The "Stagnation" of Employment and the "Inflation" of Prices: The Federal Reserve's Dilemma

· Investment Highlights ·

In July, the U.S. non-farm payrolls added fewer jobs than expected, and the data for the previous two months was also significantly revised downwards, while the unemployment rate only saw a slight increase. We believe that since May, Trump's immigration expulsion policy has gradually increased its impact on the job market, and the expulsion of immigrants will significantly drag down the number of non-farm payrolls, but the impact on the unemployment rate is relatively small, which may be the key reason for the divergence between the unemployment rate and non-farm payrolls. Looking ahead, the July employment data may not be sufficient to change Powell's hawkish stance, and we need to closely observe the Jackson Hole central bank annual meeting, being wary of the risk of renewed concerns over the loss of Federal Reserve independence.

Non-farm data: The revision exceeded market expectations. The July U.S. non-farm payrolls data fell short of expectations, and what worried the market even more was the significant downward revision of the data for May and June. Historically, the extent of this non-farm data revision is quite rare, raising doubts about the quality of the data. By comparing with ADP employment data, we tend to believe that the U.S. private sector job market has indeed weakened.

Unemployment rate vs. non-farm payrolls: Why the divergence? After the downward revision, the non-farm payrolls data has significantly declined, but the unemployment rate has only slightly rebounded. We believe that the gradual effectiveness of Trump's immigration expulsion policy may be an important reason for the divergence between the two types of data. Since May, the impact of the immigration expulsion policy on the labor market has become increasingly apparent, reflected in the significant decline in the proportion of foreign-born individuals in employment and the total population, as well as a noticeable drop in the proportion of people holding multiple jobs. The expulsion of immigrants will significantly drag down the number of non-farm payroll jobs, but the impact on the unemployment rate is relatively small, leading to different performances of the two types of data.

Federal Reserve: Inflation or Employment? The current monetary policy of the Federal Reserve faces a dilemma. It is still uncertain whether the job market has substantially deteriorated, and the potential inflation risks under the influence of immigration and tariff policies cannot be ignored. We believe that the July non-farm data may not be sufficient to change Powell's hawkish stance. The upcoming Jackson Hole central bank annual meeting in August will be an important point to observe Powell's policy statements.

Risk Warning: Rumors of Trump firing Powell have resurfaced, threatening the independence of the Federal Reserve; non-farm data continues to exceed expectations and slow down.

Non-farm Data: The Revision Exceeded Market Expectations

The July U.S. non-farm data fell short of expectations. In July, the number of new non-farm jobs in the U.S. was 73,000, below the market expectation of 104,000. Among them, private sector employment increased by 83,000, lower than the market expectation (100,000), while government employment decreased by 10,000. However, what worried and puzzled the market even more was the significant downward revision of the data for May and June. After the July data was released, the number of new non-farm jobs for May and June was revised down from 144,000 and 147,000 to 19,000 and 14,000, respectively, with a total downward revision of 258,000 for the two months. This revision nearly erased the increase in non-farm employment for May and June and caused the average number of new non-farm jobs in the U.S. to significantly drop to 35,000 over the past three months, compared to an average of 147,000 before the revision for April to June. !

Historically, the magnitude of the downward revision of this non-farm data is relatively rare. While it is common to see significant downward revisions during the first correction due to delays in questionnaire collection, it is less common for both the first and second corrections to be significantly downward, occurring only during the COVID-19 pandemic.

Why was there a cliff-like downward revision of the non-farm data? The BLS explained that this was influenced by supplementary corrections from new questionnaires and seasonal factor adjustments. Due to delays in feedback from business surveys, the BLS revises previous data upon receiving new questionnaire data. Especially since the pandemic, the phenomenon of delayed feedback from business questionnaires has become increasingly evident, leading to more frequent downward revisions of non-farm data.

The delay in questionnaire collection may not be the main reason for this data revision. The downward revisions for new jobs in May and June were 125,000 and 133,000, respectively, both at historically high levels. However, the initial questionnaire collection rates for May and June were 68.4% and 59.5%, not significantly lower than the average level since the pandemic. Additionally, the non-farm new employment data for May was slightly revised upward during the first correction in June, at which point the second survey collection rate for May had reached 92.3%. Therefore, seasonal factor adjustments and other influences may be the main reasons.

Is there data manipulation, or is the number of new jobs indeed weakening? The cliff-like drop in new employment data has raised market concerns about data quality. What is the true state of the U.S. job market? We use ADP employment data for verification and tend to believe that the number of new private sector jobs in the U.S. has indeed weakened.

Before 2022, the purpose of compiling ADP employment data was to track and predict the non-farm employment data published by the BLS. However, after 2022, the methodology for calculating ADP employment data underwent significant adjustments, no longer aiming to predict non-farm employment but instead focusing on more independently measuring the state of the U.S. private sector job market. Therefore, ADP data can provide us with an additional independent perspective on the U.S. job market situation. Since ADP data only includes private non-farm sectors and does not encompass government employment, we only compare private employment situations.

From a historical trend perspective, the revised private non-farm new employment numbers since 2022 have maintained a good fit with ADP new employment numbers. Even with the significant downward revision of the non-farm data, it still shows a high consistency with ADP data, which may indicate that the revised data does not significantly underestimate the true state of the private sector labor market. In contrast, the initially published values for non-farm private new employment have consistently been overestimated since the pandemic

In addition to the significant decline in non-farm payroll employment, there are other hidden concerns in the U.S. labor market. On one hand, the proportion of long-term unemployed individuals is rising. In July, the proportion of those unemployed for 27 weeks or more in the U.S. increased further to 24.9%. The increase in long-term unemployment may reflect the current difficulty in finding jobs in the labor market.

