TANF-ACF-PI-1998-01 (Formula for Awarding the First High Performance Bonus in Fiscal Year (FY) 1999)

To:

State Agencies Administering the Temporary Assistance for Needy Families (TANF) Program and Other Interested Parties

Subject:

Formula for Awarding the First High Performance Bonus in Fiscal Year (FY) 1999

References:

Section 403(a)(4) of the Social Security Act

Purpose:

The purpose of this transmittal is to advise States and other interested parties of the measures and formula the Department of Health and Human Services (DHHS) intends to use for awarding High Performance bonuses for FY 1998.  We are also submitting a clearance package to the Office of Management and Budget that addresses the data collection necessary to award these bonuses.

This transmittal addresses only the first-year bonus.  We are currently drafting a notice of proposed rulemaking (NPRM) that will address the formula applicable for future years.

Background:

Statutory Context

Section 403(a)(4) of the Social Security Act makes $1 billion available over a five-year period to reward States that achieve high performance levels under the new welfare block grant program, known as TANF.  It provides that the first award will be made in FY 1999 based on performance above a threshold level in FY 1998.  It calls for the Secretary -- in consultation with the National Governors' Association (NGA) and the American Public Welfare Association (APWA) -- to develop a formula within one year of enactment (or by August 22, 1997).  The formula should reflect performance in achieving the goals of the TANF program.  The total award to a State cannot exceed 5 percent of its TANF grant, known as the State Family Assistance Grant (or SFAG).

DHHS Efforts to Date

Following enactment of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, the Administration for Children and Families (ACF) developed a strategy for broad consultation with the States and other interested parties to inform its development of proposed rules on major policy issues.  General consultation on the TANF regulations began in the fall of 1996.

Consistent with the specific statutory language on consultation regarding the High Performance Bonus, we developed a process for ongoing consultation with the NGA, APWA, National Conference of State Legislatures (NCSL), and State representatives who agreed to participate in a work group.  From February through July 1997, we scheduled biweekly discussions with this work group covering the principles underlying a performance system, the viability of individual measures and data options, and the general allocation and distribution rules.

In the interest of moving the process forward, in July we shared a "preliminary proposal" with our State partners and other interested parties, which also included advocates and technical and policy experts.  This proposal provided the framework for additional, more focused consultations on the possible elements of the bonus formula.  At these consultations we received wide-ranging and very helpful feedback on the draft proposal.

NGA and APWA also provided feedback in the form of policy statements presenting their views on how the formula should be drafted.  There were several important similarities between these policy statements and ACF's draft proposal--including similar principles and goals for the bonus system and a similar focus on work measures--but there were also differences.

Based on these consultations, we incorporated a number of changes to our initial proposal in this first-year formula.

Need for Interim Measures

States need to know how we will be assessing their performance this year.  We have a responsibility to give them this information as soon as possible.

We want to ensure that States have the capacity to provide the information needed to qualify for these bonuses. Based on our discussions with States and the analysis we have done, we believe this capacity exists.  The notice now will increase the chance that a particular State will be able to make any necessary adaptations to correct any data shortcomings and compete.

Although the statute does not require it, we would strongly prefer to set the formula through rulemaking.  However, the time frames involved in issuing proposed and final rules are not compatible with providing timely notice to States of the policies that will apply during the first bonus year.  Thus, we have decided to issue this policy announcement governing the determination of bonuses for this first bonus year.

In the absence of regulations, and in taking into consideration the input we received during our consultations, we will use a simplified and streamlined formula.  The measures for the first year focus on work and are closely tied to the statutory goals.

We will give States some flexibility on how they collect the data on each measure.

Substance of Interim Measures

We intend to award $200 million in High Performance Bonus funds to States in FY 1999.  We believe that providing a full share of the bonus funds during the first year will give important impetus to State implementation efforts, particularly in the welfare-to-work area.  We understand there are limitations in the first year formula and data, but we do not believe these are sufficient to warrant deferring the award of the bonus monies.

We will award funds to States on the basis of their rankings on each of four work-related measures.  Two of these measures will address absolute State performance, and two will address State performance improvement. The four measures are:

Job Entry Rate

Success in the Work Force Measure:

a) Job Retention Rate

b) Earnings Gain Rate

Increase in Job Entry Rate

Increase in Success in the Work Force Measure:

a) Increase in Job Retention Rate

b) Increase in Earnings Gain Rate

(Note that the "success in the work force" measure is a combination of two work-related measures.)

In different ways, all four measures reflect a State's success in moving families from welfare to work.  Full success requires not only getting recipients into jobs, but also keeping them in jobs, and increasing earnings in order to reduce dependency and enable families to support themselves.  Our measures address all these aspects of success.

