Quarterly Census of Employment and Wages

Proper Monthly Employment Reporting for State Unemployment Insurance and Common Reporting Problems

The monthly employment data reported on both the Quarterly Contribution Report and Multiple Worksite Report should be a count of all full- and part-time workers who worked during or received pay (subject to Unemployment Insurance wages) for the pay period which includes the 12th day of the month. The count should be unduplicated, so that for the reference period in any month, an employee should be counted only once.

Proper Reference Period

The reference period is the pay period that includes the 12th day of the month. Since not all pay periods are of the same duration, the following examples use March 1996 as a reference month to illustrate proper reporting:

If the pay period is a calendar week, the reference period is March 10-16. If the pay period is weekly but does not match the calendar week, it is whatever seven-day period includes the 12th.

If the pay period is biweekly, corresponding to two calendar weeks, there are two possibilities: the 12th may fall into the first or second week of the pay period. In the first case, March 10-23 is the proper reference period. If the 12th falls in the second week of the pay period, the reference period is March 3-16. Again, If the pay period is biweekly but does not match calendar weeks, it is whatever fourteen-day period that includes the 12th.

Some employers pay twice a month according to the calendar month. The pay periods are generally the first through the fifteenth and the sixteenth through the end of the month. The proper reference period in this case is March 1-15. The 12th always falls in the first half of each month.

This case is straight forward. If the employer pays only once a month, according to the calendar month, the pay period must include the 12th of the relevant month. The reference period here is March 1-31.

Generating a Proper Count

Some types of incorrect counts based on an improper reference period are described in this section. Other conceptual misunderstandings result in erroneous data as a result of counting wage records, checks, cumulative employment, or available employees—none of which is necessarily equal to the proper count of employees. These improper methods are described in this section.

Some employers may mistakenly provide a count of employment as of the time the Quarterly Contribution Report is received—possibly the pay period at the end of the third month of the quarter or the beginning of the following month. This is not generally the proper reference period and, as a result, the employment level reported for that month is somewhat inaccurate. Even more serious, employment for the first two months may be reported as the same figure or estimated. Thus, data for the first two months of the quarter may be even more inaccurate.

The wage record count may be a more common problem, probably because this count is also required by most States to be reported on the Quarterly Contribution Report. In what are called "wage reporting" States, employers are required to report what are called "wage items" or "wage records." This report is a listing by Social Security number of all persons who received pay during the quarter along with their total wages. Typically, the data are reported along with a count of the number of records (persons). Some employers mistakenly believe that this wage record count is also the employment count requested for the pay period including the 12th. They then report that same value for each month on the Quarterly Contribution Report.

Only in an extreme case where the employment level is absolutely constant for the entire quarter is the wage record count equal to the employment count for each of the three reference periods (months). If there is any turnover or change in staffing level, the wage record count will overstate the true employment level for each reference period. The greater the degree of turnover or staffing changes, the larger the overstatement.

Some employers report employment by counting the number of checks written within the pay system for a particular period. This approach provides an accurate employment count only if the pay system limits a person to one check per pay period for all types of wage payments combined. Otherwise, the employment count is overstated to the degree that employees receive more than one pay check for the reference period. Possible types of additional payments include bonuses, commissions, overtime pay, vacation pay, sick pay, holiday pay, moving expenses, severance pay, and contributions to an employee savings plan (e.g., 401(k)). Generating a proper employment count is facilitated by ensuring that any Social Security number is counted no more than once.

Using the check issued date does not guarantee reporting for the proper reference period. The reference pay period is the pay period for which—not in which—the employee is being paid. If an employer pays with a time lag, a check for work in the reference period will typically be dated in a later pay period.

Some employers have mistakenly been providing cumulative counts of everyone who has worked for them since the beginning of some time period. That period might be a calendar quarter, calendar year, tax year, fiscal year, or something else. The resulting pattern of data reported shows employment for each month at or above the level of the preceding month. When the employer's file is purged, as it is periodically, a precipitous decline in reported employment results. Thus, employment is overstated by an amount which grows each month until the overstatement becomes substantial (again, the degree depends on turnover). Then a sizeable drop in reported employment appears, but it is due to an administrative practice, rather than any economic phenomenon.

One practice that causes this type of overstatement of employment to occur is when an employer reports the number of "active employees" each month. Again, if the employer's active file is not updated every pay period to reflect turnover, an overstatement of monthly employment will result until the file is subsequently purged.

This count is provided typically by employers who maintain a file or list of employees who may potentially be called upon to work. The count reported is the number of people on that list, rather than those who actually worked or received pay (e.g. sick pay). Employment is overstated because the count adds those who were potentially available but did not actually work or received no pay during the relevant period to those who did work. This situation is most likely to occur in certain industries, such as education, retail sales, and temporary help. In these industries, employers frequently maintain lists of certified substitute teachers, contingent sales staff, and available temporaries.

Impact of Proper Reporting

The employment and wage data provided on the Quarterly Contribution Report and Multiple Worksite Report impact directly or indirectly a large number of important economic statistics, including the Gross Domestic Product and other measures of the labor market and overall economy. The accuracy of these and other economic indicators relies on the accuracy of the pay figures reported by employers.

If a review of your state Quarterly Contributions Reports indicates that you have been filing inaccurate monthly employment counts, please contact Michael Searson at (202) 691-6469. The BLS will assess the impact of the incorrect reporting and assist in developing a plan to implement the necessary changes. In this manner, real changes in employment levels will not be significantly affected by corrected employer reporting procedures. Questions on proper employment reporting should also be addressed to Mr. Searson at:

BUREAU OF LABOR STATISTICS Postal Square Building Suite 4840 2 Massachusetts Ave., N.E. Washington, D.C. 20212 FAX 202-691-6645

Last Modified Date: October 16, 2001

Recommend this page using: