This chapter examines the effects the employee leasing industry has on State Unemployment Insurance (UI) trust funds. To measure these effects, we used wage-record data from four States to identify firms engaged in employee leasing and their client firms. Using State UI tax data, we were then able to analyze the differences in tax rates between the identified leasing and client firms. The chapter begins by detailing the States selected for the study. It then describes the methodology used in the analysis, the characteristics of both the identified leasing firms and the identified client firms, and then presents the results of the tax rate analysis.
Massachusetts and New York expressed interest in participating early on in the study. However, limited wage-record histories prevented their inclusion in the analysis. For example, Massachusetts had only 1 year of archived wage-record data at the time the study began.
In this study we address some of the difficulties of identifying employee leasing firms by developing an algorithm that uses State UI wage records to identify firms engaged in employee leasing. Appendix D addresses the methodology of the analysis. The appendix begins with a brief description of the State UI wage records. Then, it describes in detail the two-step process used to identify leasing firms and their client firms. Finally, it discusses some of the limitations of using State UI wage records to monitor the employee leasing industry. In the remainder of this chapter, we discuss the results of the analysis of the data on the potential leasing companies identified and the implications for State UI trust funds.
Using State UI wage-record data and the algorithm described above, this study identified firms engaged in leasing arrangements, including the leasing firms and their client firms (i.e., those firms to which the employees were leased). This section describes general characteristics of the firms identified by the study and examines characteristics of the leasing occurrences identified by the criteria.
The number of leasing firms identified by the algorithm varied across the States. In Florida, 101 firms were identified as being engaged in leasing. Texas had the second largest number of identified leasing firms with 93. In Oklahoma, only 18 firms were identified by the criteria as being leasing firms. Maryland had 15 firms identified. The differences in the number of leasing firms are due to the size of the States (e.g., in the first quarter of 1994, both Texas and Florida had more than 300,000 employers, while Oklahoma had 65,000); the industry mix prevalent in the States; and the age of the leasing industry in the State.
As expected, most of the identified leasing firms were large. The mean employment levels for the leasing firms for the first quarter of 1994 for all four States were more than 600 employees. Oklahoma was first with 1,996 employees. Maryland had the smallest leasing firm mean employment, 633 employees2. Also, more than three-quarters of all leasing firms in all four States had more than 100 employees in the first quarter of 1994; and in Florida and Texas, more than 60 percent of the identified leasing firms had more than 500 employees in the first quarter of 1994.
The number of identified client firms for each identified leasing firm also varied across the States. The mean number of client firms per identified leasing firm in Oklahoma was approximately five, the lowest of the three States for which the number of separable clients could be identified. Identified leasing firms in Texas had 13.5 clients on average, while the identified leasing firms in Florida had slightly more than 23 clients.
The most common SIC code for the identified leasing firms was 7363. Given both the criteria used to identify the leasing firms and the classification procedures of most States, this result is not surprising. Roughly one-fourth of the identified leasing firms in both Texas and Florida had a SIC code of 7363, while 8 of the 18 (44 percent) leasing firms in Oklahoma had a SIC code of 7363. Of Maryland's 15 identified leasing firms, 6 had a SIC code of 7363 (40 percent), and another 7 were in the specialized services series (8700). The other SIC codes of the leasing firms varied across the States, although a number of the leasing firms in each State had a two-digit SIC code of 80 (Health Services). Exhibit 4-1 displays the number, size distribution, mean number of client firms, and major SIC codes for the leasing firms identified in each of the four States.
The algorithm identified more leasing firms in Florida than in Texas, and it identified more total client firms in Florida than in Texas. In Florida, 2,350 client firms were identified; in Texas, 1,999 client firms were identified. In Oklahoma, with its much smaller employer and identified leasing firm base, 93 firms were identified as client firms.