The multiple measures also support State flexibility in designing their TANF programs.  They enable States that have different work philosophies to fairly compete for bonuses.  Compared to a single measure, they are less likely to distort State policy decisions or to cause unintended negative consequences.  (The TANF work participation rates also serve to counterbalance the potential effect of the individual measures by encouraging States to be expansive in their efforts to move recipients from welfare to work and discouraging creaming.)

In the case of the first two measures, we will look at State rankings on absolute performance in FY 1998; for the latter two measures, we will rank States based on the extent to which their performance in FY 1998 exceeded their performance on that same measure in FY 1997.  By including improvement measures, we intend to:

1. provide a mechanism by which all States can compete;
2. motivate all States to improve their performance; and
3. moderate problems associated with uneven playing fields.

For this first year, we will make the ten States with the best scores on each measure eligible for bonus monies.  Sixty-five percent, or $130 million, of the $200 million will be available to the ten best States on each State performance measure.  The remaining 35 percent, or $70 million of the award money, will be available to the ten best States on each State improvement measure.

The following percentages and amounts have been allocated to each measure.

As previously mentioned, the law requires that no State may receive a high performance bonus greater than 5 percent of its SFAG. If the 5 percent cap on an individual State's award prevents the full allocation of these individual award pools, we plan to re-allocate the extra funds.  For all four measures, we plan to look at performance over the course of an entire year.  We made this decision, in large part, to avoid any potential bias in work data due to seasonal differences in employment.

In all cases, we will compute State performance only with respect to adults who were receiving assistance [as defined in the policy guidance we issued January 31, 1997 (TANF-ACF-PA-97-1)].  In order to compete for the High Performance Bonus, States must report comparable data for adults receiving assistance in separate State programs for FY 1997 and FY 1998.  This means reporting the data requirements in Sections 1, 2 and 3 of the Emergency TANF Data Report (Form ACF-198) for the separate State programs beginning with the month the State is subject to the TANF reporting requirements, as well as reporting the seven HPB data elements in Section D of the Appendix for adults in separate State programs.

As indicated in the January 1997 guidance, we are requesting data on separate State programs to assess States overall performance in moving recipients from welfare to work.  The success in meeting TANF performance goals may be affected by State choice in funding separate State programs with Maintenance of Effort (MOE) dollars, thus advantaging one State over another.  For example, if a State had a separate State assistance program similar to TANF in which it put more difficult to employ recipients, its TANF performance results could be unfairly inflated.  In such cases, we would need to consider including outcomes for the caseload in separate State programs in the performance measures.

We will analyze separate State program data required above, as well as other information we receive on the characteristics of the caseload and the nature of benefits provided in separate State programs, in assessing how and whether to adjust a State's TANF performance data.

We are willing to consider arguments by States that they should not be required to report all the data outlined above for certain specific separate State programs because:

1. the scope and nature of the specific separate State program will not unfairly inflate performance under TANF or the High Performance Bonus; and
2. the State is incapable of meeting the reporting requirements due ta the nature of the benefits.

Under such limited circumstances, we may conclude that a State may report a more limited subset of the required Data on that separate State program and still compete for the High Performance Bonus.  States that wish to be considered for this limited exception will need to provide detailed information about the nature and scope of their separate State programs for our consideration, along with a detailed explanation of what data cannot be provided and why.  We will determine whether the exception is appropriate and, if so, what data must still be provided.

The Appendix to this transmittal includes five Sections: Section A - a description of the formula we will use in calculating the FY 1999 High Performance Bonus; Section B - how the best 10 States will be determined for each measure; Section C - how the bonus monies will be calculated for the ten best States on each measure; Section D - the specific data collection requirements; and Section E - the reporting dates.

We will shortly publish an emergency form clearance for comment and may make technical changes to the content of Sections D and E based on comments we receive.

Flexibility in Data Sources

For this first year, we will allow States to submit the best data they have available on each measure.  Thus, they may use matches with quarterly Unemployment Insurance (UI) data, surveys, administrative records or a combination of these data sources.

We will be requiring States to report quarterly data on a quarterly basis.  We believe quarterly reporting is necessary in order to provide us with sufficient feedback as to how well the system is working and whether States are on track to report the data and compete this year.

Because of the lags associated with using UI subsequent quarterly periods to measure job retention and earnings gain of employed adult recipients, we will not ask States to report the new performance data in the same time-frames as specified for other TANF program and financial data.  In essence, they will have additional time to report their performance data to us.

Along with aggregate reporting on the measures, we will require each State to:

1. specify the data sources and methodology it has used to determine the aggregate numbers;
2. include a certification that it has provided the best data available on the measures in the formula;
3. maintain records that adequately document the derivation of its performance data; and
4. provide access to such records, for validation purposes, upon request.

