This article is from Gelonghui's column: Tianfeng Research Author: Sun Binbin's Team

summary

The market may have an idea that the US recession is on the left side of the deal.

Looking from the bottom up, there are many details worth paying attention to in the US recession signals.

The first is the Sahm rule triggering.

How much was the impact of the hurricane? The historical rule is that there is no clear correlation between hurricanes and temporary unemployment, and there is even a tendency to reduce temporary unemployment. Instead, it is triggered by successive Sahm rules and the US recession tends to be accompanied by a rise in temporary unemployment. Combined with Sahm's own recent response and the easing of illegal immigration in the US, triggering the Sahm rule does not necessarily mean that the US economy has declined, but corporate recruitment demand may indeed be weakening marginally.

The second is the inversion of the US Treasury yield curve.

Historically, the inversion of the yield curve led the NBER inflection point for 5-18 months; when the inversion peaked, it was up to 15 months ahead. The current round of inversion of the US bond yield curve began in July 2022, 24 months ago; it peaked in July 2023, 12 months ago. Judging from the historical lead, it seems impossible to say that the inversion of the US bond maturity spread forecast for the recession must have failed.

The difference in this round of inversion is the deep inversion of inflation expectations, which has an impact on yield patterns. The underlying background is that early monetary and fiscal easing and fiscal contrarian expansion under high interest rates may have delayed the deterrent effect of high interest rates. In other words, the US economy may have already entered recession, but early income subsidies and countertrend fiscal expansion are equivalent to having carried out countercyclical stimulus and cross-cycle management ahead of schedule. On the one hand, the risk of economic recession is hedged at the cost of short-term inflationary resilience. On the other hand, supply-side policy orientation can stabilize long-term inflation expectations and prevent them from becoming anchored. Furthermore, it should be noted that in the course of the US economic downturn bottoming out in history, term interest spreads will indeed correct and rise early.

As reflected in the NBER reference index, recent revenue, consumption, and sales performance has been relatively stable. The weak performance is mainly employment and production-related data. The current problems facing the US economy are probably mainly due to the fact that companies' production investment and expectations may begin to change, leading to weakening labor demand.

Currently, different leading indicators are showing conflicting signals, which also means that the US economy may be in a complex situation where factors of resilience and recession are intertwined.

From the top down, the US economy may have three possible paths.

We tend to think that the US recession deal should consider three directions: upward is no landing and cross-cycle; downward is error correction in the development path.

The middle scenario is a normal economic landing. With cross-cyclical support from countertrend fiscal expansion, the momentum of the US economy may slow more slowly than in history. Mechanistically, the process of economic recession may first be transmitted from enterprise production investment to employment, which in turn affects residents' income expectations and consumer confidence, as well as the entire economic cycle. Furthermore, it is also important to note that although the US private sector leverage ratio experienced a decline during the interest rate hike cycle, it is still at a historically high level. As the economy cools down, local risk events may occur, such as risks for small to medium banks and falling asset prices.

Regarding medium- to long-term factors, we tend to think that, on the one hand, the international order and dollar system may be slow variables, and it is less likely that they will “break” beyond expectations in the short term. On the other hand, if technological changes in the industry are not realized beyond expectations, asset prices driven by earlier expectations may be adjusted in stages. To some extent, it is possible to compare the US economic and financial performance from 1999 to 2001 and the macro background of the Federal Reserve's first interest rate cut, as a frame of reference for understanding the path of subsequent US economic and financial changes. The variable is that technological changes in the industry may indeed be realized beyond expectations.

The US economic data for July has once again raised expectations of a recession in the US economy. Since 2023, the market has first discussed whether the US economy will have a “hard landing” or a “soft landing”. As the 2023Q3 and Q4 economy's resilience exceeded expectations, the market began discussions on whether it would “not land,” and then until the recent weakening of data and further adjustments in expectations, the market may have an idea on the left side of the transaction.

1. How do you define recession?

According to the NBER's definition of the US business cycle, a recession involves a sharp decline in economic activity. This decline will spread throughout the economy and continue for several months or more. The economy is thought to have entered a recession phase usually the first month after the peak.