The most common two-digit SIC code of identified client firms in Florida was 17 (Special Trade Contractors); in Texas, it was 73 (Business Services); and in Oklahoma, it was 50 (Wholesale Trade Durable Goods). In all three of these States, the two-digit SIC codes of 80 (Health Services), 73, and 65 (Real Estate) accounted for more than 5 percent of the identified client firms' SIC codes. In addition, the two-digit SIC codes of 17 and 50 were also relatively common among all three States' identified client firms. Exhibit 4-2 displays the most commonly occurring two-digit SIC codes for the identified client firms in all three States. To the extent that the identified leasing firm list includes cases that are, in fact, successorship, this may affect the distribution of SICs of "client" firms. This may help explain the ranking of Real Estate firms in Exhibit 4-2.
The differences in the identified client firms' SIC codes and the identified leasing firms' SIC codes indicate substantial shifts of employment from client SICs to leasing company SIC categories as a result of employee leasing. This can be misleading in terms of reported employment by industry categories. It is this shift that the earlier described BLS study is attempting to rectify in the ES 202 program.
An example of the pattern of leasing occurrences over time is shown in Exhibit 4-3 for Texas. The first three columns display information about the identified leasing firm: the first column displays its scrambled EIN, the second column lists its SIC code, and the third column shows its number of employees as of the first quarter of 1994. The fourth and fifth columns display information about the identified client firm: its scrambled EIN and its SIC code, respectively. The remaining columns provide information about the leasing occurrences identified for that particular leasing firm/client firm combination for the preceding years (the first quarter of 1993 back to the first quarter of 1990).
Exhibit 4-1 Unemployment Insurance Service Department of Labor Characteristics of Identified Leasing Firms By State |
||
| Florida | ||
| Number of Estimated Leasing Firms, 1994 | 101 | |
| Average Employment, 1994:I | 1,996 | |
| Average Number of Client Firms, 1994:1 | 23.3 | |
| Size of Leasing Firms | ||
| Employees | Firms | |
| 6 - 25 | 2 | |
| 26 - 100 | 8 | |
| 101 - 500 | 25 | |
| 501+ | 66 | |
| Major SIC Codes | 7363 |
26 |
| 87 |
9 | |
| 80 |
8 | |
| Maryland | ||
| Number of Estimated Leasing Firms, 1994 | 15 | |
| Average Employment, 1994:I | 633 | |
| Average Number of Client Firms, 1994:I | n/a | |
| Size of Leasing firms | ||
| Employees | Firms | |
| 6 - 25 | 0 | |
| 26 - 100 | 1 | |
| 101 - 500 | 8 | |
| 501+ | 6 | |
| Major SIC Codes | 7363 |
6 |
| 87 |
7 | |
| Oklahoma | ||
| Number of Estimated Leasing Firms, 1994 | 18 | |
| Average Employment, 1994:I | 636 | |
| Average Number of Client Firms, 1994:1 | 5.2 | |
| Size of Leasing Firms | ||
| Employees | Firms | |
| 6 - 25 | 1 | |
| 26 - 100 | 3 | |
| 101 - 500 | 10 | |
| 501+ | 4 | |
| Major SIC Codes | 7363 |
8 |
| 1389 |
2 | |
| Texas | ||
| Number of Estimated Leasing Firms, 1994 | 93 | |
| Average Employment, 1994:I | 1,582 | |
| Average Number of Client Firms, 1994:I | 13.5 | |
| Size of Leasing firms | ||
| Employees | Firms | |
| 6 - 25 | 3 | |
| 26 - 100 | 15 | |
| 101 - 500 | 38 | |
| 501+ | 93 | |
| Major SIC Codes | 80 |
10 |
| 13 |
9 | |
Exhibit 4-2 Unemployment Insurance Service Department of Labor Most Frequent Two-Digit SIC Codes of Client Firms in Florida, Oklahoma, and Texas |
|||||||||||
| Florida | Oklahoma | Texas | |||||||||
| SIC | Description | Freq. | Pct. | SIC | Description | Freq. | Pct. | SIC | Description | Freq. | Pct. |
| 17 | Special Trade Contractors | 208 | 8.85% | 50 | Wholesale Trade--Durable Goods | 8 | 8.60% | 73 | Business Services | 150 | 7.50% |
| 80 | Health Services | 147 | 6.26% | 73 | Business Services | 6 | 6.45% | 80 | Health Services | 150 | 7.50% |
| 73 | Business Services | 132 | 5.62% | 49 | Electric, Gas, ad Sanitary Services | 5 | 5.38% | 50 | Wholesale Trade-- Durable Goods | 147 | 7.35% |
