The chip shortage does explain a considerable degree of weaker car sales, but even if the impact of this factor is excluded as much as possible, the actual car consumer demand in the third quarter has also gone through the high point of cyclical recovery, and the current round of automobile consumption recovery is weakening.
If there is no impact on the supply side of the lack of core, under the condition of other conditions remain unchanged, the total social zero in July may increase by about 1.5%, and the year-on-year growth rate of the total social zero in August may be less than 1 percentage point, and the judgment that the recovery of residents' consumption is weak is still valid.
In the next six months, even if the chip shortage can be substantially alleviated, the corresponding potential consumer demand has begun to weaken. With the combination of upward supply and downward demand, passenger car sales in the fourth quarter are still difficult to return to the same period in 2020, and may continue to drag down the recovery of overall consumption.
——Zhu He, Deputy Director and Young Researcher of the Research Department of the China Finance Forty Forum; Sun Zihan, Research Assistant of the China Finance Forty Forum
What happens to car sales if there is no shortage of chips?
Text | Zhu He, Sun Zihan
Passenger car consumption directly reflects the consumer demand for durable goods of residents and is an important indicator for judging total demand in the macro analysis of various countries. Since May, China's passenger car sales growth rate has begun to decline rapidly, with passenger car sales in August reaching 1.45 million units, down 14.7% year-on-year, and even lower than the same period in 2019.
The market generally attributed the decline in car sales to a reduction in production caused by a shortage of chips. This is obviously an important reason, but changes in terminal demand are also masked by supply shocks. Therefore, in order to better observe the recovery momentum of current resident consumption, it is necessary to analyze from a quantitative point of view how much the supply impact caused by chip shortage has explained the decline in passenger car sales since the second quarter.
Our analysis shows that if there is no chip shortage caused by the epidemic, the year-on-year growth rate of passenger car sales in the second quarter of 2021 is about 7.6%, but the theoretical year-on-year growth rate in the august month may be only 2.8%.
The basic meaning of the above calculation is that after excluding the interference of supply factors, the recovery cycle of passenger car sales in this round has begun to weaken. Since passenger car sales accounted for more than 10% of total retail sales of consumer goods, if this factor is taken into account, the growth rate of total social zero in August may increase by less than 1 percentage point year-on-year. In the next six months, even if the chip shortage can be substantially alleviated, under the intertwining of supply and downward demand, passenger car sales in the fourth quarter will still be difficult to return to the same period in 2020, which may drag down residents' consumption.
First, the two basic facts of the decline in passenger car consumption: "three quantities of the same decline" and "endogenous and external decline"
(1) "Three quantities falling together": production, sales volume and inventory volume decline at the same time
Since the beginning of 2021, automobile production and sales have both declined, excluding the data affected by the Spring Festival in February and March, the growth rate has always been negative, and the magnitude and duration of the decline are far more than in previous years. At the same time, the inventory level of car companies during the year is also low, and the overall trend of destocking is present. The passenger car industry has shown the characteristics of "three volumes falling at the same time" in terms of output, sales volume and inventory.
The six-month "three-volume decline" is the first occurrence in the past decade. As shown in Figure 1, in the past decade, we have occasionally seen a brief "three quantities and one drop" phenomenon at the beginning of the year, most of which lasted only two months, and the situation of synchronous changes in the three will change after entering the second quarter. Compared with the off-season of automobile sales in the second and third quarters of 2018, there was a small peak in the inventory of car companies at that time, which showed that the sluggish consumption would lead to inventory accumulation. This also shows from the side that the weakening demand is not enough to explain the continuous weakening of car sales year-to-date, which must have a supply-side impact, otherwise the decline in sales will not be accompanied by obvious destocking.
Figure 1 Since the beginning of 2021, there has been a continuous phenomenon of "three quantities of co-decline" in passenger cars
Source: Wind, Caucus
(2) "Internal rise and external decline": The proportion of sales of domestic brand cars has increased significantly
Despite the overall decline in passenger car sales, by brand, this year's passenger car retail market presents two very different scenes. Domestic brand passenger car sales are hot, and the performance of imported cars/joint venture cars is relatively "deserted", showing a structural change of "internal rise and external decline". Especially since April, the market share of independent brands has gradually expanded, reaching a record high of 46% in the same period in July.