The NBER definition mainly focuses on three criteria: depth, diffusion, and duration. When one standard shows extreme performance, even if the performance of the other standards is poor, it is considered to have entered a recession. For example, when economic activity peaked in February 2020, the NBER believes that the subsequent decline in activity was so large and spread across the economy that it should be classified as a recession, even if the process was quite short.

The NBER mainly relies on a set of indicators to determine economic peaks and troughs, including actual personal income minus transfer payments (PILT), non-farm employment, employment measured by household surveys, actual personal consumption expenditure, wholesale and retail sales adjusted for price changes, and industrial production. The NBER does not give fixed quantitative rules on how to define economic inflection points based on indicators. However, the NBER also emphasized that the two indicators they pay the most attention to are actual personal income minus transfers and non-farm payrolls.

Furthermore, financial media often define a recession as two consecutive quarters of decline in real GDP. In fact, most recessions defined by the NBER include two or more consecutive quarters of decline in real GDP, but there are exceptions, such as 2001.

As the US non-agricultural data for July fell short of expectations, the market may have an idea that the US economic recession is on the left side of the deal.

2. A bottom-up perspective: How do you view the signs of the US recession?

2.1. What do you think of the Sahm recession index?

2.1.1. Did the hurricane cause temporary rise in unemployment and is the Sahm indicator overestimated?

According to the Sahm rule, when the three-month moving average of the national unemployment rate (U3) rises 0.5 percentage point or more from the minimum of the three-month average of the previous 12 months, it may mark the beginning of a recession.

Regarding the July Sahm recession indicator of 0.53% surpassing the empirical standard of 0.5%, one current understanding is that US Hurricane Beryl caused an abnormal increase in temporary unemployment in July (July), which may cause disturbances in the unemployment rate.

The most immediate answer to this question is the BLS announcement on August 2: “Hurricane Beryl made landfall on the central coast of Texas on July 8, 2024, and reference periods for household surveys and institutional surveys were ongoing at the time. Hurricane Beryl had no significant impact on national employment and unemployment data for July...”

This conclusion is in fact in line with historical statistics. Judging from the past, there is actually no clear correlation between hurricanes and temporary unemployment, and there is even a tendency to reduce temporary unemployment.

We believe that logically, a hurricane should be a temporary, seasonal disaster, and if the business is expected to operate well, it probably won't lay off employees as a result. Demand for infrastructure maintenance and housing repair in areas affected by the hurricane may also increase intensively after the disaster.

Furthermore, since the pandemic, there has been no change in the standard weekly working hours included in employment in the US, so it seems difficult to say that weather factors have had a greater temporary impact on short-term part-time jobs/gig workers in the US than in history.

On August 16, the BLS will release state-level unemployment statistics. At that time, the impact of the hurricane on unemployment will be more clear.

2.1.2. Can a temporary rise in unemployment prove that the Sahm rule is invalid?

Historically, the Sahm rule recession index hit the 0.5% level 9 times in total. We can observe four phenomena:

First, when the indicator broke through the 0.5% “collision line” 9 times, temporary unemployment also tends to rise.

Second, it “hit the line” 9 times, and 100% of the economic recessions occurred.

Third, the Sahm rule recession index has a good trend. It continues to rise every time it breaks 0.5% in history.

Fourth, the SAHM rule recession indicators are generally 2-8 months behind when compared to the peak inflection point of the NBER economy.

As can be seen from this, the temporary rise in unemployment cannot at least directly explain the invalidity of the Sahm rule. Also, judging from the relationship between temporary unemployment and GDP, when the US economy is in recession, it tends to be accompanied by an increase in the number of temporary unemployed people.

2.1.3. How should the Sahm rules be understood in the post-pandemic era?

At the press conference after the July interest rate meeting, Powell emphasized that the Sahm rule is a statistical law (statistical regularity), not an economic rule (economic rule). Our understanding is that understanding the Sahm rule requires reading the changes behind the statistics.

