UI Benefit Accuracy Measurement Annual Report Overpayment Rate in Detail
Annual Report Overpayment Rate In Detail
The estimated BAM Annual Report overpayment rates since the beginning of the program in calendar year (CY) 1988 are summarized in the following chart.
A subset of the Annual Report rate is the operational overpayment rate, which includes those overpayments that the states are reasonably expected to detect and establish for recovery -- fraud and nonfraud recoverable overpayments, excluding work search, employment service (ES) registration, base period wage issues and miscellaneous causes, such as benefits paid during a period of disqualification, redeterminations, and back pay awards. The following chart plots the BAM Annual Report rate and the operational rate for the last five years. Because the operational rate is more narrowly focused than the Annual Report rate, it is less likely to fluctuate from year to year.
State Annual Report Overpayment Rates
Readers are strongly cautioned that it may be misleading to compare one state's payment accuracy rates with another state’s rates. No two states' written laws, regulations, and policies specifying eligibility conditions are identical, and differences in these conditions influence the potential for error. States with stringent or complex provisions tend to have higher improper payment rates than those with simpler, more straightforward provisions.
| State | CY 2004 Overpayment Rate |
Change From CY 2003 (Percentage Pts.) |
Leading Cause (Percent Of UI Benefits Paid) |
|---|---|---|---|
| Kansas | 32.82% | +4.87 | ES Registration (28.10%) |
| Arizona | 30.58% | +8.63 | Work Search (17.58%) |
| Virginia | 26.86% | +6.27 | Work Search (13.49%) |
| Louisiana | 19.64% | +3.86 | Benefit Year Earnings (5.81%) |
| Ohio | 18.41% | +7.30 | ES Registration (5.80%) |
| State | CY 2004 Overpayment Rate |
Change From CY 2003 (Percentage Pts.) |
Leading Cause (Percent Of UI Benefits Paid) |
|---|---|---|---|
| Vermont | 2.84% | -0.03 | Benefit Year Earnings (1.62%) |
| Oklahoma | 3.01% | +1.21 | Separation Issues (1.31%) |
| Hawai'i | 3.77% | -1.41 | Benefit Year Earnings (1.16%) |
| West Virginia | 4.57% | +2.56 | Benefit Year Earnings (2.22%) |
| Tennessee | 4.67% | -3.91 | Benefit Year Earnings (3.23%) |
| State | Amount of UI Benefits Paid (In Millions) |
CY 2004 OP Rate (Percentage Pt. Change From CY 2003) |
State | Amount of UI Benefits Paid (In Millions) |
CY 2004 OP Rate (Percentage Pt. Change From CY 2003) |
|---|---|---|---|---|---|
| California | $5162.2 | 6.16% (-0.64) | Michigan | $1,781.8 | 9.53% (+0.36) |
| New York | $2,558.5 | 8.94% (+1.10) | Texas | $1,665.2 | 16.32% (-4.43) |
| Pennsylvania | $2,391.8 | 7.44% (-0.39) | Massachusetts | $1,497.8 | 5.41% (+2.07) |
| Illinois | $2,125.4 | 12.18% (+4.13) | Ohio | $1,367.0 | 18.41% (+7.30) |
| New Jersey | $1,967.3 | 11.29% (-0.08) | Washington | $1,080.1 | 10.93% (-0.13) |
Note: Rates are the percentage of UI benefits that were overpaid.
Changes in Annual Report Overpayment Rates
The estimated CY 2004 overpayment rate of 9.92 percent of benefits paid in the State UI, UCFE, and UCX programs is an increase of 0.61 percentage points from CY 2003. It is important to note that BAM estimates are based on sample data. Taking into account sampling error, the 95% confidence interval for the CY 2004 estimate is 9.36% to 10.48%. The probability that the increase from CY 2003 is due to chance (sampling variation) is less than approximately 7% (p < .07). An estimated $3.4 billion of benefits were overpaid in CY 2004, compared with $3.81 billion overpaid in CY 2003. Most of the national increase in the overpayment rate is attributable to an increase in the percentages of UI benefits that were overpaid due to benefit year earnings issues (+0.52). The changes in the percentages of UI benefits overpaid for all other causes were 0.15 percentage points or less.
The observed changes in the cause components are less than the variation attributable to sampling.
| Overpayment Cause | CY 2004 Percentage of UI Benefits Paid |
Percentage Point Change From CY 2003 |
|---|---|---|
| Benefit Year Earnings | 2.76% | +0.52 |
| Separation Issues | 2.06% | -0.08 |
| Work Search | 1.55% | +0.06 |
| ES Registration Issues | 0.98% | +0.01 |
| Able & Available | 0.65% | +0.10 |
| Base Period Wage Issues | 0.61% | -0.01 |
| Other Eligibility Issues* | 0.46% | -0.14 |
| Dependents and Other Issues+ | 0.85% | +0.15 |
* Includes work refusals, self-employment, and illegal alien status.