If a State opts to use sampling to comply with the reporting requirements, it must use an acceptable sampling methodology and sufficiently large samples to make estimates over various subpopulations, e.g., the earnings gain rate of employed recipients in the second subsequent quarter.  The samples must be representative of the entire calendar quarter; it is not appropriate to use only one month to represent activity for the quarter.

A State need not submit data if it does not want to compete on any particular measure.  However, as indicated previously, to the extent that a state wants to compete on a given measure, it will need to submit comparable performance and caseload characteristics data for separate State programs in order to qualify for a High Performance Bonus, as described above.

Future Bonuses

As mentioned earlier, the formula and data collection presented in this transmittal apply only to the performance awards made in FY 1999.  We are committed to issuance of an NPRM and final rule that will cover the High Performance Bonuses to be awarded in future years, i.e., fiscal years 2000 through 2003.  The schedule for this rulemaking provides for NPRM publication in the spring and final rules in early FY 1999.

As we work towards publication of proposed and final rules, we will be looking at possible refinements to the performance measures used this first year.  The rulemaking process will allow ample opportunity for all interested parties to comment on the formula we will implement for future bonuses.

In this further work we remain committed to our process of consultation and collaboration with the NGA, APWA, NCSL, individual States, and other organizations and individual experts.  We also remain committed to developing a performance system that is simple to understand and administer, has integrity, and fosters the goals of the TANF program.

Effective Date: 

Immediately

Attachment:

Appendix - High Performance Bonus Award

Inquiries:

For further information, you should contact the appropriate Regional Administrator.

Diann Dawson
Director
Office of Family Assistance

/s/

Howard Rolston
Director
Office of Planning, Research and Evaluation

Appendix

High Performance Bonus Award

Section A.  Bonus Formula

We will use the FY 1997 and FY 1998 data reported by the States (see Section D) in the calculations for the FY 1999 High Performance Bonus Awards.

1.  Performance Measures

1. FY 1998 Job Entry Rate

This is the unduplicated number of adult recipients who entered employment for the first time in FY 1998 (job entries) as a percent of the total unduplicated number of FY 1998 recipient adults unemployed for the first time in FY 1998.  Adult recipients participating in workfare or in fully subsidized employment are not included in the numerator but are included in the denominator.

Unduplicated number of job entries
-------------------------------------------------- x 100
 

Unduplicated number of unemployed adult recipients

(b) Success in the Work Force Measure (a composite measurement of the following two rates)

(1) FY 1998 Job Retention Rate

This is the FY 1998 average of the sum of the unduplicated number of employed adult recipients in one quarter who were also employed in the first subsequent quarter, as a percent of the sum of the unduplicated number of employed adult recipients in each quarter.  (At this point, they might be former recipients).  Adult recipients participating in workfare or in fully subsidized employment are not included in either the numerator or the denominator.

Sum of employed adult recipients in each 
of Qtrs 1 thru 4 who continue employment
in each subsequent Qtrs
---------------------------------------- x 100
Sum of employed adult recipients in Qtrs 1 thru 4

(2) FY 1998 Earnings Gain Rate

This is the sum of the gain in earnings between the initial and second subsequent quarter in each of quarters 1 through 4 of FY 1998 adult recipients employed in both these quarters as a percent of the sum of their initial earnings in each of quarters 1 through 4.  (At this point, they might be former recipients.)  Earnings gain of adult recipients participating in workfare or in fully subsidized employment are not included in either the numerator or the denominator.

Sum of total dollars earned in second
subsequent quarter in each of Qtrs 1 thru 4 
minus sum of total dollars earned in initial
quarter in each of Qtrs 1 thru 4 of adult
recipients employed in both these quarters
------------------------------------------------- x 100
Sum of total dollars earned in initial
quarter in each of Qtrs 1 thru 4

1.  Improvement Measures

(a) Increase in FY 1998 Job Entry Rate

This is the percentage change between FY 1997 and FY 1998 job entry rates.

FY 1998 job entry rate - FY 1997 job entry rate
-------------------------------------------------- x 100
FY 1997 job entry rate

(b) Increase in the Success in the Work Force Measure

(1) Increase in FY 1998 Job Retention Rate

This is the percentage change between FY.1997.and FY 1998 job retention rates.

FY 1998 retention rate - FY 1997 retention rate 
-------------------------------------------------------- x 100
FY 1997 retention rate

(2) Increase in FY 1998 in Earnings Gain Rate

This is the percentage change between FY 1997 and FY 1998 in the earnings gain rates.

FY 1998 earnings gain rate - FY 1997
earnings gain rate
-------------------------------------- x 100
FY 1997 Earnings gain rate

Section B.  How the Best States for HPB Money on Each Measure will be Determined

High Performance Bonus monies will be awarded to States on the basis of their rankings in two absolute State performance measures (job entry rates and the success in the work force measure), and in two State improvement measures (increase in the job entry rates and increase in the success in the work force measure).

