Unemployment nuances, NFP data and signaling quality

Yesterday the initial jobless claims number was released in US, which will be followed by the NFP data today. Initial claims which?comes out weekly represents the number of people who went out of their active employment and ended up filing for jobless claims during the week. The number was a pandemic era low at 340k.The NFP data expectations are closing in around 720k to 740k as per different estimates. Unemployment rate is also expected to come down to 5.2% from 5.4%. Today's NFP report gains significance at is the last job report before the September FOMC. Given the mighty focus and lengthy debates on employment dynamics, a stellar number can move the taper start deadline to October itself. Right now the market is expecting that FOMC will indicate a slight taper that too in December or January. Lift off that is the raising of rates is way too in future. As per the Fed funds futures quoted on CME the first bets of any rate hike start appearing in the January 2023 contract, whether that positioning also gets changed post NFP will be interesting to see.

The start of the month is the time of PMI data points to get released across the world. Today the Chinese PMI data came out which showed a fall in business activity during the month of August. The resurgence of the delta variant was cited as the reason but the survey respondents looked upbeat about the year ahead outlook. The PMI indicators are generally considered a leading indicator of economic activity as the purchasing managers are the ones who are thought to be with the pulse on the ground and eye on the future. India services PMI is also due to be released today.

Now coming back to the employment debate. The debate is much more nuanced then what can be summarised by one unemployment number figure.As readers can see that the unemployment number is a percentage, it depends on how we define the numerator and the denominator. In US, BLS ie Bureau of Labor statistics surveys about 60000 completely random households, each person who is 16 years or above in the selected households is put into one of 3 categories, number one is employed (working full time or part time during the previous week), second is unemployed?( is unemployed but is making efforts to find work) and then final category is "out of the labor force". A person out of the labor force is the one who did not work and also did not look for work in the past four weeks. People like full time students, unpaid homemakers, retirees are generally out of the labour force.

The labor force is defined as summation of category one and two ie Employed and Unemployed. The unemployment rate is then defined as number of unemployed as a percentage of the total labor force. Participation rate is defined as labor force as a percentage of total working age population. Readers can see above that in calculation of unemployment rate the number of people who have opted out or are not looking out for a job also plays an important role. This opting out can be voluntary or involuntary. During the pandemic times the risk of contagion made many people sit at home, they believed the risk of going out was too much, so these people just quietly fell off from the unemployment number. The parents who had to compulsorily be at home because the support system of creches and child care home collapsed also fell off the statistics. As the support services open these people will come back to the labor force and make the unemployment number look puzzling.?

As we have written previously that the markets however are generally impatient to assimilate these nuances, they jerk their knee on the basis of the headline number. The further details then get priced in slowly. But when the FOMC decision makers sit all this gets debated at length. We have seen in the previous minutes that not only general population participation rate is discussed, they also debate on how the participation is across different races. They also need to ensure that their decisions comply with the test of fairness and equality. As per me though this is a noble endeavor it muddles the signaling quality of data points.