| 58 | Eating & Drinking Places | 115 | 4.89% | 65 | Real Estate | 5 | 5.38% | 17 | Special Trade Contractors | 146 | 7.30% |
| 65 | Real Estate | 107 | 4.55% | 80 | Health Services | 5 | 5.38% | 65 | Real Estate | 127 | 6.35% |
| 50 | Wholesale Trade--Durable Goods | 102 | 4.34% | 42 | Trucking & Warehousing | 4 | 4.30% | 87 | Engineering & Mgmt. Services | 81 | 4.05% |
| 72 | Personal Services | 69 | 2.94% | 55 | Automotive Dealers & Service Stations | 4 | 4.30% | 60 | Depository Institutions | 73 | 3.65% |
| 87 | Engineering & Mgmt. Services | 69 | 2.94% | 17 | Special Trade Contractors | 3 | 3.23% | 42 | Trucking & Warehousing | 68 | 3.40% |
| 7 | Agricultural Services | 65 | 2.77% | 27 | Printing & Publishing | 3 | 3.23% | 64 | Insurance Agents, Brokers & Srvs. | 62 | 3.10% |
| 75 | Auto Repair, Services, & Parking | 64 | 2.72% | 34 | Fabricated Metal Products | 3 | 3.23% | 51 | Wholesale Trade-- Durable Goods | 61 | 3.05% |
| 55 | Automotive Dealers & Service Stations | 60 | 2.55% | 56 | Apparel & Accessory Stores | 3 | 3.23% | 13 | Oil & Gas Extraction | 58 | 2.90% |
| 59 | Miscellaneous Retail | 57 | 2.43% | 72 | Personal Services | 3 | 3.23% | 59 | Miscellaneous Retail | 54 | 2.70% |
| 15 | General Building Contractors | 54 | 2.30% | 95 | Environmental Quality & Housing | 3 | 3.23% | 15 | General Building Contractors | 52 | 2.60% |
| 99 | Nonclassifiable Establishments | 54 | 2.30% | 88 | Private Households | 47 | 2.35% | ||||
| 51 | Wholesale Trade-Nondurable Goods | 53 | 2.26% | 72 | Personal Services | 38 | 1.90% | ||||
| 86 | Membership Organizations | 46 | 1.96% | 55 | Automotive Dealers & Service Stations | 37 | 1.85% | ||||
| 57 | Furniture & Homefurnishings Stores | 44 | 1.87% | 49 | Electric, Gas & Sanitary Services | 36 | 1.80% | ||||
| 88 | Private Households | 44 | 1.87% | 81 | Legal Services | 36 | 1.80% | ||||
| 42 | Trucking & Warehousing | 33 | 1.40% | 47 | Transportation Services | 34 | 1.70% | ||||
| 70 | Hotels & Other Lodging Places | 33 | 1.40% | 76 | Miscellaneous Repair Services | 33 | 1.65% | ||||
Exhibit 4-3 Unemployment Insurance Service Department of Labor Leasing Occurrences by Year for Sample Texas Employer |
||||||||||||
| Client Firm | ||||||||||||
| Leasing Firm Information | Information | 1993: First Quarters | 1992: First Quarters | 1991: First Quarters | 1990: First Quarters | |||||||
| 1994:1 | No. in | Client Firm | No. in | Client Firm | No. in | Client Firm | No. in | Client Firm | ||||
| EIN | SIC | Emp. | EIN | SIC | Changeover | Emp. | Changeover | Emp. | Changeover | Emp. | Changeover | Emp. |
| 202728876 | 5541 | 1 | 2 | |||||||||
| 203032964 | 6510 | 10 | 53 | |||||||||
| 204549212 | 1799 | 1 | 1 | |||||||||
| 205425130 | 3089 | 9 | 21 | |||||||||
| 205425156 | 3089 | 7 | 17 | |||||||||
| 206585383 | 2339 | |||||||||||
| 224621663 | 7363 | 1,050 | 8 | 22 | ||||||||
| 206632451 | 5810 | 11 | 30 | |||||||||
| 207105319 | 7622 | 7 | 8 | |||||||||
| 207440048 | 5810 | 1 | 2 | |||||||||
| 221995842 | 5148 | 2 | 3 | |||||||||
| 222406896 | 5064 | 11 | 25 | |||||||||
| 229535365 | 6553 | 1 | 2 | |||||||||
In Texas, the average number of employees involved in leasing occurrences increased between 1990 and 1992 and then decreased between 1992 and 1993. In Florida, the average number of employees involved in leasing occurrences increased between 1990 and 1991 but then fell between 1991 and 1993. For both States, increases in the average number of employees involved in leasing occurrences were always accompanied by increases in the average employment level of the identified client firms; likewise, decreases in the average number of employees involved in leasing occurrences were accompanied by decreases in the average employment level of the client firms. As a result, the percentage of an identified client firm's workforce changing employers (to an identified leasing firm) remained somewhat consistent. The only anomaly occurred in Texas in 1992 when, on average, 45 percent of the identified client firms' workforces were involved in leasing occurrences.