According to Sahm's own recent response (the difference between this round and the previous one), the US labor market experienced huge turmoil during the pandemic. The return of labor to the market now, as well as the influx of immigrants facing the US on a larger scale, may have boosted the unemployment rate.

However, Sahm also pointed out that the unemployment rate also reflects a trend of declining demand for workers and weakening workers' ability to spend. I don't think the US economy is in recession now, but it is true that (the US economy) is not moving in a good direction. ... The Federal Reserve needs to start acting, the rest of the economy is cooling down, and the Federal Reserve needs to act in concert.

A trend change in line with Sahm's response is that the phenomenon of illegal immigration in the US has abated in recent months, and the labor force growth rate has also slowed.

On the one hand, it may be that Mexico has stepped up enforcement efforts. On the other hand, the Biden administration also tightened asylum access at the US-Mexico border at the beginning of June this year: when the number of illegal immigrants detained by the US Border Patrol exceeded 2,500 in a day, people who had crossed the border illegally into the US were quickly returned to Mexico and suspended the acceptance of political asylum applications.

This shows that behind the rise in the US unemployment rate in recent months, it does not necessarily mean that the US economy has declined, but corporate recruitment demand may indeed be weakening marginally.

2.2. What do you think of the US Treasury yield curve pattern?

2.2.1. Has the US debt maturity spread expired?

Another typical leading indicator of a recession is the maturity spread in the US Treasury yield curve. In extensive research on cyclical inflection points, the US bond yield maturity spread is probably the most classic and stable leading indicator.

Historically, the first time the yield curve was inverted is usually 5-18 months ahead of the NBER economic downturn point. At the peak of the reversal, they were at most 15 months ahead.

The first time the US bond yield curve was inverted in July 2022, 24 months ago; the margin peaked in July 2023, 12 months ago, so judging from the leading margin in historical statistics, it seems impossible to say that the inversion of US bond maturity spreads predicts an economic recession.

Judging from recent changes in expectations, the significance of subsequent indicators does need to be continuously tracked and verified.

2.2.2. Why might the meaning of this round going backwards be different?

Of course, in terms of internal mechanism, there is indeed a certain difference in the inversion of interest spreads during this round:

First, there is a deep inversion of inflation expectations this time around.

Since the pandemic phase, inflation has been greatly affected by the cost side. While short-term inflation expectations have risen, medium- to long-term inflation expectations have been relatively stable, and may even be suppressed to a certain extent by potential recession risks.

Therefore, judging from the difference between the University of Michigan's 5-year inflation forecast and 1-year inflation forecast, this round of inflation expectations also showed a deep reversal, which is relatively rare in history. This will affect the pricing of US bonds and increase yield inversion.

However, the inversion caused by this factor is more due to short-term supply-side disturbances, and not necessarily because the market expects downward economic and deflationary pressure in the medium to long term.

More importantly, large-scale monetary and fiscal stimulus during the 2020-2022 period, and the Biden administration bucked the trend of fiscal expansion in a high-interest rate environment in 2023-2024, may have delayed the inhibitory effect of high interest rates on residents' income consumption and corporate production and investment activities.

In other words, the US economy is probably already in recession, but early income subsidies and countertrend fiscal expansion are equivalent to having carried out countercyclical stimulus and cross-cycle management ahead of schedule.

On the one hand, stimulus policies hedge against the risk of recession at the cost of short-term inflationary resilience.

On the other hand, supply-side policy orientation can stabilize long-term inflation expectations and prevent them from becoming anchored.

Furthermore, it should be noted that although the inversion of the 2024 US bond maturity spread continues to shrink, as the US economy bottoms out in history, maturity spreads will indeed correct and rise early.

2.3. What do you think of the NBER business cycle reference indicators?

Judging from recent performance, revenue, consumption, and sales performance are relatively stable.

(1) Actual personal income minus transfer payments (PILT). According to St. Louis Federal Reserve data, there was a positive increase of 0.1% month-on-month and 1.8% year-on-year in June.

