Dunhe Asset Management: U.S. economy returns to low growth, more indicators approaching pre-pandemic levels

Recent U.S. economic data has started to weaken across the board, with the most noticeable change coming from the previously resilient service sector. While the U.S. manufacturing PMI fell below 50 as early as last year, the service sector PMI has remained above 50. However, in April and June of this year, it fell below 50 twice, hitting a new low since the pandemic. By observing more economic indicators, it is not difficult to see that the growth rates of consumption, household income, employment, and inflation have all significantly slowed down to pre-pandemic levels, indicating that the U.S. economy is transitioning from high growth in recent years to low growth.

Firstly, the real disposable income of U.S. residents has started to significantly decrease on a monthly basis compared to pre-pandemic levels this year. The average monthly growth rate of U.S. residents' compensation in the first half of this year was 0.45%, only slightly higher than the pre-pandemic 0.36%. More importantly, the month-on-month growth rate of disposable income after tax deductions is almost the same as before the pandemic. On one hand, the government subsidies during the pandemic have been greatly reduced, and on the other hand, the tax relief effects brought by the inflation subsidy law are gradually disappearing. Furthermore, the increase in income has not kept up with the speed of price increases. After excluding the factor of price increases, real disposable income has not significantly increased since 23 years ago and is still below the historical trend line. In addition, according to statistics from the Federal Reserve Bank of San Francisco, the excess savings accumulated by residents due to government subsidies during the pandemic were already depleted by March this year Income slowing down and reduced savings have led to a significant decrease in residents' consumption capacity. Although the monthly resident consumption amount is still slightly higher than the pre-epidemic average, the growth in goods consumption has dropped to around 0%, with only service consumption significantly higher than before the epidemic. After excluding price factors, the overall month-on-month growth rate of actual retail consumption has fallen below pre-epidemic levels, with even the most resilient service consumption only slightly higher than before the epidemic.

Next, the continuously rising unemployment rate contrasts sharply with the high level of non-farm employment, indicating that the quality of employment is deteriorating, and the risk of qualitative change caused by quantitative change is increasing. Since 23 years ago, the monthly average of non-farm employment has remained above 220,000, still higher than the pre-epidemic average of 190,000 from 2015 to 2019. It seems that the US employment market has not cooled significantly, but the unemployment rate has rebounded since the third quarter of last year, currently rising to 4.1%, exceeding its 27-month rolling average level. Historically, this often corresponds to the US economy entering a recession. Why are different employment indicators so different? The unemployment rate data is based on the number of employed persons under the household survey, while non-farm employment is based on the establishment survey. The former's monthly average has been consistently lower than the latter's since the epidemic, with only two-thirds of the latter last year. In the first six months of this year, household survey employment has not shown significant growth, far below the 1.33 million increase under the non-farm employment indicator. According to the Quarterly Census of Employment and Wages (QCEW), which covers over 95% of US employment positions, the growth rate of US employment is closer to the household survey indicator. Since last year, there have been multiple downward revisions to non-farm data, indicating an exaggeration of the employment growth trend by non-farm data.

There are two reasons for this deviation: first, in the past three years, a large number of illegal immigrants have entered the US border. The quarterly wage survey data is based on employment insurance payment records and cannot cover individuals without work permits. However, the employment of this group will be reflected in the establishment survey, and in fact, for native-born Americans, the number of job positions has not significantly increased compared to before the epidemic. Instead, the employment brought by immigrants continues to rise Second, the recent improvement in employment is mainly reflected in part-time workers. Part-time work will increase employment in enterprise surveys, but not in household surveys. In the past year, the United States has lost 1.55 million full-time job positions, replaced by 1.05 million part-time positions. This means that the employment status of U.S. residents is not stable, and the risk of unemployment for part-time workers is greater compared to full-time workers.

Third, the core CPI in the United States has been below 0.2% month-on-month for two consecutive months, especially the most resilient service prices have begun to return to pre-pandemic levels. The prices of goods in the core CPI have been in a state of deflation since August 23 years ago, but the month-on-month central value of service prices has always been significantly higher than pre-pandemic levels, only falling from 0.4-0.5% to around 0.2% in May this year, while the pre-pandemic average was 0.24% The main manifestation of pushing service prices down is mainly reflected in non-housing sub-items, with the month-on-month change staying around 0 in the past two months, even lower than the pre-epidemic average level. This is a clear result of the significant easing of the labor market supply-demand imbalance, with the ratio of job vacancies to the number of unemployed people returning to a level similar to pre-epidemic levels. The University of Michigan Consumer Survey shows that consumers' inflation expectations for the next one year and five years have also begun to fall to within 3%.