+ Includes reductions in payments due to severance and vacation pay, Social Security and pension income, back pay awards, redeterminations, and payments made during a period of disqualification.
In CY 2004, 28 State Workforce Agencies (SWAs) reported increases in their overpayment rates from the previous year, and 23 SWAs reported decreases in their overpayment rates. The overpayment rate was unchanged in Indiana.
| State | Percentage Point Increase From CY 2003 |
Leading Cause of Increase (Percentage Point Increase) |
|---|---|---|
| New Mexico | 10.44 | Separation (+4.46) |
| Arizona | 8.63 | Work Search (+3.39) |
| Ohio | 7.30 | ES Registration (+5.80) |
| Virginia | 6.27 | ES Registration (+3.17) |
| Alaska | 5.40 | ES Registration (+7.39) |
| State | Percentage Point Decrease From CY 2003 |
Leading Cause of Decrease (Percentage Point Decrease) |
|---|---|---|
| Montana | -9.51 | ES Registration (-7.61) |
| Nebraska | -8.51 | ES Registration (-7.22) |
| Texas | -4.43 | Work Search (-4.39) |
| Tennessee | -3.91 | ES Registration (-1.99) |
| North Carolina | -3.46 | Separation (-1.69) |
Overpayment Cause and Responsibility
Causes of Overpayments in CY 2004
The leading cause of overpayment errors in CY 2004 was unreported or erroneously reported benefit year earnings and income, which accounted for over one-fourth of the $3.4 billion overpaid and 2.8 percent of the $34.4 billion in UI benefits paid in CY 2004. Separation issues were the second leading cause of overpayments, accounting for 21 percent of the amount overpaid and 2.1 percent of benefits paid.
Leading Cause of Overpayments by State
Unreported or erroneously reported benefit year earnings and income were the leading cause of overpayment errors in 23 of the states. Work search issues were the leading cause in 12 states, and employment service registration issues were the leading cause of overpayments in eight states. Separation issues were the principal cause of overpayment errors in seven states.
Overpayment Cause Trends
Unreported or erroneously reported benefit year earnings (BYE) account for the largest proportion of overpayments, and separation issues are the second leading cause of overpayments. In CY 2004 the percentage of UI benefits that were overpaid due to BYE issues increased, while the percentage of UI benefits that were overpaid due to separation issues was unchanged after increasing in each of the three previous years. Work search issues, the third largest cause of overpayments, increased for the third consecutive year. ES registration issues have also increased over the last five years. The percentage of UI benefits that were overpaid because of errors in reporting or recording base period wage issues has decreased slightly in recent years as the automation of wage reporting has improved.
Responsibility for Overpayments in CY 2004
Claimants alone were responsible for 58 percent of the dollars overpaid; claimant responsibility accounted for overpayments representing 5.8 percent of total UI benefits paid in CY 2004. Errors resulting in overpayments that were attributed exclusively to the state agency accounted for 15 percent of the amount overpaid and represented 1.5 percent of the UI benefits paid. The claimant and agency shared responsibility for an additional 9 percent of the overpayments. The category "All Other" includes overpayments for which more than one party has responsibility, such as agency and employers, as well as third parties.
The following chart displays the overpayment responsibility of the major responsibility classifications for the period CY 2000 to 2004.
Filing Method and Payment Error
In recent years state agencies have facilitated the filing of UI claims through the implementation of automated telephone and Web-based claims filing systems. Are claims filed using these methods more likely to have overpayment errors than claims filed in-person or by mail?
Because different types of payment errors are likely to occur at different points in the UI claims process, separate analyses were conducted for initial claims filing methods and continued claims filing methods.
Initial Claims Filing Methods
Although BAM samples and investigates individual paid weeks of UI, BAM auditors, as part of the BAM data collection record, code the filing method used when the claim for unemployment compensation was initially filed. BAM investigates all eligibility issues affecting the claim, including those that are evaluated through the normal claims taking procedures when a new initial claim is filed -- base period wage, separation, illegal alien, ES registration, identity theft, and dependents allowance issues. Rates of improper payments attributed to these issues were computed for each of the major filing methods.
The initial claims for more than half of the UI payments sampled in CY 2004 were filed by telephone. Initial claims filed in-person represented 18.7 percent of the total, and 17.5 percent of the initial claims were filed through the Internet.
Claims filed by telephone had the highest overpayment rates, both in terms of the percentage of UI weeks paid (10.5 percent) and the percentage of UI benefits paid (4.8 percent). Between 6.2 percent and 7 percent of the UI weeks paid for which the initial claim was filed in-person, through the Internet, or other methods (mail and employer-filed claims) were improperly paid. The UI benefit overpayment rates for these claims varied little, from 3.1 to 3.4 percent.