In Oklahoma, the average percentage of the identified client firms' workforces that changed over increased each year with the number of leasing occurrences, from 12 percent in 1991 to 17 percent in 1992, to 25 percent in 1993. Somewhat surprisingly, the average number of employees involved in the leasing occurrences declined between 1991 and 1992 but then rose again between 1992 and 1993. Exhibit 4-4 displays the number of leasing occurrences, the average number of employees involved in each leasing occurrence, the average number of employees in the client firms' workforces, and the average percentage of the client firms' workforces involved in the leasing occurrences for the years 1990 through 1993 for these three States.
Some UI administrators contend that the employee leasing industry has had an adverse effect on State UI trust funds. As noted earlier in this report, this can occur in a number of ways: changes in the experience rated tax rate; turnover of the corporate identity of the firm; and, delinquency/default on tax payments. Using the leasing firms and client firms identified by the process described in this report, this study examined the differences between the UI tax rates of the leasing and client firms. By comparing the tax rates of both types of firms, it is possible to determine whether leasing arrangements adversely affect State UI trust funds and to estimate the total effect on the trust fund. This section describes the analysis of the UI firm tax-rate data and the results of that analysis.
Measuring the effects the leasing industry has on State UI trust funds using UI wage records is not a straightforward task. Ideally, one would want to calculate the taxes paid by and benefits charged to the client firm for the year of the leasing arrangement and the taxes paid by and benefits charged to the leasing firm for the year after the leasing arrangement for only those employees who were leased. This is impractical because of the difficulty involved in accurately identifying the leased employees who collected benefits charged to either the leasing or client firms. This is particularly the case because, by virtue of the algorithm used to identify leased employees, those employees who changed over to an identified leasing firm had to remain employed by that firm until the first quarter of 1994. Instead, a more practical approach is to compare the UI tax rates paid by both the leasing and client firms in the year in which the changeover occurred.
Exhibit 4-4 Unemployment Insurance Service Department of Labor Leasing Occurrence Characteristics by State and Year |
|||||
| State | Year | Number | Avg. No. Of Employees |
Avg. Client Firm Employment |
Pct. of Client Workforce |
| 1990 | 80 | 45.8 | 184 | 25% | |
| Florida | 1991 | 176 | 54.9 | 202 | 27% |
| 1992 | 370 | 23.1 | 106 | 22% | |
| 1993 | 1743 | 14.9 | 50 | 30% | |
| 1991 | 7 | 15.9 | 133 | 12% | |
| Oklahoma | 1992 | 12 | 6.8 | 39 | 17% |
| 1993 | 75 | 35.3 | 141 | 25% | |
| 1990 | 240 | 29.3 | 100 | 29% | |
| 1991 | 260 | 60.2 | 204 | 30% | |
| Texas | 1992 | 477 | 92.3 | 207 | 45% |
| 1993 | 1098 | 30.8 | 90 | 34% | |
Upon first consideration, it would appear that the best comparison of tax rates would be the client firm's tax rate the year in which the leasing arrangement occurred, as compared with the leasing firm's tax rate the year after the leasing arrangement. However, because of changes in State tax schedules from one year to the next, this may not be an accurate measurement in the differences between the leasing firms' and client firms' tax rates. For example, if the State trust fund were to experience an increase in its balance from one year to the next, it is possible that nearly all firms would experience a decrease in their tax rates as a result of a shift in the tax schedules for the following year.