The slowdown in U.S. economic growth shows that the marginal effects of fiscal stimulus have begun to diminish. As we have analyzed in previous articles, fiscal expansion, even without a hard constraint on the debt ceiling, does not mean that the economic stimulus is sustainable. The central tendency of U.S. fiscal expenditure/GDP has significantly increased after the epidemic, but fiscal revenue/GDP still fluctuates within the original range, indicating that interest payments on existing national debt are more dependent on consuming new fiscal deficits of the fiscal year. Currently, the cumulative deficit for the 2024 fiscal year has started to decrease compared to the same period in 2023, and the scale of fiscal expenditure rolling one year after excluding interest payments has returned to near pre-epidemic levels. If interest rates remain high, the proportion of interest payments to fiscal expenditure is expected to soar to 20-30% in the next 1-2 years, becoming the largest expenditure item for the U.S. government. The biggest beneficiaries of interest income are still financial institutions and wealthy individuals, which goes against the original intention of fiscal policy to provide subsidies to the middle and low-income groups through transfer payments, significantly reducing the effectiveness of economic stimulus.

The U.S. Treasury yield curve is being driven by enhanced rate cut expectations, showing a steep downward trend, but term spreads have not turned significantly positive yet, indicating that the downward trend in U.S. Treasury yields has just begun. Since July, the 10Y-2Y Treasury spread has significantly narrowed from -50bp to -12bp, breaking through the upper limit of the fluctuation range in the past 23 years. The probability of the first rate cut in September has now reached 100%, and the expected number of rate cuts for the year has risen from less than two to close to three. Historically, once a rate cut cycle begins, the 10-2 spread will mostly reach over 150bp. This means that the downward trend in yields driven by rate cuts will not end until term spreads turn significantly positive. After experiencing many unexpected events such as Trump's impeachment and Biden's withdrawal from the race, the outlook for the U.S. presidential election has become more uncertain. Democratic candidate Harris's support has significantly increased, rapidly narrowing the gap with Trump. The possibility of a Republican sweep has greatly decreased, and even Trump's chances of winning have become uncertain again. Despite Trump's promotion of tariffs, immigration expulsion, and loose fiscal policies, which may imply a resurgence of inflation, U.S. Treasury yields have continued to decline even during the period when Trump's approval rating surged to nearly 70% after the shooting incident. This indicates that the current main driver is the downward economic fundamentals rather than the more variable medium-term policy outlook. Against the backdrop of an upcoming rate cut cycle, the downward trend in short-term rates remains relatively certain. At the same time, the impact of Trump's trade policies on long-term bond yields has become limited

However, US tech stocks do not seem to have benefited from the rate cut trade, and have recently started to adjust significantly. We believe that the liquidity support of commercial bank reserves for US stocks has been weakening, coupled with the downward economic fundamentals, which could easily drive the risk premium of US stocks to rise again. Since mid-July, major US stock indices have fallen by around -3%, with the Nasdaq experiencing a maximum decline of nearly -8%. In previous rate hike cycles, even when US stocks rose, the risk premium often expanded. However, in this rate hike cycle, the risk premium has shown an abnormally sustained decline. The main reason for this deviation is the excessive liquidity, with the commercial bank reserves as a percentage of GDP, a measure of US stock liquidity, remaining at historically high levels. When this ratio exceeds 11%, the risk premium of US stocks mostly continues to decline. Interestingly, the change in this ratio does not have a clear correlation with the direction of the benchmark interest rate, but is more easily influenced by changes in the Fed's balance sheet. For example, during the rate hike cycle from 2016 to 2018, this ratio was also above the threshold, but it fell sharply during the rate cut cycle in 2019-2020, and the risk premium of US stocks also experienced a process of first falling and then rising.

! On a trend basis, even if the Fed cuts interest rates, the liquidity support for US stocks is likely to continue to weaken. Due to the impact of the US tax payment period, commercial bank reserves fell from around 36 trillion to around 33 trillion in April, but have since remained low and volatile, far below the first quarter of 24-year range. During this round of Fed balance sheet reduction, reserve sizes have not decreased but increased, mainly because the continuous release of reverse repurchase funds has offset the negative impact of balance sheet reduction on liquidity. However, the current balance of reverse repurchase is not much left, and with the Fed maintaining balance sheet reduction, the probability of further decline in reserves is high.

The Bank of Japan's initiation of balance sheet reduction may also unexpectedly impact US stock liquidity. Market expectations for the July Bank of Japan meeting rate hike probability have reached nearly 70%, while also announcing the initiation of balance sheet reduction operations. Historically, Bank of Japan rate hikes have often meant that US stocks will begin to adjust. In 2000, Japan's exit from negative interest rate policy corresponded to the bursting of the dot-com bubble, while in 2006, Japan's rate hike corresponded to the peak and subsequent decline in US house prices, triggering the subprime mortgage crisis. Over the past 22 years, global central bank total assets have begun to shrink, with only the Bank of Japan expanding its balance sheet again in the third quarter of 22 years, corresponding to the bottoming out and rebound of US stocks at that time. If the Bank of Japan also starts to reduce its balance sheet, it means that the final provider of liquidity for US stocks will also exit This article is from: Dunhe Asset Management (ID:gh_7ba8a04da885), original title: "Dunhe Market Observation | Overseas Observation: US Economy Returns to Low Growth"