The largest downward revision in 15 years: Preliminary downward revision of 818,000 in the US employment population for the past year until March

On Wednesday, August 21, before the release of the minutes of the July FOMC meeting by the Federal Reserve, the Bureau of Labor Statistics (BLS) of the U.S. Department of Labor announced the preliminary revised data for non-farm employment for the past year ending in March.

The data shows that from April 2023 to March 2024, the preliminary revision of the increase in non-farm employment in the U.S. over the past year was significantly reduced by 818,000 people, resulting in a growth rate of 1.3% for the past year ending in March of this year.

Wall Street News has mentioned that if the downward revision of employment exceeds 501,000, it will be the largest downward revision adjustment in fifteen years. Goldman Sachs has also pessimistically estimated that the highest downward revision could reach 1 million people, setting a historical record.

According to media summaries, the latest data indeed shows that the revised total employment for the year before March is -0.5%, the largest revision since 2009. Previously, mainstream economists expected the downward revision to be 500,000 people.

In other words, the actual non-farm employment growth in the U.S. from April 2023 to March 2024 was nearly 30% lower than the initially reported 2.9 million people, averaging about 242,000 net job additions per month during this period, which suddenly changed to an average of 173,500 new jobs per month.

Commentary: The annual employment growth rate in the U.S. until March is not as strong as originally reported, and it does not rule out Powell hinting at a significant rate cut

The consensus on Wall Street is that this indicates that the U.S. labor market is not as robust as imagined, and the cooling of the job market may last longer or continue for a longer period, with the data possibly affecting the tone of Federal Reserve Chairman Powell's speech at the Jackson Hole Global Central Bank Annual Meeting on Friday. The market is also concerned that the Fed's rate cut action may lag significantly behind, as the labor market is not as resilient as previously reflected in the data.

Renowned financial journalist Nick Timiraos, known as the "New Fed News Agency," commented that the annual employment growth rate in the U.S. until March is indeed not as strong as originally reported, reducing the monthly average of new jobs during the period from 246,000 to 68,000.

LPL Financial's Chief Global Strategist Quincy Krosby stated that Powell has mentioned that the Fed is closely monitoring the labor market, looking for signs of deterioration, and is ready to intervene when necessary. If the revised report shows that the number of new job additions is significantly lower than the data initially announced in the monthly non-farm report, Powell may highlight this concern in his high-profile speech on Friday.

Analysts also mentioned that on Tuesday, U.S. bond yields fell across the board by more than 5 basis points, with the more rate-sensitive two-year yield experiencing a relatively larger decline and breaking below the 4% level, leading to a cumulative decline of nearly 34 basis points so far in August. This indicates that "short-term bond yields reflect a more severe U.S. economic situation than risk assets or credit spreads," and does not rule out Powell hinting at a 50 basis point rate cut in September In the latest market reaction:

The 2/10-year US Treasury yield rose and then fell back, with the yield curve flattening in the United States after the non-farm payroll revision was announced. The 10-year benchmark bond yield hit a daily low of close to 3.78%, but rose to a daily high of 3.8294% at 22:00 Beijing time when the revised data was scheduled to be released. The two-year yield hit a daily low of 3.9263%, but also rose to 4.0041% at 22:00. The 2/10-year US Treasury yield spread rose by more than 3.14 basis points, hitting a daily high of -14.488 basis points.

The trend of US stock indices first rose sharply to a daily high, then significantly fell back to a daily low, with the S&P 500 and Nasdaq approaching a reversal, while the Dow directly reversed. One hour after the data was released, the US stock indices basically returned to a slight uptrend before the data release, but the Dow's uptrend was noticeably weaker than at the opening. The Russell 2000 small-cap stocks, which are more sensitive to the economic cycle, maintained a 0.6% increase, leading the way.

In addition, the Chinese concept stock index expanded its gains, rising 2.3% within an hour after the data was released, significantly outperforming the broader US stock market.