To eliminate the effects of shifts in schedules, it is best to compare tax rates for the same year for both the client and the leasing firms. Because many client firms lease their entire workforce, many client firms cease to have UI covered employees (and thus a UI tax rate) in years subsequent to the leasing. To maximize the number of leasing firm/client firm combinations analyzed, this study compares tax rates for the leasing and client firms for the year of the leasing occurrence. This analysis will not reflect changes in the tax rates of the leasing firms brought about by the States' experience-rating systems, but it will provide a measure of the change in the tax rate (and hence payroll taxes) for the leased employees as a result of leasing. Because most States use multiple years' (generally, 3 to 5) worth of payroll taxes and charged benefits to calculate their experience-rated tax rate, an analysis of the effects of leasing arrangements on leasing firms' experience rating would require the use of data taken over an extended period of time. However, such an analysis would be unsatisfactory because, with the passage of time, it would be difficult to attribute such experience-rating changes only to the leasing arrangements.
The analysis consists of the comparison of the mean tax rates for the leasing firms and client firms in each of the three States. We first calculate the unweighted mean tax rates of the leasing and client firms. Because occurrences involving a greater number of employees changing firms (and thus tax rates) would have a greater effect on the trust funds, a second component of the analysis calculates the mean tax rates for the leasing and client firms, weighted by the number of employees involved in each leasing occurrence. Thus, a leasing occurrence involving 500 employees would have a greater effect on the weighted mean tax rates for the leasing and client firms than one involving 50 employees. Using these weighted mean tax rates, we then estimate the total effect, in dollar amounts, the identified leasing arrangements had on the State trust funds.
Some States, including Florida, regulate the employee leasing industry within their State boundaries. To do so, they maintain lists of licensed employee leasing firms. Florida identified both their licensed leasing firms and their licensed client firms for this study, and additional analyses were conducted comparing the mean tax rates of the licensed firms (both leasing and client), with those firms identified using Florida UI wage records. These results make up the fourth component of this analysis.
In all three States (Florida, Texas, and Oklahoma) the average tax rates over the 4 years for the identified leasing firms were lower than the average tax rates for their client firms. The greatest disparity was in Florida, where the leasing firms' average rate was 0.90 percent lower than that of their clients. Texas had the least difference, with the mean tax rate of the identified leasing firms being only 0.28 percent lower than that of their clients (a 16-percent reduction). Oklahoma had a 0.40-percent difference between the mean rates of the leasing firms and client firms against an average tax rate for the client firms of 1.94 percent (a reduction in the rate of 21 percent). Exhibit 4-5 displays the unweighted mean tax rates for both the leasing firms and their clients for all three States.
When weighted by the number of employees involved in the leasing occurrence, the mean tax rates for the leasing firms in all three States were lower than the average tax rates for their client firms. The weighted mean tax rates of the leasing firms in both Texas and Florida were approximately 0.40 percent lower than those of their clients (19 and 26 percent reductions, respectively). In Oklahoma the leasing firms' weighted mean tax rate was 0.11 percent lower than their clients' (a 6-percent reduction). Exhibit 4-6 displays the weighted mean tax rates for the leasing and client firms for the three States.
The differences in the weighted average tax rates were created based on the actual tax rates of the leasing and client firms in the same year the year in which the changeover of employees from the client firm to the leasing firm occurred. This avoids the issue of the effects of changes in tax schedules from year to year based on overall trust fund effects.
