Statistics on metro-area employment are among the most timely and comprehensive source of information about economic conditions on a local and regional level. In fact, the monthly data from the Bureau of Labor Statistics (BLS) are often featured prominently on these pages. Based on a survey of employers, these data are compiled by the BLS as part of its Current Employment Statistics (CES) program.

Despite representing a broad sample of employment, the CES survey is incomplete. (See below.) Each year, the BLS carries out a benchmark revision, in which it uses information from the more comprehensive Quarterly Census of Employment and Wages (QCEW) to revise the monthly payroll data. The quarterly report is a very comprehensive measure of employment, based on information about workers covered by state and federal unemployment insurance programs. Although comprehensive, the QCEW reports are available only after a lag of six to seven months. Consequently, they are of limited value for gauging current economic conditions.

To bridge the gap, economists are left with a two-step process for evaluating local labor markets. Each year, the BLS uses information from the QCEW to establish new benchmarks for the CES data, bringing the sample data more closely in line with the census data. Between benchmark revisions, monthly changes reflect the incomplete nature of the CES survey.

As a result, once per year (in early March), we are presented with an employment picture that is sometimes dramatically different than recent data had been indicating. This is particularly true around turning points in economic activity, when incomplete survey data are more likely to miss important developments in local labor markets.

This year, in the midst of a recession, data revisions for metro areas in the Eighth Federal Reserve District might be expected to be particularly dramatic. As it turns out, this year’s revisions are relatively small.

Employment in District Metro Areas

Prior to this year’s benchmark revisions, employment data for Eighth District metro areas were showing job losses across the board. More variation was evident among smaller metro areas, but most were showing employment declines for the year as a whole.

The revisions to payroll employment resulted in employment gains for some metro areas in the District and losses for other metro areas for December 2008. Despite the upward revision in December 2008 employment for some metro areas, all major metro areas in St. Louis experienced a decline in jobs between December 2007 and December 2008.

St. Louis

Employment in the St. Louis metro area for December 2008 is now estimated at 1,354,200, up from the previous estimate of 1,346,300 (an increase of 7,900 jobs). New estimates from December 2007 to December 2008 reveal that job growth in St. Louis fell 1.4 percent over this period. This revised estimate is less than the previous estimate of –1.7 percent, in part due to a relatively smaller upward revision in December 2007 employment (revised from 1,369,300 to 1,374,000, an upward revision of 4,700 jobs). This pattern of revisions has the effect of improving the job growth figures for 2007. It had been earlier estimated that employment in St. Louis increased by a meager 0.1 percent, whereas the new figures show an increase of 0.5 percent.

Louisville

For Louisville, revised payroll employment for December 2008 is 613,800, down 3,400 jobs from the previous estimate. Revised estimates from December 2007 to 2008 reveal that job growth in Louisville fell 2.7 percent over this period. This revised estimate of job growth is a bit larger than the initial estimate of –2.5 percent. The downward revision for 2007 data is also reflected in a slower estimate of growth for that year. The new data show growth of 0.7 percent, compared with 1.1 percent in the earlier estimates.

Memphis

In Memphis, employment growth for 2008 was unaffected by the revisions, but only because a dramatic downward revision affected both December 2008 and December 2007. For both months, the revised figures show 5,500 fewer jobs than did the unrevised data. The revised data show a total of 633,500 jobs in the Memphis area at the end of 2008. The revisions had a substantial impact on job growth in 2007. Before the revision, the data showed an expansion of 5,400 jobs over the year, amounting to a growth rate of 0.8 percent. After the revision, 2007 employment appears to have been stagnant, with a net decline of about 100 jobs.

Little Rock

Of the four major metro areas in the District, Little Rock has fared the best over the past two years, and the revised data do little to change that perception. Revised data for December 2008 show total employment of 345,900 jobs in the metro area, an upward revision of 900 jobs. The revision for December 2007 represented only a slight decrease compared with the pre-revision levels. As a result, Little Rock employment growth for 2007 is essentially unchanged at 1.5 percent. For 2008, the new data show smaller job losses than previously estimated. Employment is now measured at –1.3 percent for the year, compared with –1.7 percent in the pre-revision estimates.

