Source: Zhao Wei Macro Exploration Authors: Zhao Wei, Cao Jinqiu (Zhao Wei is the chief economist of China International Finance Securities and a director of the China Chief Economist Forum) Report key points Driven by post-epidemic stimulus policies and high employment boom, U.S

Source: Zhao Wei Macro Exploration

Author: Zhao Wei, Cao Jinqiu (Zhao Wei is the Chief Economist of Guojin Securities and Director of the China Chief Economist Forum)

Report Highlights

Driven by post-epidemic stimulus policies and high employment boom , U.S. residents’ consumption continues to “prosper.” At present, the policy tide is receding and the job market is still booming. Can the "boom" of consumption continue? Latest review for reference.

The “appearance” of nominal consumption in the United States? Consumption of goods remains high, while the recovery of service consumption lags behind

Overall, the performance of private consumption expenditures and retail sales data in the United States in April was very strong. After the outbreak of the epidemic in 2020, private consumption in the United States suffered a heavy blow, with the nominal year-on-year growth rate once falling to -16.1%. As the U.S. government implemented large-scale fiscal subsidies for residents and the impact of the epidemic on the economy tended to weaken, the compound growth rate of private consumption in the United States rose to 6% in April, far exceeding the 3.9% average of the past 10 years. At the same time, another important consumption indicator, the compound growth rate of retail sales rose to 9.9% in April, which was also much higher than the average level in the past 10 years.

In terms of structure, commodity consumption remains at a high level, while the recovery of service consumption is relatively lagging behind. In terms of commodities, the compound growth rate of durable goods consumption fell back to 13.3%, still at a high level, and the compound growth rate of non-durable goods consumption was 8.1%, which also remained at a high level. Among them, except for some individual commodities such as home appliances, most of the commodities are very eye-catching. Compared with goods, the compound growth rate of service consumption is "only" 4.1%, and repairs are significantly lagging behind. In terms of industries, consumption such as medical care has exceeded the pre-epidemic level, but consumption of catering, accommodation, entertainment, etc. has not yet recovered to pre-epidemic levels.

The “truth” about actual consumption in the United States? Goods consumption has fallen from highs, while service consumption has maintained a weak recovery.

Historically, the nominal and actual consumption trends in the United States have been basically consistent, but there has been a clear differentiation recently. The continued high inflation readings in the United States have obviously interfered with the robust relationship between nominal and actual consumption, leading to a clear divergence between the two. After excluding inflation factors, the compound growth rate of real consumption in the United States in April was 2.5%, slightly exceeding the average level in the 10 years before the epidemic. Retail sales. After excluding inflation factors, the actual year-on-year growth rate of U.S. retail sales in April was only 2.3%, which was also slightly higher than the average level of the 10 years before the epidemic.

According to the 2-point structure, actual commodity consumption has dropped from a high level, but is still high, and service consumption maintains a relatively weak recovery trend. Amid high inflation, the actual compound growth rate of durable goods consumption in the United States in April was 9.5%, still higher than the 6.1% average of the 10 years before the epidemic. The compound growth rates of automobiles, building materials, etc. are all at historical lows. In terms of non-durable goods, the compound growth rate was 4.1%, which was also higher than the average level of 2.1% in the 10 years before the epidemic. Compared with the decline in commodity consumption from high levels, the actual compound growth rate of service consumption in April only recorded 0.9%, maintaining a relatively weak recovery trend.

Where will American consumption go? Service consumption "takes the lead" and the "erosive effect" of inflation needs to be alerted.

Residents' wages have increased and employment has been restored faster than expected. As well as the scenario is restored, service consumption growth is expected and is expected to support overall consumption, but it is necessary to pay attention to the continued accumulation of service inflation pressure. As a leading indicator of service consumption in the United States, the increase in residents' wages and the restoration of employment have continued to exceed expectations and have entered a virtuous cycle. Under the background of the ebb of the epidemic, whether the consumption of leisure and entertainment and other services can be significantly boosted during the peak travel season will also be a "touchstone" to test the quality of consumption. However, it should be noted that the recovery in demand for services may further release inflationary pressure.