The elements of the analysis are shown in Exhibit 4-7. Because leased employees are generally full-time workers, we assume that each employee who changes over to a leasing company receives at least the maximum taxable wage in the State in that year. Given that the highest, Oklahoma, is $10,700, this seems plausible. We then calculate the per-employee annual tax using the tax rate of the client firm and the leasing company applicable to each employee that changed over in the year they changed over. The difference in the average annual tax per employee is shown in column 6 of the exhibit. This amount, multiplied by the number of employees involved in the changeovers in each State yields an estimate of the loss to the State trust fund over the period of the analysis as the result of employee leasing. This is shown in column 7 of the exhibit.
To the extent that the employee who changed over did not earn at least the maximum taxable amount of wages in the State in each year, these estimates of loss to the trust fund are overestimates. However, more importantly, the estimates are generated only for the year in which the changeover occurred. By definition, once a worker changed over to the leasing company, that worker remained employed by the leasing company through the first quarter of 1994. Thus, to the extent that the workers remain with the same leasing company and client firms, the estimates presented here may be thought of as 1-year losses to the State trust fund.
Exhibit 4-5 Unemployment Insurance Service Department of Labor Unweighted Mean Tax Rates of Identified | ||||
| Leasing Firms and Their Clients | ||||
| State | Leasing Firms' Mean Tax Rate |
Client Firms' Mean Tax Rate |
Difference | Percent Reduction |
| Florida | 0.97% | 1.86% | -0.89% | -47.85% |
| Oklahoma | 1.54% | 1.94% | -0.40% | -20.62% |
| Texas | 1.47% | 1.75% | -0.28% | -16.00% |
Exhibit 4-6 Unemployment Insurance Service Department of Labor Mean Tax Rates of Identified Leasing Firms and Their Clients Weighed by Employees Involved in Leasing Occurrence |
||||
| State | Leasing Firms' Mean Tax Rate |
Client Firms' Mean Tax Rate |
Difference | Percent Reduction |
| Florida | 1.16% | 1.56% | -0.40% | -25.64% |
| Oklahoma | 1.65% | 1.76% | -0.11% | -6.25% |
| Texas | 1.78% | 2.19% | -0.41% | -18.72% |
Exhibit 4-7 Unemployment Insurance Service Department of Labor Estimates of Trust Fund Loss |
||||||
| State | Maximum Taxable Wage |
Total Employees in Changeovers |
Annual Client Firm Taxes per Employee |
Annual Leasing Firm Taxes per Employee |
Annual Tax Difference per Employee |
Estimated Annual Trust Fund Loss |
| Florida | $7,000 | 39,851 | $109.20 | $81.20 | $28.00 | $1,115,828 |
| Oklahoma | $10,700 | 747 | $188.32 | $176.55 | $11.77 | $8,792 |
| Texas | $8,500 | 87,552 | $197.10 | $160.20 | $36.90 | $3,230,669 |
Because we identified only 747 changeovers in Oklahoma, the fact that the trust fund loss estimate is small ($8,700) is not surprising. At the other end of the spectrum, the estimates for Texas, with 87,000 changeovers, runs to $3.2 million. Florida, with 39,800 changeovers has an estimated trust fund loss of more than one million dollars. Put in perspective, using 1992 taxable wages as the base, the largest loss was six tenths of a percent of the taxable wages in Texas. Thus, it would reduce the average tax rate by that amount.4 For Florida, it amounts to three tenths of 1-percent. In Oklahoma, the small size of the leasing industry had infinitesimal effect.
As noted earlier, Florida provided both their licensed leasing firms and those firms' licensed clients, coupled with the firms (both leasing and client) identified by using Florida's UI wage records, which allows comparison between the following combinations of leasing and client firms:
Identified unlicensed leasing firms and their identified client firms
Unidentified licensed leasing firms and their licensed client firms
Identified licensed leasing firms and their identified client firms
Identified licensed leasing firms and their licensed client firms
Only unweighted means were calculated for those leasing/client firm combinations involving unidentified licensed clients, because the licensed clients were not identified using the UI wage records and thus had no leasing occurrences and, therefore, no weighting variable.
For all four leasing/client groups examined, the unweighted mean tax rate was lower for the leasing firms. The identified licensed leasing firms had lower unweighted tax rates than both the identified unlicensed leasing firms and the unidentified licensed leasing firms.
