Abstract

The number of new non-farm payrolls in the United States in September was only 194000, less than the 500000 expected by the market. Faced with this non-farm report, investors may ask three questions: 1) how to view the recovery trend of the US labor market in the future? 2) the additional unemployment allowance has been stopped, why is there no significant improvement in employment? 3) how will the lower-than-expected non-farm data affect the Fed's monetary policy?

In response to the first question, we believe that the recovery of the US labour market has not been plain sailing. Although the demand for labor is still strong, employment growth has been dragged down because of supply constraints. In view of this, we believe that the market should lower its expectations for US job growth, and it may be normal for around 300000 jobs to be created in a single month.

In response to the second question, survey data show that early retirement is still the main reason for the decline in labor supply. In addition, the balance sheet of the US residential sector has remained strong since the outbreak, reducing the willingness of workers to work. After the epidemic, workers have become more selective in choosing their careers, which has also aggravated the labor shortage in some industries.

On the third question, we don't think the Fed will change its Taper plan because of this non-farm data. Under the benchmark scenario, the Fed will still announce the launch of Taper in November and December. The reason is that as supply contradictions become more prominent and inflationary pressures rise, the need for the Fed to withdraw from easing increases. We believe that if the Fed continues to hesitate, it may trigger higher inflation expectations and eventually lead to sustained inflation becoming self-fulfilling, which is an outcome that the Fed does not want to see.

Text

How to interpret non-agriculture in September?

Before the release of the non-farm report, expectations for new jobs in September were high, in part because the federal government's additional unemployment benefits had been stopped in the first week of September, and with the marginal relief of the Delta epidemic, it seemed that people should go back to work. But the data show that not only did the number of new jobs fall short of expectations, but the labor force participation rate also declined. On the other hand, hourly wages have accelerated and exceeded market expectations, indicating that demand in the labour market is not weak.

Overall, the non-farm report was a slight disappointment to the market and the Fed. The combination of employment, wage and labour force participation data shows that the recovery in the US labour market has not been plain sailing. Although the demand for labor is still strong, employment growth has been dragged down because of supply constraints. In view of this, we believe that the market should lower its expectations for US job growth, and it may be normal for around 300000 jobs to be created in a single month. Specifically:

New non-farm payrolls: 194000 new jobs were created in September, down from the previous month. In terms of sectors, 317000 new jobs were created in the private sector, the same as the previous month, while the decline of 123000 jobs in government departments was a major drag. Why is there a decline in employment in government departments? One explanation is related to the quarterly adjustment factor of the education industry. Many schools in the United States are public, and their employment is counted as a government department. School holidays and class resumption times have changed this year because of the COVID-19 epidemic. In this case, the use of conventional quarterly adjustment factors will increase statistical errors and increase the volatility of employment data. But even if you look at the private sector, there has been no retaliatory rebound in employment. In the transport sector, where labour shortages are severe, including truck drivers, railway and port workers, job creation has also slowed.

Labor force participation rate: the labor force participation rate in September was 61.6%, down 0.1 percentage points from the previous month and 1.5 percentage points lower than before the epidemic.[1]The corresponding non-labor force population (Not in labor force) increased by about 4.7 million compared with before the outbreak. In terms of age groups, the labor force participation rate of people over 55 years old was 38.6%, 1.7 percentage points lower than that before the epidemic, and the labor force participation rate of people aged 25-54 was 81.6%, 1.1 percentage points lower than that before the epidemic. Thus it can be seen that the decline in the participation rate of the middle-aged and elderly is an important reason for dragging down the overall labour force participation rate, which we think is related to early retirement. But at the same time, the participation rate of young people has also declined, indicating that there are other reasons that prevent workers from returning to work, which we will analyze below.

Hourly wage growth: hourly wages rose 0.6% month-on-month in September and 4.6% year-on-year, both rising further from the previous month and exceeding market expectations. The rise in wages shows that the demand for labor is still strong. From a sub-industry point of view, low-wage industries, such as hotel leisure, transportation and warehousing, retail wages have increased rapidly, and these industries are also industries with a serious labor shortage. In other words, if enterprises want to recruit workers, they must pay higher wages. We believe that this will increase the risk of rising inflation in the United States in the future.