The "endogenous and external descent" since April is also the first time in the past few years. As shown in Figures 2 and 3, from 2014 to 2019, the market share of domestic car sales is about 40%, and the joint venture brand is about 60%, and the change in the market share of the two types of cars has obvious seasonal laws. Even in 2020, passenger car sales have been significantly affected by the epidemic, and the market share of the two types of brands has basically returned to the historical average level of the same period in the past six years in the fourth quarter. This shows that the structural characteristics of market share are relatively stable. However, since April this year, the market share of domestic cars and joint venture cars has deviated significantly from the average, and the market share of independent brands has increased significantly, significantly higher than the historical average level, which cannot be explained by seasonal changes.
Figure 2 Market share of self-owned brand passenger cars from 2014 to 2021
Source: Wind, China Automotive Industry Information Network, China Automobile Association
Figure 3 Market share of passenger cars of foreign brands from 2014 to 2021
The "structural shortage" of chips is a key clue to understanding the above phenomenon. The shortage of automotive chips will affect all car production, but not all automotive chips will be affected equally. In reality, the impact of the epidemic on MCU chips is significantly greater than that of other types of chips, because such chips belong to the high-end chips in automotive chips, Taiwan is the most important production center, and Southeast Asian countries such as Malaysia are responsible for a considerable share of packaging and testing. Since the end of 2020, the repeated outbreaks of the epidemic in Southeast Asia have directly affected the packaging and testing of Countries such as Malaysia, and ultimately led to the supply of related chips being affected. In contrast, many low-end types of automotive chips have basically achieved localization production, so the impact of the epidemic on the supply of chips here is relatively controllable, and it is not as great as the impact of MCU chips.
The impact of the "structural shortage" of chips on joint venture brand cars and domestic brand cars is also significantly different. MCU chips are widely used in the automotive field, but in general, the more high-end cars carry the more types and numbers of MCU chips. Therefore, the impact of the "structural shortage" of chips on high-end cars will be relatively large, and the pressure on low-end cars will be relatively small. Further, mid-to-high-end cars occupy a more important position among joint venture brand car companies, so joint venture brand cars, which are mainly based on high-end car sales, are more likely to be inhibited in terms of production and sales. In contrast, domestic brands have been less affected, coupled with the substitution effect of domestic brand cars on joint venture brand cars, which has finally shown a higher growth rate, and the market share of domestic cars has also increased significantly.
In summary, it can be seen that the "three quantities and the same decline" indicates that the current decline in sales cannot be fully explained by the changes in the demand side, and the shrinking supply is an important reason for the decline in automobile sales. The main reason for the shrinking supply is the "structural shortage" of chips, that is, the supply chain of high-end chips is more affected. Further, the "structural shortage" of chips has a significant heterogeneity on the impact of domestic brands and joint venture brand cars, and the impact of domestic brand cars is significantly smaller than that of joint venture brand cars, resulting in "internal rise and external decline".
Second, the calculation ideas and results of counterfactual inferences on passenger car sales
As mentioned earlier, from 2014 to 2019, the market share of joint venture brand cars and domestic brand cars remained basically stable, and the recent changes mainly came from the impact of structural shortage of chips. Based on this long-term stable structural relationship, and the reality that domestic brand cars are relatively less affected, we give the following counterfactual inference logic:
If there is no chip shortage caused by the epidemic, the corresponding sales proportion of domestic brand cars should be at a balanced level, which can reverse the potential level of total sales and growth rate of passenger cars.
It should be pointed out that in the counterfactual inference designed in this article, the domestic brand car plays the role of the control group, but it is not directly used to match the sales or growth rate of the domestic brand car to match the sales or growth rate of the joint venture car. This is because there is no stable long-term relationship between the two cars in terms of sales or growth rate, and the "common trend" condition cannot be met. Comparatively speaking, whether it is the seasonal change law or the absolute value level, the market share of the two has a relatively good stability in history, and it is precisely since April this year that there has been a significant deviation, which is obviously more suitable as the basis for counterfactual inference.