For those licensed leasing firms identified by the UI wage-record algorithm, the unlicensed client firms (i.e., those identified by the algorithm but not by Florida) had a higher unweighted mean tax rate than the licensed clients. Furthermore, the difference between the leasing firms' and unlicensed client firms' unweighted mean tax rates was greater than that between the leasing firms' and licensed client firms' unweighted mean tax rates.
Even though the identified unlicensed leasing firms' unweighted mean tax rate is lower than their (identified) client firms' unweighted mean tax rate, the weighted mean tax rates of the client firms are lower than those of the leasing firms. This suggests that, although the identified leasing firms' tax rates are generally lower than their client firms', the larger leasing occurrences that occurred for the identified, yet unlicensed, leasing firms took place with those clients that had lower tax rates. Exhibit 4-8 displays the mean tax rates, both
Exhibit 4-8 Unemployment Insurance Service Department of Labor Mean Tax rates of Licensed and Unlicensed Florida Leasing and Client Firms |
|||||||
| Leasing Firms' Weighted Mean Tax Rate | Client Firms' Weighted Mean Tax Rate | Diff. | Leasing Firms' Unweighted Mean Tax Rate | Client Firms' Unweighted Mean Tax Rate | Diff. | Pct. Reduction | |
| Unidentified licensed Leasing Firms and Licensed Client Firms | n/a | n/a | n/a | 1.48% | 1.82% | -0.34% | -19% |
| Identified Licensed Leasing Firms and Licensed Client Firms | n/a | n/a | n/a | 0.99% | 1.78% | -0.79% | -44% |
| Identified Licensed Leasing Firms and Identified Client Firms | 0.72% | 1.79% | -1.07% | 0.84% | 1.94% | -1.10% | -57% |
| Identified Unlicensed Leasing Firms amd Identified Client Firms | 1.49% | 1.38% | 0.11% | 1.44% | 1.61% | -0.17% | -11% |
unweighted and, where appropriate, weighted, for the combinations of licensed and unlicensed leasing and client firms for Florida.
In addition to the reasons given above why leasing firms could be expected to have lower payroll tax rates than the average of their client firms, there are several ways in which leasing firms can manipulate their payroll tax rate over time to lower their UI contributions. These include the following:
Establishing a low rate for an employer account by maintaining an account with minimal (1 or 2 employees) employment and then moving their leasing employment into that account
Letting an account whose experience-rated tax rate has increased become inactive and transferring the leasing employment to another account (with a lower payroll tax rate)
Because Florida provided the most complete data of the three States, we attempted to identify instances of these methods of tax rate manipulation. To do this, we first determined how many of the identified leasing firms had tax rates lower than the beginning tax rate (2.7 percent in Florida) at the time of their first changeovers. Then, we identified firms that experienced a substantial increase in their employment level subsequent to becoming experience-rated. Finally, we examined the tax rate patterns of the leasing firms that became inactive in 1994 or 1995 and the universe of firms that became inactive in those same years.
Because Florida maintains only a 5-year archive of its firms' tax rates in electronic format, and we received the firm's data after the assignation of the 1995 tax rates, we were unable to determine the tax rate of all 101 identified leasing firms at the time of their first leasing occurrence. Of the identified leasing firms that did have a tax rate the year of their initial leasing occurrence, only 27 percent had a tax rate of 2.7 percent. This suggests that most firms have previously established tax rates before experiencing any leasing occurrences.
Using the tax rate and employment data, we also identified 33 firms that had a starting tax rate of 2.7 percent with a very low employment level (less than 5 employees) and experienced a decline in their tax rate with a subsequent substantial increase in their employment. Six of these firms were licensed leasing firms (three of which we also identified). All these firms' employment levels increased by more than 400 employees the year of or the year after the decrease in their tax rates.
Examining the tax rate patterns of firms that became inactive in 1994 or 1995 reveals that most of the 10 identified leasing firms that became inactive experienced increases in their tax rates in their previous years. On average, these firms experienced an annual increase of 0.26 percent in their tax rates. On the other hand, the universe of firms that became inactive in 1994 or 1995 experienced no increase in their tax rates in earlier years.