Unemployment rate: the better performance in September was the unemployment rate, which fell to 4.8% from 5.2% last month. How to understand the further decline in unemployment? One explanation is that after the outbreak, due to the decline in the labour force participation rate, there is less idle available labour, making it easier for the labour market to tighten. The implication for monetary policy is that the natural unemployment rate may rise after the epidemic, making it easier for the labor market to enter a state of "overheating". If monetary policy does not respond in a timely manner, it may increase the upward risk of inflation. in turn, it will disturb the economic and financial operation.

Subsidies have stopped, why has American employment not improved?

According to the U.S. Department of Commerce, as of September 13, the proportion of people who did not work in the past week was the highest for retirement reasons, reaching 42.1%. The remaining main reasons are: suffering from other diseases other than COVID-19 (6.5%), not wanting to be employed (5.3%), taking care of children (4.9%), being laid off because of the epidemic (4.1%), suffering from COVID-19 or taking care of other COVID-19 patients (4.1%), worried about the epidemic (3.1%), and so on.

In terms of trend, the proportion of people who do not work because of retirement has continued to rise since the beginning of the epidemic (figure 7), rising to 42.1% from 31.5% at the end of April 2020, indicating that the epidemic has prompted many middle-aged and elderly people to retire early. It is worth noting that, unlike other temporary non-work, the lack of work brought about by early retirement will not be reversed by the improvement of the epidemic and may have a relatively long-term impact on the labour market. In addition, the proportion of people who do not work due to concerns about the epidemic has continued to decline since 2021, but has risen again since July as the Delta epidemic rebounded. The proportion of children who did not work decreased significantly after the start of school in July and August, but increased again in early September. We believe that this may be related to the significant increase in the number of children infected and seriously ill during the current round of Delta.

The good news is that due to the layoffs of COVID-19, or the proportion of employers temporarily closed and not working under the influence of the epidemic continues to decline, indicating that the demand in the labor market is still relatively good. This also shows once again that people who do not go to work at present may be based on active choices based on long-term consideration. In addition, the increase in the proportion of other and unexplained reasons suggests that there are other factors hindering the return of the labor force to work.

We think there are two other possible reasons: first, the balance sheet of the US household sector remained strong after the outbreak, reducing the willingness of workers to work. According to the data, by the end of the second quarter of 2021, the total net worth of the household sector in the United States was 28% higher than that at the end of the first quarter of 2020, and the total cash assets of the household sector was 2.6 times that of the first quarter of 2020 (figure 8-9). It can be said that compared with the previous two recessions, the balance sheet of American residents is very strong. The reason for this is that, on the one hand, the US government has carried out generous bailouts to the residential sector, such as cash and subsidies. On the other hand, Fed monetary policy easing has pushed up asset prices, doubling the value of households holding these assets.

Second, after the epidemic, workers have become more selective in choosing their careers, and people are no longer willing to engage in jobs with inflexible working hours and low wages, such as those in the hotel leisure industry. According to Joblist's investigation,[2]1/3 of current hotel and leisure industry employees said they were "dissatisfied" or "very dissatisfied" with their work, double the pre-epidemic level (figure 10). In view of this, 58 per cent of employees in the industry said they planned to resign by the end of 2021. About 25 per cent of former hotel leisure employees surveyed said they did not want to return to work in the industry. The main reasons include: low salary (55%, the same below), desire for a new career path (50%), lack of benefits (39%), customer unfriendly (38%), inflexible working hours (34%), worry about epidemic risk (23%), etc. (chart 11).

What is the impact on Fed policy?

We don't think the Fed will change its Taper plan because of this non-farm data. Our benchmark scenario remains that the Fed will announce the launch of Taper in November and December.

Our argument is that as supply contradictions become more prominent and upward inflation risks increase, the need for the Fed to withdraw from easing increases. Although used car prices and other previously fierce price items have fallen in the past two months, the triple impact of labor, supply chain and energy supply is becoming more and more obvious. We believe that if the Fed continues to hesitate, it may trigger higher inflation expectations and eventually lead to sustained inflation becoming self-fulfilling, which is an outcome that the Fed does not want to see.

In addition, since non-farm payrolls were revised up in July and August (a combined increase of 169000 in two months), bringing the average monthly job creation from July to September to 550000, a figure acceptable to most of the Fed. After all, judging from Powell's statement after the FOMC meeting in September, the threshold for Taper is already very low. To sum up, we don't expect the Fed to slow down the pace of Taper.

This article is selected from CICC Macro, author: Liu Zhengning, Zhang Wenlang, etc.