Job Entry Measure

The rates for States submitting data for this measure will be ranked from high to low with "1" being the rank for the State with the best score.  States not submitting data for this measure will each be assigned a rank which is the number following the last rank for States that submitted data.  States with rankings 1-10 will share in the bonus dollars allocated to this performance measure.

Success in the Work Force Measure

This is a composite weighted score of the job retention and the earnings gain measures.  A separate ranking of each of the two rates will be made for the States that submit the required data.  Each State that does not submit data for one of these measures will be assigned a rank which is the number following the last rank for States that submitted data for that particular measure.

Because it is a more familiar measure with a more substantial history, we will weight each State's job retention ranking score by 2.  The weighted job retention score will then be combined with the earnings gain score for each State.  The sum of these two scores will then be ranked from low to high with "1" being the rank for the State with the best score.  States ranked 1-10 will share in the bonus dollars assigned to this performance measure.  (See example below.)

The same procedure will be followed for the two improvement measures, i.e., increases in the job entry and in the success in the work force measures.

Example:

Assume that

• 49 States submitted data for the Job Entry and Job Retention rates.

• 42 States submitted data for the Earnings Gain rate. (States that did not submit data, e.g., State C, will be assigned "43" as their rank.)

• State Job Retention Ranks have a weight of "2"

Job Entry Measure

(States ranked from 1 to 49; States with ranks 1-10 receive bonus money.)

Success in the Work Force Measure

(States with the lowest 10 ranks, after ranking the combined weighted scores in item (e) for all States, will receive bonus money for the Success in the Work Force measure.)

Section C.  How the FY 1999 HPB Dollars will be Calculated for the 10 Best States in Each Measure

The 10 States with the best ranked scores on each of the following measures will be eligible for bonus monies.  For each measure, the percent and total amount of the available $200 million award money will be as follows:

Bonus dollars for each measure will be distributed based on each State's percentage of the total SFAG of the 10 States that will receive a bonus.  For example, if the total SFAG of the 10 best States for the Job Entry measure is $1,737 million, and if State A's SFAG of $229 million is 13.2% of all the awarded States' SFAG, then State A will receive 13.2% of the $80 million, or approximately $10,560,000 of the bonus money.  However, if State A is also eligible for bonus money for any of the other measures, the total amount of the award monies State A receives cannot exceed $11,450,000 (5% of the SFAG amount of $229 million).

Section D.  HPB Data Requirements [Reporting requirements will become effective after clearance by the Office of Management and Budget.]

In order to compete for the High Performance Bonus, States must report comparable data for adults receiving assistance in separate State programs for FY 1997 and FY 1998.  This means reporting the data required in Sections 1, 2, and 3 of the Emergency TANF Data Report (Form ACF-198) for the separate State programs beginning with the month the State is subject to the TANF reporting requirements, as well as reporting-the seven HPB data elements specified in this section for adults in separate State programs.  (Data for FY 1997 are necessary in order to measure the extent to which the FY 1998 performance of each measure exceeded its FY 1997.)

Separate information by calendar quarter must be submitted for the TANF and the TANF-MOE programs for FY 1998.  (Comparable information is also required for FY 1997.)  

(1) Cumulative unduplicated number of unemployed adults receiving-assistance in FY 1998.  (For example, the Qtr 1 report will show the unduplicated count only in Qtr 1; the Qtr 2 report will show the unduplicated count in the first two quarters; Qtr 3 will show the unduplicated count for first 3 quarters; and Qtr 4 will show the total unduplicated count for all 4 quarters.)

(2) Total unduplicated number of adult recipients entering employment for the first time in FY 1998.

(3) Total unduplicated number of adult recipients employed in each quarter.

(4) Total number of adult recipients reported in item (2) who were also employed in the subsequent quarter.

(5) Total number of adult employed recipients reported in item

(3) who were also employed in the second subsequent quarter.

(6) Total earnings in the initial quarter of adult recipients reported in item (5).

(7) Total earnings in the second subsequent quarter of adult recipients in item (5).

Examples:

Schedule E.  HPB Reporting Dates

Data will be reported on a quarterly basis.  Each quarterly report will contain the Emergency TANF Data Report for the separate State programs and the tables of performance data covering

TANF and TANF-MOE data for the FY 1997 and FY 1998 specific quarter.

A report form and instructions will be sent to the States after clearance by the Office of Management and Budget.

OFA Homepage

HHS Homepage

Search From Here

Page Posted: 9/4/98.  Questions?  Visit the Office of Family Assistance Home Page or e-mail us