Questions and Answers: Food Stamp Program High Performance Bonuses Performance Measures
Q1 - How will the money be divided up?
A1 - The money will be divided up among State agencies in proportion to the size of their
caseloads. For example, if 6 states are to split $6 million and State A accounts for 40 percent of all
food stamp participants in these 6 states, State A will receive 40% of $6 million, or $2.4 million.
We believe this is the most equitable way to divide the money up. A State with a large participant caseload
that improves its payment accuracy rate by 4 percentage points has a greater national impact than a State
with a small participant caseload that improves its payment accuracy rate by 4 percent. All States, large
and small, will have sufficient incentive to compete for the bonuses if the money is divided proportionally.
Q2 - How will the most improved payment error rate be determined, by percent increase
or by percentage point increase?
A2 - We will determine the most improved payment error rate by measuring the percentage
points improved. For example, if State A has a payment error rate of 10% in 2003 and a payment error rate
of 6% in 2004, its improvement is 4 percentage points, or a 40 percent improvement. If State B has a payment
error rate of 6% in 2003 and a payment error rate of 4% in 2004, its improvement is 2 percentage points,
or a 50 percent improvement. We would rank State A higher than State B because its absolute improvement
is greater even though its relative improvement is less. We believe absolute improvement has more of an
impact on the national program than relative improvement. For example, if States A and B both issued $100
million in benefits, State A would have reduced its payment error by $4 million while State B would have
reduced by only $2 million.
Q3 - Will a State agency that has improved its payment accuracy rate by the most percentage
points, but whose payment accuracy rate is still above the national average be able to win a bonus?
A3 - Yes, for fiscal year (FY) 2003, a State agency that improves its payment accuracy
rate by the most percentage points will be able to win a bonus even if its payment accuracy rate is above
the national average. For further explanation, see question and answer number
Q4 - How will the most improved negative error rate be determined, by percent decrease
or by percentage point decrease?
A4 - We will determine the most improved negative error rate by measuring the percentage
points improved. For example, if State A has a negative error rate of 3% in 2003 and a negative error
rate of 1% in 2004, its improvement is 2 percentage points, or 66 percent. If State B has a negative error
rate of 10% in 2003 and 5% in 2004, its improvement is 5 percentage points, or 50 percent. We would rank
State B higher than State A because its absolute improvement is greater, even though its relative improvement
is less. We believe absolute improvement has a greater impact than relative improvement.
Q5 - What data will we use to calculate the participant access rate? Will we make adjustments
for special situations, such as the Supplemental Security Income (SSI) population in California and the
individuals residing on reservations that choose to receive commodities?
A5 - We will use a variety of data sources to calculate the participant access rate.
The denominator is composed of data from the Census Bureau’s Current Population Survey. We will
use annual state counts of persons in poverty from the Census Bureau shortly after it is released, usually
in late September. These counts are based on income received in the previous calendar year. Because persons
receiving SSI are ineligible for food stamps in California, we will reduce the number of poor persons
in California by the percentage of poor who received SSI in the previous year. This data is collected
from the Census Bureau as well.
For the number of Food Stamp participants, we will use administrative counts of participants by State
over the same calendar year for the Census Bureau's persons in poverty, averaging 12 months of data. Because some individuals residing
on reservations may choose to receive food assistance from either the Food Stamp Program or the Food Distribution
Program on Indian Reservations (FDPIR) but not both simultaneously, we will add to the Food Stamp participants
the number of FDPIR participants using administrative data over a calendar year, averaging 12 months of
data. The sum of these two numbers is the numerator. The adjusted numerator divided by the adjusted denominator
produces the participant access rate.
Q6 - What is the Federal application-processing standard that we will use to measure
performance under the application processing timeliness measure?
A6 - We will use the application-processing standard of 30 days. An applicant must be
given the “opportunity to participate” (as defined in 274.2) within thirty days (or 7 days
for expedited service). New applications that are processed outside this standard will be considered untimely
for this measure. We recognize that the regulations at 7 CFR 273.2 provide procedures for State agencies
that, for one reason or another, are unable to meet the 30-day standard. Some may argue that we should
measure compliance with these regulations rather than performance under a 30-day standard. However, we
believe this measure is a test of performance not compliance, and of customer service regardless of who
is at fault for the delay. We would not want to reward a State that is weak in meeting the 30-day standard
but good at getting benefits out within 60 days. Furthermore, it would be difficult, based upon certification
records, to consistently distinguish between delays that are applicant vs. agency caused.
Q7 - Will both approvals and denials be included in the determination of timeliness?
A7 - No, this measure is focused on meeting the 30-day standard for providing eligiblehouseholds
the opportunity to participate.
Q8 - Are there revised quality control (QC) requirements for application processing timeliness?
A8 - Yes, the revised QC instructions are in the 310 handbook as item 71B. However, the
final handbook has not been published yet nor put on the web page. We anticipate this happening soon and
will let the Regional offices know as soon as it does.
Q9 - Since QC pulls a case for a specific review month, will the Food and Nutrition Service
only look at timeliness of applications if the review month is the month the application was processed?
Or will every case pulled be evaluated to see if an application was filed in FY 2003, and those applications
will be evaluated for timeliness?
A9 - We will look at new applications that were filed on or after the beginning of FY
2003 because they were filed within the performance measurement year for which the bonuses are awarded.
Q10 - Can a State agency win more than one bonus, i.e., the best and the most improved?
A10 - No, a State cannot be awarded two bonuses in the same category, i.e., the best
and most improved participant access rate. If a State is among the most improved in a category, it would
not be counted among the best. This allows the “next best” State to receive an award as being
among the best States. A State may be awarded bonuses for different categories, such as most improved
negative error rate and highest participant access rate.
Q11 - Can a State agency that is assessed a liability receive a bonus?
A11 - For FY 2003, no State agency will be assessed a liability. The farm bill provides
that no State may receive a bonus that has been assessed a quality control liability. Since a State must
have a payment error rate above the national average for two years in a row in order to be assessed a
liability, and FY 03 is the first year, no State will be assessed a liability in FY 03. Therefore, all
States could be eligible to receive a bonus in FY 2003. The prohibition against awarding bonuses to States
that have been assessed a liability will take effect for FY 04 regardless of whether that liability is