Skip to main content

Dallas Fed’s Texas jobs estimates provide early, accurate assessment

Keith R. Phillips and Alexander T. Abraham

The Bureau of Labor Statistics (BLS) annually revises regional job estimates in a process called benchmarking. A Federal Reserve Bank of Dallas adjustment provides researchers a more current means of assessing Texas economic conditions.

Much regional economic analysis is derived from monthly Current Employment Statistics (CES) industry-level employment data for U.S. states and metropolitan statistical areas.

We show here how the Dallas Fed’s early benchmarked data for 2018 were very close to the final BLS benchmarked CES figures and, thus, provided a timelier view of economic growth in the state. To improve employment data accuracy prior to the annual revision, the Dallas Fed has since the mid-1990s applied several adjustments to the Texas series, including a preliminary benchmark estimate.

Nationally, the BLS uses a year’s worth of figures from the Quarterly Census of Employment and Wages (QCEW) in its benchmark process. The data are derived from forms that employers must fill out for unemployment insurance programs. The QCEW is comprehensive, encompassing more than 95 percent of total jobs.

Early benchmarking accuracy

Preliminary QCEW data for Texas—covering each month—are released with about a four-and-a-half month lag after the reporting period. These figures are usually very close to the final annual benchmark and, thus, are a good tool to accurately estimate the benchmark revision.

To adjust the BLS’ official employment series using these preliminary data, special care must be taken to account for seasonal patterns, which differ between the QCEW data and the more recent months of survey-based CES data.

In February 2019, Dallas Fed early benchmarked data showed job growth of 2.5 percent from September 2017 to September 2018; official CES data at the time indicated 3.4 percent growth.

When the BLS’ annual benchmark was released in March, the official data were revised to 2.6 percent—very close to the earlier Dallas Fed figure.

What the data showed at the end of 2018 may be even more important since that is typically when analysts make their forecasts for the coming year. Having a better idea of growth in the current year significantly improves the forecast for the upcoming year.

For example, an analyst or model might predict growth in 2019 would be about the same as in 2018. Using CES data at year-end 2018 would have predicted 2019 growth of about 3.1 percent (based on growth through November 2018).

Alternatively, Dallas Fed data on Dec. 21, 2018, which incorporated the early benchmark through June 2018, suggested 2.4 percent—very close to what the annual benchmark showed when it was released in March 2019 (Chart 1).

Chart 1: Dallas Fed's Texas Year-End 2018 Employment Data Approximate Final Benchmark

Downloadable chart | Chart data

Revising Houston job growth, statewide service sector activity

The annual benchmark affects some regions and industries more than others.

Estimates of growth in the September 2017–September 2018 period were revised significantly lower for Houston (by 43,000 jobs), Dallas (about 29,000 jobs) and Austin (about 7,000 jobs). The changes become even more pronounced for Houston when depicted in percentage terms—annual growth fell to 2.8 percent from 4.2 percent. Revisions were less severe in Dallas, down to 2.5 percent from 3.7 percent, and Austin, down to 3.5 percent from 4.2 percent.

Dallas Fed early benchmarked data in late December 2018 captured the downward revision for each metro, estimating growth during the benchmark period at 3.1 percent for Houston, 2.4 percent for Dallas and 3.6 percent for Austin. Chart 2 shows Dallas Fed data for Houston as of December 2018.

Chart 2: Dallas Fed’s Houston Employment Near Final Benchmark

Downloadable chart | Chart data

The benchmark revisions subtracted a large number of jobs from estimated growth in the service sector. Benchmark-period growth fell by more than 28,000 jobs in professional and business services, followed by 23,500 jobs in trade, transportation and utilities, 10,000 in education and health services, and 6,000 in financial activities.

In the goods-producing sector, estimated construction growth declined by more than 27,000 jobs. The Dallas Fed early benchmark captured the bulk of these revisions.

2019 job growth likely to slow

The Dallas Fed’s most recent forecast is for job growth of 1.8 percent this year. This outlook may well change as leading indicators and job estimates are updated.

Dallas Fed data produced in late May will incorporate QCEW data for the final three months of 2018. A subsequent adjustment in late August 2019 will draw on the QCEW through first quarter 2019, while the late November figures will use QCEW data through June.

Dallas Fed estimates not only will likely provide a more accurate picture of recent economic performance, they will also yield a more accurate forecast of where state and local growth is headed.

About the Authors

Keith R. Phillips

Phillips is an assistant vice president and senior economist at the San Antonio Branch of the Federal Reserve Bank of Dallas.

Alexander T. Abraham

Abraham is an economic programmer and analyst in the Research Department at the Federal Reserve Bank of Dallas.

The views expressed are those of the authors and should not be attributed to the Federal Reserve Bank of Dallas or the Federal Reserve System.