Small and Medium Metro Areas

Several of the smaller metro areas in the District experienced downward revisions for both years. Data for Bowling Green, Ky., Evansville, Ind., Jackson, Tenn., and Springfield, Mo., all show downward revisions for growth in 2007 and 2008. Data for Texarkana were revised downward for 2008, but job growth in that metro area remains positive for both years. The revisions also show positive job growth for Columbia, Mo., in 2008, but only because data for 2007 were revised sharply downward. Employment in Jefferson City, Mo., was unaffected by the revisions for 2007, but job losses were revised downward in 2008. In Fort Smith and the Fayetteville areas of Arkansas, small upward revisions for 2007 were balanced by downward revisions in 2008.

Metro-Area Employment Changes

	December 2007-December 2008				December 2006-December 2007
Large Metro Areas	Original Estimate as of January 2009		Revised Estimate as of March 2009		Original Estimate as of January 2009		Revised Estimate as of March 2009
Little Rock- N. Little Rock, Ark.	–5.8	–1.7	– 4.7	–1.3	5.2	1.5	5.0	1.5
Louisville, Ky.-Ind.	–16.1	–2.5	–16.9	–2.7	6.9	1.1	4.3	0.7
Memphis, Tenn.- Ark.-Miss.	–15.7	–2.4	–15.7	–2.4	5.4	0.8	–0.1	0.0
St. Louis, Mo.-Ill.	–23.0	–1.7	–19.8	–1.4	2.0	0.1	6.7	0.5
Small and Medium Metro Areas
Fayetteville-Springdale-Rogers, Ark.	–2.5	–1.2	–2.6	–1.2	0.9	0.4	1.2	0.6
Fort Smith, Ark.-Okla.	–1.6	–1.3	–1.4	–1.1	1.7	1.4	2.1	17
Texarkana, Texas-Ark.	1.2	2.1	0.9	1.6	0.7	1.2	0.9	1.6
Bowling Green, Ky.	–0.8	–1.3	–1.5	–2.4	1.8	2.9	1.6	2.6
Evansville, Ind.-Ky.	–2.5	–1.4	–4.6	–2.6	1.4	0.8	–0.2	–0.1
Jackson, Tenn.	–0.9	–1.4	–1.7	–2.7	0.3	0.5	0.0	0.0
Columbia, Mo.	0.0	0.0	1.1	1.2	1.0	1.1	–0.1	–0.1
Jefferson City, Mo.	–1.0	–1.2	–0.7	–0.9	1.5	1.9	1.	1.9
Springfield, Mo.	0.1	0.1	–4.6	–2.3	5.2	2.6	4.4	2.2

SOURCE: Bureau of Labor Statistics

The table shows how the estimates of jobs lost and gained changed between January and March 2009. For example, according to the estimate released in January 2009, the St. Louis MSA had lost 23,000 jobs between December 2007 and December 2008, and it had gained 2,000 jobs between December 2006 and December 2007. But, according to the revised estimate that was released in March 2009, the St. Louis MSA had lost 19,800 jobs between December 2007 and December 2008, and it had gained 6,700 jobs between December 2006 and December 2007.

Sidebar 1

A Tale of Two Data Sets

Current Employment Statistics (CES) is a monthly survey that is compiled from information from about 160,000 businesses and government agencies, representing approximately 400,000 individual work sites around the United States. Although the survey covers hundreds of thousands of employers, these employers make up only a small percentage of all businesses and work sites in the country.

The Quarterly Census of Employment and Wages (QCEW) is a tabulation of employment information for workers covered by state and federal unemployment insurance programs. As its name suggests, the QCEW is a census that achieves nearly 100 percent sampling of the nation’s employment and is, therefore, very accurate. Lags in the compilation of the data, however, mean that the QCEW is not a very good source for up-to-date information.

To bridge the gap, the Bureau of Labor Statistics (BLS) needs to augment the CES with an estimate of the number of establishments in the area. This can be difficult: When the economy is going into a recession, for example, old firms might be going out of business, while the formation of new firms might be slowing. The BLS doesn’t find out about the changes until the unemployment insurance records are updated, which can take several months or more. This lag is compounded by the fact that small firms, which provide the bulk of jobs, might need to provide unemployment insurance information only once a year rather than monthly or quarterly, as is required of larger firms.

Because of the lags and revisions to the QCEW data, the annual benchmarking affects employment data from the CES going back 21 months. Consequently, the estimates that were released in March have affected the yearly employment changes for 2007 and 2008. Note also that the estimates for job growth in 2008 will change again in March 2010, when the data for 2008 will once again be revised in the annual benchmark revision process.

Luke Shimek provided research assistance.