Against the background of the ebbing of stimulus and interest rate hikes, the consumption of durable goods is at the forefront or is coming to an end; the consumption of non-durable goods may remain stable, but we need to be wary of the "erosive effect" of inflation. After the epidemic, the consumption of durable goods in the United States has obvious characteristics. "Excess" consumption may exceed 430 billion US dollars, equivalent to nearly 30% of the consumption of durable goods in 2019.In the context of ebbing stimulus and raising interest rates, demand for interest-rate-sensitive durable goods such as automobiles may be suppressed. In terms of non-durable goods, rigid demand may support stable consumption, but we need to be wary of the intensifying "erosive effect" of high inflation on non-durable goods.

Risk warning: The recovery of U.S. service consumption is less than expected; single-month data statistics fluctuate greatly.

Report text

. What is the “appearance” of nominal consumption in the United States? Consumption of goods remains high, while the recovery of service consumption lags behind. Overall, the performance of U.S. private consumption expenditures in April and the latest April retail sales data were very strong. After the outbreak of the epidemic in 2020, private consumption, the "ballast stone" of the U.S. economy, suffered heavy losses, with the nominal year-on-year growth rate once falling to -16.1%. As the U.S. government implements large-scale fiscal subsidies for residents and the impact of the epidemic on the economy tends to weaken, the compound growth rate of U.S. private consumption② has recovered to an extremely high level of 6%, far exceeding the 3.9% average of the past 10 years. At the same time, another important consumption indicator, the compound growth rate of retail sales rose to 9.9% in April, which was also much higher than the average level of 4.3% in the past 10 years.

① Private consumption is a measure of consumer spending by the U.S. Bureau of Economic Analysis. It is a more comprehensive indicator of consumer spending than retail sales and is also the largest component of GDP (accounting for about 70%). Retail sales are survey data obtained by the U.S. Bureau of Commerce through interviews with thousands of retailers. They mainly reflect changes in nominal consumption (mainly commodity consumption).

② In order to eliminate the interference of the base effect, the data in 2021 and later are based on the data of the same period in 2019, and then calculate the compound growth rate, the same below.

In terms of commodities, the growth rate of durable goods consumption has declined but remains high, and the growth rate of non-durable goods consumption continues to recover. Commodity consumption can be further divided into durable goods and non-durable goods. The former mainly includes cars, furniture, home appliances and other goods with a service life of more than 3 years. The latter mainly includes food, clothing, energy products and other relatively easy-to-consume goods. commodity. Specifically, data in April showed that the compound growth rate of durable goods consumption dropped to 13.3% from the previous high of 20%, still at a high level; the compound growth rate of non-durable goods consumption was as high as 8.1%. Against the background of continued expansion of commodity consumption, the proportion of durable goods and non-durable goods consumption in private consumption increased from 11% and 20% before the epidemic to 13% and 22% respectively.

point of view, except for individual products such as home appliances, the consumption of most products is very eye-catching. The retail sales sub-indicator mainly reflects the consumer spending of U.S. residents on goods. In terms of structure, the retail sales of motor vehicle and parts stores, non-store retail, daily necessities, and gas stations are relatively high, accounting for 20%, 16%, 12%, and 9% of total retail sales respectively. Based on retail sales data, except for the relatively low retail growth rate of individual commodities such as home appliances, the compound growth rate of consumption of most commodities such as automobiles, furniture, clothing, and food has remained high, basically at the historical percentile. More than 75%.

