Analysing the tightness in the US labour market
We analyse the stress in the labour market of the largest economy in the world: are there more companies wanting to hire or fewer workers? Most importantly, how can job supply and demand be rebalanced?
The unemployment rate in the US stood at 3.4% in January, a level not seen since the 1960s, while job creation in the same month reached more than 500,000 jobs, a rate twice the average observed between 2015 and early 2020. Moreover, based on the data available up to December, in the US there are 1.9 job vacancies for every unemployed person. This is well above the historical average of 0.65 and demonstrates just how tight the world’s largest economy’s labour market is.
When we say there is stress in the labour market, we are referring to the fact that there are many more companies seeking to hire workers than there are people looking for jobs. The first question we must ask ourselves is thus whether this post-pandemic tightness is the result of higher job demand (more companies wanting to hire) or a fall in supply (fewer workers). The sum of the total number of employees plus the unfilled job vacancies, which serves as an indicator for total job demand, is not far from its historical trend, suggesting that we are not facing a demand shock. On the other hand, if we perform a similar calculation for the supply of jobs,1 we see that the pandemic did indeed trigger a major supply shock, as there are now around 4.5 million fewer workers than would be expected.
There are two factors that can explain most of this lower job supply: lower immigration and a reduction in the labour participation rate. The first factor explains approximately 1.2 of the 4.5 million fewer workers, as since the beginning of 2019 a change is apparent in the data regarding the number of people born outside the US who are seeking work in the country. If the growth trend of this population segment observed between 2015 and 2019 had continued, then by the end of 2022 there would have been just over 47 million people born outside the US over 16 years of age, rather than the 45.9 million reported by the BLS in January. The second factor is the fall in the labour participation rate – i.e. the number of people over the age of 16 who either work or would like to work – from 63.3% before the pandemic to the 62.4% registered in January, which equates to some 2.7 million fewer workers.
- 1. In order to forecast the level of employment which we would “expect” to see, we use the latest US population estimates (2017) to predict the population over 16 years of age and assume constant unemployment and labour participation rates (3.5% and 63.3%, respectively) since the beginning of the pandemic.
At the height of the pandemic, the decline in this rate was particularly pronounced and was attributed, in part, to temporary factors such as the fear of working in jobs more exposed to human contact or the fact that many people were unable to combine work with caring for minors or elderly people.2 Nevertheless, these factors should no longer play a significant role. In fact, the factor that is now driving this lower rate is, above all, the departure of the population over 55 years of age from the labour market. The changes in composition within this (large) age range have been particularly pronounced between 2016 and 2022, as the relative weight in the labour market of the population between 55 and 59 years of age (whose labour participation rate is much higher than that of the population over 60) has fallen.3
On the other hand, it is important to note that this fall in the number of workers has occurred in the extensive margin of the labour supply, that is, the number of people who work. But at the same time as the number of people who are in the labour market is falling, those who remain active in it are working fewer hours, thus reducing the intensive margin. This phenomenon has been named quiet quitting and is due to the desire of many workers to strike a better work-life balance. Thus, as can be seen in the third chart, according to the data from a recent investigation, the reduction in the hours worked between 2019 and 2022 in the US is explained by a reduction in both the extensive and the intensive margin, with the intensive reduction being most notable among men with university studies and between the ages of 25 and 39.4
- 2. This phenomenon, which was particularly intense in 2021, was named the great resignation. For further details, see «US: (in)complete recovery of the labour market» in the MR07/2021.
- 3. In fact, the change in composition within the population over 55 accounts for 2.4 million of the 2.7 million fewer workers discussed above.
- 4. See D. Lee et al. (2023). «Where Are the Workers? From Great Resignation to Quiet Quitting». National Bureau of Economic Research, No. w30833.
This mismatch between the supply and demand for jobs is leading to wage dynamics which are not exactly helping the Federal Reserve to bring inflation back down to its 2% target. In order to fill their vacancies and attract workers, businesses must offer higher wages, which in turn drives up inflation. For this reason, through its rate hikes, the Fed is trying to rebalance the supply and demand for jobs and hopes to do so without triggering an unemployment rally: monetary tightening alone will not encourage workers who are outside the labour market to join it, but it should discourage businesses from hiring more people, thereby reducing the number of job vacancies.
So will the Fed manage to reduce the job vacancy rate without necessarily increasing unemployment? The answer is not clear, and the conclusion of the debate over whether there will be a recession in the US will largely depend on how this episode unfolds. There are arguments in favour of both outcomes: on the one hand, the Beveridge curve, which links the vacancy rate to that of unemployment, is not linear and, provided the historical relationships are maintained, at the point at which it currently lies there should be some scope to reduce the number of job vacancies without an excessive rise in unemployment.5 In addition, given the difficulties in filling vacancies in recent years, many firms may choose not to destroy jobs if they fear that they will find it difficult to find replacement workers later on.6 On the other hand, job vacancies are not evenly distributed across the different US economic sectors. In fact, vacancies are highly concentrated in accomodation and food services sector (precisely where the hourly wage is lower). Thus, if we assume that the Fed’s rate hikes will cool demand (and thus destroy job vacancies) to a relatively similar extent across all sectors, then it could lead to job destruction in the rest of the sectors and thus increase the unemployment rate.
- 5. See B. Bok et al. (2022). «Finding a Soft Landing along the Beveridge Curve». FRBSF Economic Letter, 24.
- 6. In fact, the employment report for January documented fewer redundancies than usual for this month of the year, which could be explained by companies’ fear of difficulties in finding workers in the future.