(2) Actual personal consumption expenditure (PCE) was 0.2% month-on-month and 2.6% year-on-year in June, all showing positive growth;

(3) Wholesale-retail sales adjusted by price changes. In the second quarter, final sales in US domestic purchases were 3% year-on-year, and total domestic final sales without R&D were 3.1%. In terms of high-frequency data, the growth rate of Red Book commercial retail sales slowed slightly year-on-year over week, but there was no significant weakening.

The weak performance was mainly due to employment and production-related data.

(4) The number of non-farm payrolls, with 0.114 million new non-farm payrolls added in July falling short of expectations, which is one of the factors exacerbating expectations of the recession;

(5) The number of employed people measured by household surveys mainly depends on the labor participation rate and unemployment rate performance. The unemployment rate rose in July and triggered the Sahm rule.

(6) Industrial production, ISM PMI recorded 46.8% in July, down 1.7 percentage points from month to month; new orders, output, and employment segments were 47.4%, 45.9%, and 43.4%, respectively, down 1.9, 2.6, and 5.9 percentage points from month to month.

In fact, as a country with 70% consumption, the NBER's reference index covers the main processes and transmission chains of the US economic cycle: income - consumption - sales - production (investment) - employment - income.

The current problems facing the US economy may be mainly due to the fact that corporate production investment and expectations may begin to change, further weakening corporate labor demand.

2.4. What are the other leading metrics?

Currently, different leading indicators are showing conflicting signals, which means that the US economy may be in a complex situation where factors of resilience and recession are intertwined.

Generally speaking, leading indicators designed by authorities take into account many aspects of the macroeconomy. This also means that the composite index may not be particularly obvious in leading the recession. Compared with a certain sub-indicator, the composite indicator may be closer to the verification index.

3. A top-down perspective: How do you view the three paths of the US economy?

An economic recession is a stage in the economic cycle.

Economic development is a historical spiral upward trend.

When discussing the current US economic cycle, it may be difficult to leave aside development trends and talk about experience.

We tend to think that the US recession deal should consider three directions:

3.1. Upward does not land, crosses cycles

We judge that if the US economy wants to cross cycle and not land, it may require variables such as the AI industry to support the US economy to cross the contraction phase and directly start a new round of expansion.

3.2. The downside is error correction and correction of the development path

We believe that if there is error correction in the development path, it means a deep adjustment of the economic structure, which may be accompanied by fiscal consolidation, changes in the international order and geographical pattern, adjustments in the US dollar system, and economic depression.

3.3. The middle scenario is a normal economic landing

Supported by cross-cycle fiscal expansion, the momentum of the US economy is likely to slow down more slowly than the historical process of raising interest rates.

Mechanically, the process of economic recession may first be transmitted from enterprise production investment to employment, which in turn affects residents' income expectations and consumer confidence, as well as the entire economic cycle.

Furthermore, it is also important to note that although the US private sector leverage ratio experienced a decline during the interest rate hike cycle, it is still at a historically high level.

As the economy cools down, local risk events may occur, such as risks for small to medium banks and falling asset prices. The outbreak of each incident may drive a phased adjustment and re-understanding of social expectations.

Regarding medium- to long-term factors, on the one hand, we tend to think that the international order and dollar system may be slow variables, and it is less likely that they will “break” beyond expectations in the short term.

On the other hand, if technological changes in the industry are not realized beyond expectations, asset prices driven by earlier expectations may be adjusted in stages.

To some extent, it is possible to compare the US economic and financial performance from 1999 to 2001 and the macro background of the Federal Reserve's first interest rate cut, as a frame of reference for understanding the path of subsequent US economic and financial changes.

The variable is that technological changes in the industry may indeed be realized beyond expectations.

Risk Alerts

The US job market exceeded expectations, asset price fluctuations exceeded expectations, financial risk events

Note: This article was selected from the securities research report “Two Perspectives on the US Economic Recession” published by Tianfeng Securities on August 8, 2024. Report analysts:

Sun Binbin SAC Practice Certificate Number: S1110516090003

Sui Xiuping SAC Practice Certificate Number: S1110523110001