In terms of service consumption, consumption of medical care and other services has exceeded the pre-epidemic level, while catering consumption has not yet returned to pre-epidemic levels. The epidemic situation recurred in the early stage, which obviously interfered with the service production recovery process. Judging from April data, compared with the compound growth rate of goods consumption of up to 9.9%, the compound growth rate of service consumption is "only" 4.1%, and the recovery is obviously lagging behind. In order to characterize service consumption more comprehensively, we use credit card consumption data to capture changes in service consumption sub-items. Credit card consumption data shows that consumption on repairs and maintenance (home-based businesses) and outpatient health care (medical services) has exceeded pre-epidemic levels. Catering consumption, etc., because the indoor mask order was relaxed relatively late, related consumption activities have not yet returned to pre-epidemic levels.

. The "truth" about actual consumption in the United States? Goods consumption has fallen from highs, while service consumption has maintained a weak recovery.

Historically, the nominal and actual consumption trends in the United States have been basically consistent, but there has been a clear differentiation recently. The continued high inflation readings in the United States have obviously interfered with the robust relationship between nominal and actual consumption, leading to a clear divergence between the two. After excluding inflation factors, the actual consumption level in the United States is not as strong as the nominal statistical value. The actual compound growth rate in April was 2.5%, slightly higher than the 2.2% average level in the 10 years before the epidemic. After excluding inflation factors, the actual year-on-year growth rate of U.S. retail sales in April was only 2.3%, which was also slightly higher than the average level of the 10 years before the epidemic.

In terms of durable goods, inflation remains high. Behind the beautiful nominal data, actual consumption of durable goods has fallen sharply. After excluding inflation factors, the actual compound growth rate of durable goods consumption in the United States in April was 9.5%, which was significantly lower than the previous high of 18.9%, but still higher than the 6.1% average level of the 10 years before the epidemic. From a structural point of view, completely opposite to the nominal growth rate, the actual compound growth rates of retail sales of major durable goods such as home appliances, furniture, automobiles, and building materials have all fallen sharply, and are respectively at 30%, 28%, 8%, and 5% of the historical percentile. %.

In terms of non-durable goods, the actual consumption growth rate is not as strong as the nominal growth rate, and the overall trend remains stable. After excluding inflation factors, the actual compound growth rate of non-durable goods consumption in the United States in April was 4.1%, which was significantly lower than the previous high of 7.1%, but it was also higher than the 2.1% average level in the 10 years before the epidemic. Judging from historical trends, unlike durable goods consumption, non-durable goods generally correspond to rigid consumption demand, and their proportion in actual total consumption is generally stable at just over 20%. Looking at the breakdown of data, the actual retail sales of clothing stores, sports goods stores, daily necessities shopping malls, etc. are compounded year-on-year and basically maintain a stable trend.

Compared with the decline in commodity consumption from a high level, actual service consumption has shown a relatively weak recovery trend. PCE prices in the United States reached 6.3% year-on-year in April, much higher than the average level of less than 2% in previous years. Among them, there is a huge gap between the inflation pressure of goods and services. In April, the price of durable goods and non-durable goods reached 8.4% and 10.1% year-on-year respectively; in comparison, the price of services was "only" 4.6% year-on-year. The reason is that, unlike the hot consumer demand for goods, the demand for service consumption is restricted by factors such as repeated epidemics, and the actual compound growth rate is only 0.9%. Among them, in addition to relatively rigid service consumption such as household operations and medical care, the compound year-on-year compound consumption remains positive, while the compound year-on-year compound consumption of non-essential services such as transportation and entertainment is still in the negative range.

. Where will American consumption go? Service consumption "takes the lead", and we need to be wary of the "erosive effect" of inflation.

As a leading indicator of service consumption, salary increases and employment restoration continue to exceed expectations and have entered a virtuous cycle. Historical review shows that the year-on-year trend of total salary income in the United States is in sync with or weakly ahead of consumption, especially service consumption, and the former mainly depends on the employment boom. At present, the number of job vacancies in the United States has been above 10 million for eight consecutive months, indicating that labor supply still does not fully meet demand. Under the strong recruitment demand, "post-epidemic trauma" will still restrict labor supply and intensify the contradiction between supply and demand, forcing companies to significantly increase wages to recruit people. Private salary income maintains high growth, which will provide strong support for service consumption (For a detailed analysis of the US job market, please see "How far is the US recession from the job market?").