Claimants who filed their initial claims through the Internet received higher weekly UI benefits, compared with other filers, indicating that, on average, these claimants earned higher wages prior to their unemployment. Because of these higher weekly benefit amounts, Internet-filed claims also had highest average overpayment errors, although their overpayment rate was lower than claimants who filed by telephone.
Continued Claims Filing Methods
Nearly two-thirds of the continued claims for UI during CY 2004 were filed by telephone. A little more than one-fifth of the continued claims were filed by mail, and fewer than ten percent were filed through the Internet. Claims filed by other methods (in-person or employer-filed) made up less 3 percent of all continued claims.
Rates of improper payments attributed to issues affecting the claimant’s continued eligibility to receive UI benefits were computed for each of the filing methods. These eligibility issues include benefit year earnings, other benefit year income (such as severance and vacation pay, Social Security benefits, and pension income), able and available issues, work search, refusal of suitable work, and self-employment.
As is the case with initial claim filing methods, continued claims filed by telephone had the highest overpayment rates, both in terms of the percentage of UI weeks paid (11.5 percent) and the percentage of UI benefits paid (7.2 percent). The overpayment rates for claims filed by mail (6.3 percent of UI weeks paid and 4.2 percent of UI benefits paid) or through the Internet (6.9 percent of UI weeks paid and 3.7 percent of UI benefits paid) were lower than those filed by telephone.
Also consistent with the initial claims data, claimants who filed their continued claims through the Internet received higher weekly UI benefits, compared with other filers. However, unlike claimants who filed their initial claims via the Internet, claimants who used the Internet to file their continued claims did not have the highest average overpayment error. Continued claims filed by mail had the highest average overpayment ($163.74), although the differences among the filing groups are not large enough to be statistically significant.
Average Amount Overpaid
Another approach to measuring UI benefit payment integrity is to examine the average amount overpaid. Average overpayments can be expressed either in terms of the number of erroneous payments or in reference to a workload measure.
To identify an appropriate workload measure, overpayment errors are classified by cause: monetary (base period wage and dependents allowance issues), separation (voluntary quit and discharge), and nonseparation (benefit year earnings and income, able and available, work search, ES registration, refusal of work, alien status, and other continuing eligibility issues). The workload measure relates to points in the UI claims process at which an overpayment error can occur. For example, the state agency must determine that the claimant is monetarily eligible for benefits and that no disqualifying separation issue exists before a week of UI benefits can be claimed. Nonseparation issues relate to the claimant's eligibility to receive benefits for continued weeks of unemployment compensation claimed. Workload measures are summarized in the following table.
| Cause | Workload Measure | CY 2004 |
|---|---|---|
| Monetary | Number of first payments | 8,489,349 |
| Separation | Number of first payments + Number of additional claims, excluding ACs for which a nonmonetary determination resulted in a denial | 14,060,925 |
| Nonseparation | Number of UI weeks paid | 137,894,102 |
Monetary overpayments average approximately $32 per erroneous payment. These errors generally represent a reduction of the claimant's weekly benefit amount due to erroneously reported or recorded base period wages. However, most of these claimants remain eligible for some UI benefits. In contrast, separation overpayments often involve complete disqualification (due mostly to voluntary quits and discharges), resulting in an average overpayment of $215 per erroneous payment. Nonseparation errors average $160 per overpayment and encompass several issues. Some of these, such as able and available and work search violations, result in total disqualification for the compensated week of unemployment, whereas other issues, such as earnings from part-time employment or income from social security or pensions result in a partial reduction of benefits for the week of unemployment that is affected.
Workload averages place overpayments in the context of the number of transactions and eligibility determinations made in the UI system. The average overpayments per workload are computed by dividing the amount of benefits overpaid in each cause category by the appropriate workload measure. For example, BAM estimates that approximately $229 million were overpaid due to monetary issues. Dividing this amount by the 8,489,349 first payments results in an average overpayment of approximately $27. Average overpayments per workload for separation and nonseparation errors are significantly smaller than the average amount per overpayment. For example, dividing total overpayments due to separation issues ($710.6 million) by the 14 million first payments and eligible additional claims* produces an average overpayment of $51. Nonseparation overpayments, which comprise issues related to the claimant's continued eligibility during his or her benefit year (for example, benefit year earnings, able and available, and work search issues) totaled nearly $2.5 billion, an average of $18 per UI week paid.
* Additional claims are filed when a claimant returns to work during the benefit year before exhausting the maximum amount of UI benefits to which he or she is entitled, is separated from that employment, and files a claim for additional weeks of unemployment. The agency determines eligibility based on the circumstances of the claimant's separation from the intervening employment.
Prepared by:
U. S. Department of Labor
Employment and Training Administration
Office of Workforce Security
Division of Performance Management
August 2005