to a statewide fixed clock at any time. Although, USDA encourages states to adopt this option around December 1999, states continue to be free to adopt a fixed clock.

Many individuals subject to the three-month cut-off will qualify for a new three months of benefits sooner under a "statewide fixed clock" than under either of the other systems; no individual would be made worse off. The diagram below and accompanying text provide a simple example of how an individual could qualify for three months of food stamps under a statewide fixed clock but be denied benefits under either of the other "clocks."

States' administrative burdens and risk of QC errors also would be reduced under a statewide fixed clock, because an individual's benefit history during the period prior to the beginning of the second 36-month clock would become irrelevant to his or her eligibility once the state restarted its clock. For example, months before February 2000 would become irrelevant in a state starting a statewide fixed clock on February 1, 2000. In such a state, any mistakes made in recording benefit receipt or non-exempt status prior to February 1, 2000 could not cause errors after that date. Moreover, since everyone in the state would have a new three months of benefits, the state could not be charged with any QC errors related to the three-month cut-off provision during the first three months after the state established a new statewide fixed clock (e.g., the state could not be charged with any errors under the time limit provision for February, March or April 2000 if it started a new statewide fixed clock on February 1). States that have already acted to institute a statewide fixed clock include Georgia, Illinois, Massachusetts, South Dakota, Tennessee and Washington state.

On September 1, 1999, USDA's National Office issued guidance to its regional food stamp directors that for the first time explains these three options in some detail and urges its regional staffs to raise these issues with states and encourage them to reconsider their polices in this area. On December 17, 1999, USDA published proposed rules under which the "three-year period may be measured and tracked as the State agency deems appropriate;" the preamble to these rules contains a discussion of the possible benefits of a statewide fixed clock. The guidance and proposed rules should put to rest the fears a number of states have had about whether USDA was genuinely willing to permit statewide fixed clocks. The guidance clearly encourages states to consider this option, which could benefit both states (through reduced administrative burdens and errors) and recipients.

Example of the Impact of a Statewide Fixed Clock. The benefits of a statewide fixed clock can be seen in a simple example. Consider the case of a childless individual who was employed and not receiving food stamps when the three-month cut-off took effect in December 1996 and who continued to work without receiving food stamps until being laid off more than two years later in April 1999. If she applied on May 1, 1999, she would exhaust her three months of eligibility at the end of July 1999.

Under either a rolling clock or an individualized fixed clock, she would not be eligible again until May 2002 - some 33 months after exhausting her initial three months of eligibility - unless she requalified by working at least half-time for a month. (Even if she did requalify by working for a month, she could receive only an additional three months of food stamps after she again lost her job and then could not get food stamps again until May 2002 unless she became exempt.)

If, on the other hand, her state established a new statewide fixed clock beginning on December 1, 1999, she would become eligible for a new three months of food stamps on that date. Thus, if she reapplied on July 1, 2000, she could receive food stamps for July, August and September even though less than a year had elapsed since she last received benefits. (In addition, if her eligibility worker failed to record that the food stamps she received from May through July of 1999 counted against the three-month time limit, that mistake could no longer lead to a QC error after December 1, 1999.)

Example of Permissible Spells of Participation Using a Statewide Fixed Clock
1996 Dec. 1996
1997 Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. 1997
1998 Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. 1998
1999 Jan. Feb. March April May June July Aug. Sept. Oct. Nov.   1999
1999   Dec. 1999
2000 Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. 2000
2001 Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. 2001
2002 Jan. Feb. March April May June July Aug. Sept. Oct. Nov.   2002
2002   Dec. 2002
2003 Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. 2003

SUBJECT: Tracking the 36-month period for Able Bodied Adults without Dependents

TO: All Regional Directors

Food Stamp Program

As you know, Section 6(o) of the Food Stamp Act of 1977 (the Act), as amended by Section 824 of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, limits all able-bodied adults without dependents (ABAWDs) age 18 through 50 to 3 months of participation in the Food Stamp Program during any 36-month period unless they are employed or are participating in one of the work programs specified in the Act. When this provision was due to be implemented in November 1996, we gave State agencies maximum flexibility to measure and track these two time periods. State agencies made their choices and designed their systems accordingly. It has recently come to our attention that many State agencies would like to change the way they track the 36-month period. This memo therefore, reiterates the basic concept of the 36-month period, restates the ways in which State agencies may track the 36-month period, and confirms that State agencies may change the way they track it.

The 36-month period is a background against which an ABAWD's 3-months of participation are measured. It is a continuous period, it does not start and stop. There are two basic ways a State agency could measure or track the 36-month period: as a "fixed" or a "rolling" period. A fixed period has a definite start and stop date; it starts on a given date, runs continuously for 3-years, stops exactly 3-years later, and then a new fixed 3-year period starts. Under a fixed period approach, when a new 36-month period starts, a participant's slate is "wiped clean," and he or she can be eligible for another 3-months of participation without fulfilling the work requirement. A rolling period does not have definite start or stop dates. Using a rolling period, the eligibility worker always "looks back" 3-years from the date of application and keeps looking back 3-years each ensuing month.

There are two ways to use a fixed period - for the individual or for the entire caseload. Most State agencies decided to use a fixed 36-month period for each individual, starting the "clock" either at application or when the recipient's exemption was up. Some State agencies decided to use a fixed 36-month period for the entire caseload, starting the clock on December 1, 1996, and ending it on November 30, 1999.

Some State agencies have expressed interest in switching from a rolling period or an individually based 36-month period to a fixed 36-month period for the entire caseload. State agencies may do this. Since most State agencies implemented the provision starting December 1, 1996, the first fixed 36-month period for the entire caseload would have begun on that date and would be scheduled to end on November 30, 1999. The second 36-month period would begin on December 1, 1999. Therefore, if State agencies are considering switching to this method of tracking the 36-month period, December 1, 1999, is the most opportune time to do so as it will have little or no impact on either the State agency or the recipient.

We urge you to discuss this issue with your State agencies and work with them if need be to determine the most appropriate way to track this time period.

Arthur T. Foley


Program Development Division

FNS:FSP:PDD:CPB:MoiraJohnston:DRAFT 9/1/99:final:9/1/99:va

DIR: I:CPB:ABAWD:all region -clock


1. FNS, USDA, Tracking the 36-Month Period for Able Bodied Adults Without Dependents (Sept. 1, 1999) (attached); 64 Fed. Reg. 70920, 70939-40 (Dec. 17, 1999).