Coupled with the need for natural recovery of offline consumption after the epidemic, further recovery of service consumption can be expected. With the rapid expansion of commodity consumption, the proportion of service consumption in total consumption has dropped to 60.6%, which is more than 3 percentage points lower than before the epidemic. There is considerable room for repair. More importantly, as the epidemic in the United States recedes, whether consumption of leisure, entertainment, catering, accommodation and other services can be significantly boosted during the peak travel season will also be an important perspective to test the quality of U.S. consumption. It should also be noted that the recovery in demand for services may further release inflationary pressure. Taking history as a guide, changes in server-side inflation in the United States often determine the central level of overall inflation.

Post-epidemic consumption of durable goods has obvious front-end characteristics. Taking into account the suppression of related consumer demand by the ebbing of policies, the high growth in durable goods consumption may be coming to an end. The high growth in durable goods consumption in the United States after the epidemic is inseparable from large-scale financial subsidies and the low interest rate environment. However, based on historical experience, this round of durable goods consumption has significant front-end characteristics. "Excess" durable goods consumption may exceed US$430 billion, equivalent to nearly 30% of durable goods consumption in 2019. In the context of the fiscal ebb and the Federal Reserve raising interest rates, consumer demand for interest-rate-sensitive durable goods, such as automobiles, may be suppressed. The "mania" of durable goods consumption may be coming to an end, and we need to continue to track it in the future.

In terms of non-durable goods, overall consumption may remain stable, but we need to be wary of the "erosive effect" of high inflation on some non-durable goods consumption. Since most of them correspond to rigid consumption demand, the growth rate of durable goods consumption may remain "tepid". However, it should be noted that with the further recovery of service consumption during the peak travel season, combined with limited supply, rising oil prices are expected, which may further push up inflation. May gasoline retail price data shows that the CPI energy commodity sub-category may continue to exceed 40% year-on-year. Referring to the Bank of America's consumer survey report, continued energy inflation may have a further "erosive effect" on some non-durable goods consumption. (For research and judgment on oil price trends, please see "The Russia-Ukraine conflict is heating up again, why are oil prices "very calm"?" ).

After research, we found:

(1) Overall, the compound growth rates of private consumption expenditures and retail sales in the United States in April reached 6% and 10.6% respectively, both of which were much higher than the average level in the past 10 years. In terms of structure, the compound growth rate of durable goods consumption fell back to 13.3%, still at a high level, and the compound growth rate of non-durable goods consumption was as high as 8.1%, also at a high level. Compared with goods, the compound growth rate of service consumption is "only" 4.1%, and repairs are significantly lagging behind.

(2) Historically, the nominal and actual consumption trends in the United States have been basically consistent. However, under the recent disturbance of high inflation readings, the two have clearly diverged. In terms of structure, the actual compound growth rate of durable goods and non-durable goods consumption has declined, but they are still at historically high levels. Compared with the decline in commodity consumption from high levels, the actual compound growth rate of service consumption in April only recorded 0.9%, maintaining a relatively weak recovery trend.

(3) Resident wages have increased and employment has been restored beyond expectations. As the situation recovers, service consumption growth is expected and is expected to support overall consumption. However, it should be noted that the recovery in service consumption demand may lead to further release of inflationary pressure. Against the background of ebbing stimulus and raising interest rates, consumption of durable goods may be coming to an end; consumption of non-durable goods may remain stable, but we need to be wary of the "erosive effect" of inflation.

Risk warning:

. The recovery of US service consumption is not as good as expected. Some of the impacts of the epidemic may be long-term, thus reshaping the consumption habits of American residents.

. Single-month data statistics fluctuate greatly. US private consumption and retail sales data fluctuate greatly in a single month, which may disturb the judgment of future trends.