In addition to the analysis of payroll tax rates of leasing and client firms discussed earlier, we examined the chargeable benefits and taxes paid by the leasing firms and client firms. For this we obtained additional data from Florida and Oklahoma. The employee leasing industry's effect on the State UI systems can also be measured by comparing the UI taxes paid and the benefits charged to the leasing firms to the UI taxes paid and the benefits charged to the client firms. Leasing firms argue that, because they have the ability to move employees from one client firm to another, they should have lower benefit charges than their client firms had previously. If this is the case, the ratio of taxes paid to benefits charged should be higher for the leasing firms than the client firms. To examine this issue, we requested total taxes paid and benefits charged for the firms we identified as leasing companies and for their client firms over the period used for the study. If the leasing firms' tax-benefit ratios were lower than the client firms' tax-benefit ratios, then the leasing arrangements would be having a negative impact on the State UI trust funds.
Exhibit 4-9 shows the totals of taxes paid and benefits charged to leasing companies and their client firms in Oklahoma and Florida for the period 1990 to 1994. Initially, these results would seem to contradict the argument that the leasing companies negatively affect the trust funds relative to the experience of their client firms. In both States the ratio of taxes to benefits is higher for the leasing firms than for the client firms. However, there are reasons why this conclusion may not be warranted.
In the previous analysis, we were able to compare the tax rates paid on the wages of the leased employees for the leasing firms and client firms. In this case we had to accept total taxes and benefits for the firms as a whole, not just for the leased employees. This is because the leased employees, of necessity, had to remain employed over the entire period. Therefore, no benefits would be charged for those employees. Consequently, the data cover all employees of the firms over the period.
Many leasing companies also provide temporary employees. Taxes are paid on temporary employees, but they are less likely than regular employees to qualify for unemployment benefits. Therefore, we might expect benefits charged to leasing (and temporary help) companies to be lower relative to taxes than for client firms.
Note than in Exhibit 4-9 the ratios of taxes to benefits for both leasing and client firms in both States are less than one in all but one case substantially. Because nonchargeable benefits are not included, the expectation was that these ratios would be greater than one. The fact that these ratios are less than one is perhaps a result of the recessionary period of the early 1990's. During this time, the trust fund in Florida fell from $2.4 billion to $1.6 billion and annual taxes were about two-thirds of annual benefits paid. In Oklahoma, total State taxes and benefits paid were approximately equal in these years. However, the ratio of taxes to benefits for both leasing firms and client firms declined by more than half over the period. Thus, some of the difference could result from the inability of the experience rating system to completely adjust the tax structure to cyclically higher benefit payout. Oklahoma is one of two benefit-wage ratio States.
All of this leads us to conclude from this analysis that the aggregate tax-to-benefit ratio is inadequate to measure leasing company effects on UI trust funds relative to their client firms.
Exhibit 4-9 Unemployment Insurance Service Department of Labor Dollar Tax/Benefit Ratios Leasing Companies and Client Firms 1990 to 1994 |
|||
| State | Taxes | Benefits | Ratio |
| Oklahoma | |||
| Leasing Co. | $2,021,495 | $3,466,756 | 0.58 |
| Client Firms | $2,134,200 | $10,415,962 | 0.2 |
| Florida | |||
| Leasing Co. | $54,713,627 | $59,445,823 | 0.92 |
| Client Firms | $69,722,183 | $121,000,295 | 0.58 |
1 Unfortunately, we could not obtain additional tax rate and benefit data from Maryland. Therefore, we have not included Maryland in the analysis of tax rates.
2 For the study, employment was calculated as the number of wage records for a given employer. Although this employment figure may overstate the exact number of employees on a given day, it provides a consistent method of determining employment across firms.
3Both the client firm and leasing firm were required to have a tax rate for the year of the leasing occurrence for that particular leasing occurrence to be included in both the weighted and unweighted analysis.
4 Committee on Ways and Means, U.S. House of Representatives, 1994. Overview of Entitlement Programs, July, Table 7-12, p. 297.