On the other hand, the distribution of non-farm payroll employment across industries is uneven. Currently, new jobs are mainly coming from the education and healthcare sectors, while new employment in other industries has continued to decline in recent months and remains at a low level. This also reflects that the foundation of current employment growth in the U.S. is not broad and has certain vulnerabilities.

Unemployment Rate and Non-Farm Payroll Employment: Why the Divergence?

While non-farm payroll employment has clearly weakened, the unemployment rate has remained relatively stable. In July, the U.S. unemployment rate rose from 4.1% in June to 4.2%, with a modest increase that aligns with market expectations. So, how should we understand the divergence reflected between non-farm payroll employment and unemployment rate data?

First, it is important to clarify that non-farm employment data and unemployment rate and labor participation rate data come from different sources, and there are differences in statistical criteria. New non-farm employment data comes from the Current Employment Statistics (CES) survey, while the unemployment rate and labor participation rate come from the Current Population Survey (CPS), and there are many differences in statistical criteria and methods. For example, the CPS focuses on counting the number of employed individuals, counting those with multiple jobs only once. In contrast, the CES focuses on counting the number of jobs, counting individuals with multiple jobs multiple times.

The gradual effectiveness of Trump's immigration expulsion policy may be an important reason for the divergence between these two types of data. During the Biden administration, a large influx of immigrants into the U.S. drove strong job growth. However, this year, Trump has been intensifying efforts to expel immigrants. For example, in June, the U.S. Immigration and Customs Enforcement (ICE) collaborated with multiple law enforcement agencies to conduct a large-scale operation to capture illegal immigrants in Los Angeles County, California.

Since May, the impact of immigration expulsion policies on the labor market may have become increasingly evident. In July, the U.S. labor participation rate fell by 0.1 percentage points to 62.2%, marking a decline for three consecutive months. Among these, the labor participation rate for whites remained relatively stable, while the labor participation rate for Black or African American individuals declined significantly. Additionally, the proportion of foreign-born individuals in the total employment population has also seen a noticeable decline. Furthermore, due to immigrants often holding multiple jobs, the significant drop in the proportion of multiple jobholders in the U.S. in July may also reflect the impact of Trump's immigration expulsion policy.

Expelling immigrants may significantly drag down the number of non-farm payroll jobs, but the impact on the unemployment rate is relatively small. On one hand, the expulsion of immigrants directly leads to a decline in job positions, but since immigrants leave the U.S. labor market, the labor supply decreases, making the impact on the unemployment rate less pronounced On the other hand, since immigrants often hold multiple jobs, an immigrant with several jobs may be counted multiple times in the non-farm payrolls, while the Current Population Survey (CPS) only counts them once. Therefore, when immigrants are deported, the impact on the non-farm job creation numbers may be greater than the impact on the unemployment rate.

Deportation policies not only reduce the number of new jobs but also increase inflationary pressures. Since immigrants typically occupy low-wage positions, employers may have to hire more expensive employees after deportations, which could drive up wage growth.

Federal Reserve: Which to choose, inflation or employment?

After the release of non-farm data, market expectations for a rate cut by the Federal Reserve in September have significantly increased. Previously, due to Chairman Powell's hawkish statements during the July meeting and the unexpected rebound in the second quarter GDP growth in the U.S., the market had believed that the probability of the Federal Reserve not cutting rates in September was relatively high. However, after the substantial downward revision of the non-farm data, the market's expectation for a rate cut in September rose sharply from 43% to 80.3%. Driven by the warming expectations of a rate cut, asset prices reacted strongly; following the data release, the Nasdaq index fell by 2.2%; the U.S. dollar index weakened, dropping by 1.4% to 98.69; the yield on 10-year U.S. Treasury bonds fell sharply by 14.6 basis points to 4.22%, showing a pattern of falling stocks, rising bonds, and a weak dollar, while the spot price of gold in London also surged by 2.2%.

Currently, the Federal Reserve's monetary policy faces a dilemma. The weakening non-farm employment data seems to signal a clear slowdown in the labor market. However, the potential inflation risks under the influence of tariffs and immigration policies cannot be ignored. Overall, compared to when the Federal Reserve began cutting rates in September 2024, the six-month annualized growth rate of the core PCE price index has not shown a significant decline. Additionally, under the influence of tariffs, the growth rate of the core goods PCE price index continues to rise, and with the further transmission of tariff costs, this upward trend may continue in the second half of the year. Moreover, if the effects of immigration policies continue to manifest, the loss of low-wage labor may also increase the pressure for sustained wage growth We believe that the July non-farm payroll data may not be sufficient to change Powell's hawkish stance. On one hand, the unemployment rate of 4.2% remains stable, and the number of initial jobless claims is still robust, allowing the Federal Reserve to wait another month to observe whether the labor market shows signs of sustained deterioration. On the other hand, the extent of the impact of seasonal adjustments on new employment remains to be seen. The upcoming Jackson Hole central bank annual meeting in August will be an important point for observing Powell's policy statements.

However, if Powell continues to maintain a hawkish stance, there is a risk of renewed concerns about the independence of the Federal Reserve.

Risk Warning: Rumors of Trump firing Powell have resurfaced, posing a threat to the independence of the Federal Reserve; non-farm payroll data continues to exceed expectations and slow down.

Author: Liang Zhonghua, Wang Yuqing from Cathay Securities, Source: Liang Zhonghua Macro Research, Original Title: "The 'Stagnation' of Employment and the 'Inflation' of Prices: The Federal Reserve's Dilemma"

Risk Warning and Disclaimer

The market has risks, and investment should be cautious. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their specific circumstances. Investment based on this is at one's own risk