You may be wondering how we could have such a good jobs report (i.e. number of jobs added) and yet have no change in the unemployment rate. The addition of 321,000 jobs is the highest level of jobs creation since January 2013. So, how can we create so many jobs and not lower the unemployment rate? Today’s (12/5/14) employment report highlights the fact that the employment report is actually a compilation of two separate employment surveys.
The Bureau of Labor Statistics calculates the number of jobs added by collecting data from the payroll records of a sample of businesses. This survey is called the Establishment Survey. Its purpose is to provide data on employment, hours worked, and earnings of workers in non-farm jobs. This is the survey used to report the number of non-farm payroll jobs added each month.
The Census Bureau conducts a monthly survey for the Bureau of Labor Statistics in order to provide data on the labor force, employment and unemployment. This is the survey used to calculate the unemployment rate each month.
Because they are two different surveys, they will not always give congruent results. Today’s employment report is a great example of this divergence. Even though the Establishment Survey showed strong jobs creation, the Household Survey showed only 4,000 people found work in November and 115,000 more people were added to the unemployed status. Why do these surveys vary so much? Let us look at the features of each survey.
The Household Survey survey is a sample of 60,000 households, which means that it incorporates less than 1% of the total labor force. The Establishment Survey pulls data from 144,000 companies in approximately 554,000 individual worksites. According to the Bureau of Labor Statistics, the Establishment Survey covers approximately 1/3 of the total non-farm payroll population. As a result, its margin for error is far smaller than the Household Survey.
When Survey is Conducted:
The Household Survey is conducted in the calendar week that includes the 12th day of the month. The Establishment Survey pulls data from the pay period that includes the 12th day of the month; this may or may not be a calendar week.
Who is Included in the Survey:
The Household Survey is designed to reflect the entire civilian population. According to definitions supplied by the Bureau of Labor Statistics, it includes agricultural workers, self-employed workers whose businesses are unincorporated, unpaid family workers and privvate household workers. All of these categories are excluded from the Establishment Survey. The Household Survey includes people who are employed but on unpaid leave while the Establishment Survey excludes these workers.
How The Surveys Count Jobs:
The Household Survey only counts an individual once. So, if an individual got three part time jobs, the Household Survey only counts that as one job. The Establishment Survey counts each job separately.
Level of Detail:
The Household Survey provides breakdowns by age, race, sex, and level of education. The Establishment Survey does not provide this level of detail.
Each survey has its positive and negative attributes, but because they are pursuing different objectives, it is natural that they may diverge on a monthly basis.Overall, each survey serves a useful purpose and the monthly divergences between the two reports tends to smooth out once the annual revisions occur. Even though the unemployment rate may appear out of sync with the number of new jobs added logic and math argue that the unemployment rate will decline if that pace of jobs creation continues.
One word of caution:
As I have discussed in my Economic Updates newsletter-to subscribe to this newsletter, click on the Sign Up button on the right-it is distinctly possible that, as the jobs market improves, the unemployment rate might initially rise before declining. This is due to the fact that some of the decline in the unemployment rate during this recovery was due to people giving up looking for work. Once you stop looking for work and drop out of the labor force, you are no longer considered unemployed. Now, as the jobs market continues to improve, many of those people may decide to resume looking for work. Once they resume looking for work, they are added back into the labor force and counted as unemployed, until they find a job. So if the pace of people entering the labor force is faster than the pace of people finding work, the unemployment rate will rise.
Today’s employment report was clearly good news overall and shows that progress continues to be made. If this continues, wage increases will be the next process that starts. There is no direct correlation between jobs creation and when wage increases start due to Corporate America adapting to changing regulations, taxes and expenses. Corporate America is focused on efficiency, so historical time tables for wage increases may no longer apply but the process should continue: steady jobs creation should ultimately lead to wage increases.
Steve Scranton is the Chief Investment Officer and Economist for Washington Trust Bank and is a CFA charter holder with over 30 years of investment experience with equities, tax-exempt and taxable fixed income securities. Steve actively participates on committees within the bank to help design strategies and policies related to client and bank owned investments. Steve also serves as the economist for the Bank and has been a featured speaker for both client and professional organization events throughout the Northwest.