Unemployment Rate Drops: BLS Numbers Are Imprecise and Do Not Reflect Real Picture


Measuring big things and big ideas is difficult. When it can’t fit onto a scale or be assessed with a tape, you have to get creative. Throughout my own career, I learned that the best attitude with which we can hope to measure the biggies is to understand measurement is simply reducing the level of uncertainty.*

So it is today with the numbers released by the Bureau of Labor Statistics (BLS). BLS reports the national unemployment rate is now 7.8%, an improvement from 8.1% last month. BLS also reports payroll employment rose by 114,000 in September.

How does BLS arrive at these figures? How do they reduce the level of uncertainty over how many people are gainfully employed in the country? What do they tell us about government’s capabilities overall? 

Let’s tackle methodology first. The government releases two big employment surveys on the first Friday of every month. The first is the Current Employment Statistics survey, normally called the “payroll survey.” CES surveys 140,000 business and government agencies nationwide (except Puerto Rico and other U.S. territories). It notes number of employees, hours worked, and salaries earned. It also logs employment by gender and whether the positions are full or part time with some specificity. 

Also important is what the CES doesn’t count. CES omits not only contract workers – like me – but also farm workers, many of whom are migrants. Interestingly, CES also excludes workers on strike through the 12th of a month. For example, the latest unemployment number for Chicago could be nearly 30,000 people higher because the Chicago Teachers Union went on strike between September 10th and 19th.

The “household survey” (or Current Population Survey) is conducted by the BLS and the Census Bureau, together with state government agencies. The survey calls or visits around 60,000 households and asks to speak with the head of household (over 16 years old). After surveying whether a person is working or seeking work, CPS asks supplementary questions covering things such as tobacco use and voting patterns. You can even read the script for the CPS online.

It’s popular to note the household survey excludes those individuals who are unemployed yet not seeking employment. This is true. The CPS excludes those respondents “engaged in other activities” from the overall employment figures. It only counts those either employed or seeking work actively, taking people’s word for it. 

These two surveys form the basis for the numbers we’re discussing today, but government is not the only institution that counts jobs. Private efforts use different techniques to better inform businesses. The Manpower Group conducts a quarterly survey that asks a representative sampling of employers whether they intend to hire in the next quarter. Payroll process companies such as ADP also report aggregate data, as well as the Institute for Supply Management, which surveys purchasing managers at companies on whether they expect to hire more, fewer, or about the same number of people. 

These various private surveys help to form the consensus figure by which BLS reports are often measured. A popular consensus figure for September predicted between 132,000-137,000 new jobs. Despite an official gain of 114,000 jobs, because the figure is below the consensus (and an average of gain 146,000 jobs per month in 2012), some will call today’s report “disappointing.” 

That said, it’s important to note all of these numbers are subject to revision and that none of them are precise. Every figure is an estimate, and the household survey excludes either you or people you care about – the government cannot visit every home. 

That it is difficult enough to get a handle on how many people are working in the country should give pause to those who suppose the government should be empowered to predict more about the needs and wants of individuals. If simple employment surveys miss people and circumstances, imagine how boards of medical professionals will misdiagnose the needs of individual patients and how bureaucrats in the Department of Education misidentify how individual students must be educated.

Today’s jobs numbers show an improvement, which is something to cheer, but they are important mostly because they will provide rhetorical ammunition to certain political campaigns. We should each hope for lower unemployment figures every month, but we should also recognize these are estimates, imperfect and imprecise, and that partisans will use them eagerly for their own ends.

Every government policy impacts the incentives of employers, employees, and the unemployed. As we continue to dig slowly out of the trough of the worst recession since the Great Depression, we should ask what policies reduce the level of uncertainty of job seekers and job creators, and, by implication, which ones make their decisions more difficult. If you truly care about the fortunes of the unemployed, that would be the place to start.

* For more on that exact insight, check out Douglas Hubbard’s excellent book ‘How To Measure Anything: Finding The Value of Intangibles In Business.’