The roughly 60 percent drop in oil prices to around $40 per barrel since last summer and global economic weaknesses, including less exports to China, are drags on the Texas economy. Though last week’s state-level employment report for July 2015 shows Texas’ labor market is performing better according to several employment measures, the recent steep drop in the state’s total employment is cause for concern.

Chart 1 shows the Bureau of Labor Statistics’ (BLS) measures of total employment based on the household survey and nonfarm employment based on the payroll survey typically move together. This is obvious from their 99.1 percent correlation since January 1990 when the first nonfarm employment data are available. 

Chart 1

Diving deeper into the data, Chart 2 presents monthly changes in these employment measures showing that total employment has been negative during only three periods.

Chart 2

These periods include the early 2000s during the 2001 U.S. recession, the late 2000s during the 2007 to 2009 Great Recession, and since March 2015. Nonfarm employment was also negative during these periods and others when total employment was positive. In the two early periods when both were negative, total employment rebounded more quickly.

Unfortunately, we only have access to the revised data, so these previous periods could have had a situation when total employment was negative and nonfarm employment was positive, but these data do not show that this has been the case for a sustained period since 1990.

However, Chart 3 shows this dichotomy exists this year.

Chart 3

Both measures were negative in March 2015, as total employment declined by 5,602 and nonfarm employment dropped by 25,200. But total employment has remained negative while nonfarm employment has been positive in subsequent months. Since February, total employment is down 126,348 and nonfarm employment is up 49,700 in July.

Though data aren’t readily available to determine the reason for this dichotomy, the differences between these two employment measures provide potential explanations.

The U.S. Bureau of Labor Statistics compiles state and regional labor market data from the payroll survey and household survey and explains their differences by the following:

Nonfarm employment: “The Current Employment Statistics (CES) survey, also known as the payroll (establishment) survey, excludes unpaid family workers, domestic workers in private homes, agricultural workers, proprietors, and other self-employed persons, all of whom are covered by the Current Population Survey (CPS), also known as the household survey. Moreover, the payroll survey counts a person who is employed by two or more establishments at each place of employment, but the household survey counts a person only once, and classifies the individual according to the major activity.”

Total Employment: “The household survey emphasizes the employment status of individuals and provides information on the demographic characteristics (sex, age, and race) of the labor force. The survey is not well suited to furnishing detailed information on the industrial and geographic distribution of employment. The establishment survey provides limited information on personal characteristics of workers; however, it is an excellent source for detailed industrial and geographic data.”

In addition, the household survey used to estimate total employment is based on questioning around 60,000 households nationwide to determine whether they are employed. The payroll survey used to estimate nonfarm employment is based on questioning about 143,000 businesses and government agencies nationwide to determine how many workers are employed. These results are then narrowed to Texas.

Based on these survey differences, likely factors driving the employment dichotomy include those who are self-employed, very young companies, and those in the farming sector. In particular, this could be driven by the effect of lower oil prices on rural areas such as the Eagle Ford and Permian Basin areas where oil rigs are shutdown influencing startup oil companies to lay off workers and contribute to slower output that contribute to eliminating self-employed and farming jobs. The more established companies measured in the payroll survey may be more resilient to the weaknesses in Texas but could be affected in the near future if these headwinds aren’t resolved.

This dichotomy is a new phenomenon in Texas. It will be interesting to see the result as one employment measure eventually converges with the other. It may be that one of these will be substantially revised in upcoming months and years to match the other—more than likely this would be a downward revision to nonfarm employment if the past is a good guide.

Either way, these employment measures should be watched carefully moving forward to determine the state of the economy and the well-being of Texans.