The largest downward revision comes from the private sector, with the most significant declines in professional and business services, while government and certain private sectors see job growth

Specifically, looking at the preliminary revisions to the annual benchmarks mentioned above, most of the downward revisions in new job numbers come from the private sector, decreasing by 819,000 people compared to the initial estimate. Meanwhile, the government sector, which serves as the engine of job growth from April 2023 to March 2024, saw a net increase of 1,000 jobs.

Within the private sector, the industry with the largest downward revision in annual employment is "professional and business services," which decreased by 358,000 people compared to the initial estimate. This is followed by a decrease of 150,000 jobs in the leisure and hospitality industry, 115,000 jobs in manufacturing, and 104,000 jobs in trade, transportation, and utilities. Among the trade category, retail trade saw a decrease of 129,000 jobs.

Furthermore, in the few areas where the number of recruitment has been revised upwards, employment in private education and health services increased by 87,000 people, transportation and warehousing increased by 56,400 people, and other services increased by 21,000 people.

Public information shows that the US Bureau of Labor Statistics releases monthly non-farm payroll data based on the Current Employment Statistics survey, which covers about one-third of non-farm employers in the United States.

However, more comprehensive employment data is released with a delay, including the Quarterly Census of Employment and Wages (QCEW), which requires employers to submit data used for unemployment benefits to their respective states. This dataset is more extensive, covering the vast majority of US employers In order to incorporate these additional employment data to more accurately reflect the employment situation, the U.S. Bureau of Labor Statistics will release revised preliminary values for non-farm employment benchmarks in the fall each year, and then complete the actual revision of the 2024 data in the January non-farm employment report released in February of the following year. However, typically, the revised preliminary values are quite close to the final benchmark revisions.

Data Release Side Note: Investors Highly Focus on the U.S. Labor Market, Triggering High Volatility in Assets such as Stock and Bond Markets

Analyst Stephen Stanley from Santander Bank recently stated that negative revisions to the total non-farm employment from April 2023 to March 2024 "may not necessarily tell us the current state of the labor market," and even if significantly revised downward, the average employment growth remains robust, downplaying the impact of Wednesday's revised values and "simply inferring that the following months since March will follow the same pattern."

Analysts today also mentioned that even after revisions, the number of new jobs added in the past year up to March of this year still exceeds 2 million.

However, it is undeniable that investors are currently overly focused on U.S. labor market data. The disappointing U.S. July non-farm employment report released on August 2 directly led to a sharp drop in European and American stock markets similar to "Black Monday." Subsequently, better-than-expected weekly U.S. jobless claims data seemed to suddenly ease concerns about the labor market conditions, leading to an eight-day consecutive rebound in the U.S. stock market.

Therefore, a significant downward revision of non-farm employment like today may reignite concerns about the U.S. job market softening too quickly, and elevate discussions on the necessity for the Federal Reserve to significantly cut interest rates by 50 basis points in September. Many speculate that Wednesday's revision may change expectations, prompting Fed officials to reassess the depth and breadth of rate cuts, with initial signs possibly appearing in Powell's speech.

In fact, traders increased their bets on a Fed rate cut on Tuesday, expecting a total of 100 basis points of cuts by the end of the year, compared to less than 100 basis points expected before this week. The FedWatch tool showed that the probability of a 50 basis point rate cut in September increased slightly from 29% a day earlier to nearly 31% after the data release.

Despite different interpretations of the data itself, Wall Street seems to collectively agree that this data will trigger high volatility in assets such as stock and bond markets.

Tom Essaye, founder of Sevens Report Research, stated that the market is very sensitive to soft labor market data. Based on recent data releases and market reactions, although investors can overlook most disappointing data, they have not ignored soft labor market data. If such revisions are worse than expected, then the stock market will come under pressure.

There was also a small incident today, where the revised data was originally scheduled to be released at 10 a.m. Eastern Time, but was delayed until almost 10:30 a.m. before being officially announced. Even before the update on the U.S. Department of Labor's official website, information about a downward revision of 818,000 jobs had already spread on social media platforms Forexlive, a financial website, cited a report stating that someone claimed to have personally called the US Bureau of Labor Statistics, and they were told this number over the phone. Although it is questionable that the US Bureau of Labor Statistics informed this data over the phone without officially publishing it on their website, the data ultimately matched the official website's data