Not only that, before making counterfactual inferences, it is also necessary to make a certain degree of correction to the sales of domestic cars, and remove the part of the demand for replacement joint venture cars to restore the actual demand situation as much as possible.
(1) There are two effects of the decline in sales of joint venture vehicles: the total amount effect and the substitution effect
Chip shortage has two influencing mechanisms on the market share of domestic cars, namely the total amount effect and the substitution effect. The total effect is that the decline in joint venture car sales has changed the total amount of automobile sales, thereby increasing the market share of domestic vehicles. The structural effect means that when the supply of joint venture brands is insufficient, the demand for some joint venture cars will be replaced by domestic cars, thereby increasing the sales volume and market share of domestic brands.
In order to distinguish the impact of the two effects, we first calculated how large the proportion of domestic cars will rise only because of the decline in sales of joint venture vehicles. As shown in Table 1, under this assumption, the corresponding market share of domestic cars in April-August this year should be 41%, 42%, 46%, 53% and 45%. This is still somewhat different from the actual data, and it can be considered that this gap comes from structural effects. After calculation, it can be seen that the structural effects of April and August are 1.36%, -1.05%, -1.83%, 1.82% and 0.11%, respectively, and the positive value is indicated as the replacement of domestic car demand for joint venture vehicles, and the negative value is the opposite.
Another understanding is that the above assumptions may overestimate the impact of chip shortage on joint venture cars, in fact, chip shortages may affect joint venture cars and domestic cars to varying degrees, and the negative effects obtained from this may be partly derived from the impact of chip shortages on independent brands. However, whatever the reason, we must exclude it from domestic car sales to reflect the real level of consumption that is not affected by chips.
Table 1 Domestic car sales under the influence of assuming no chip shortage (10,000 units)
Data source: China Automobile Association, calculated by the authors themselves
(2) Passenger car sales in the second quarter increased by about 7.6% year-on-year, and the theoretical growth rate in the month of August was about 2.8% year-on-year.
Based on the sales of domestic vehicles after excluding structural effects, and assuming that the data reflects the real demand without the impact of chips, we then combine the historical market share level of 2014-2019 to estimate the total sales volume and growth rate of passenger cars without chip shortages.
As shown in Table 2, if there is no impact on the supply side of the chip shortage, the theoretical growth rate of passenger car sales in the second quarter is about 7.6% year-on-year. The growth rate was higher in July, at 8.5%, but the theoretical year-on-year growth rate in August was only 2.78%. From the perspective of the two-year compound growth rate, after the July high, the two-year compound in August has dropped to 4.5%. Halfway through the third quarter, even if the chip problem gradually eases, the recovery of automobile sales is significantly weaker than in the first and second quarters, and this weakening has nothing to do with the so-called off-season of automobile consumption.
Table 2 Total passenger car sales (10,000 units) assuming no chip shortage
In summary, the chip shortage can indeed explain the weakening of car sales to a considerable extent, but even if the impact of this factor is excluded as much as possible, the actual automobile consumer demand in the third quarter has also gone through the high point of cyclical recovery, and the current round of automobile consumption recovery is weakening.
Considering that passenger car sales account for about 10% of the total retail sales of social consumer goods, if there is no impact on the supply side of the lack of cores, under the condition of other conditions remaining unchanged, the total social zero in July may increase by about 1.5%, and the year-on-year growth rate of the total social zero in August may increase by less than 1 percentage point, and the judgment that the recovery of residents' consumption is weak is still valid. In the next two quarters, even if the chip shortage can be substantially alleviated, the corresponding potential consumer demand has begun to weaken. With the combination of upward supply and downward demand, passenger car sales in the fourth quarter are still difficult to return to the same period in 2020, and may continue to drag down the recovery of overall consumption.