Goldman with the note:
The weak Q3 growth was driven by a number of factors – Covid outbreaks and chip shortages that the government has less control over on the one hand, and property tightening and power cuts that are mostly policy-driven on the other. The September activity data show evidence on the combination of various shocks to the economy (fig.3) For example, catering sales (i.e., restaurant services) rebounded sharply in September after slumping in August on Covid lockdowns in multiple provinces. Auto production and sales remained soft on chip shortages. Property sales continued to drop on the government’s deleveraging efforts and lending restrictions. Output of high-emission products such as steel and cement fell sharply on the “dual controls” of energy use and severe coal shortages which led to power cuts and production halts in these sectors.
After the anemic sequential growth year-to-date (averaging only about 2% annualized rate), the Chinese economy appears to have gone from a positive output gap at the end of last year to some excess capacity in Q3.
Looking across different sectors, Exhibit 5 shows that, with the exception of agriculture, all industries are currently at or below trend level of output, assuming a pre-Covid sector-specific trend. In the case of leasing and commercial services (e.g., travel agencies and large conferences), hotel and restaurant services, and other services (e.g., household cleaning services), the negative output gap remains significant. Eighteen months after the onset of the Covid outbreak early last year and with no end of the “zero Covid” policy in sight, activity in these sectors is at risk for longer-term scarring effects.
Household consumption was the hardest-hit part of the economy last year on both lower income growth and a higher saving rate. By Q3, household saving rate has mostly normalized to its pre-Covid level, falling from a peak of 35% in 2020Q1 to 30% now (Exhibit 6). The main constraint to consumption is income growth. As of Q3, the growth of household disposable income averaged only 6.6% per year over the past two years, compared to 8.8% in 2019. Among different sources of income, growth of business income underperformed the most, averaging 3.6% per year over the past two years compared to 8.0% in 2019 (Exhibit 7).
September PPI inflation reached the highest level since the data were available in 1997, raising questions about both the duration of the high PPI inflation and its potential passing through into CPI inflation which has remained low. On the first question, PPI inflation is likely to stay high in the near term, but should soften notably in six months on base effect. If prices were to remain unchanged from here, PPI inflation would drop to about 2% in mid-2022. On the latter, we expect the pass-through from PPI to CPI to be limited for two reasons.
First, CPI has three distinct components – food, non-food goods, and services (Exhibit 11). Food inflation and service price inflation are likely to remain low in the coming months on depressed pork prices (which dominate food prices) and negative output gap (which is a key driver of service inflation). Second, historically the sensitivity of CPI non-food goods inflation to PPI inflation is statistically significant but economically small. Exhibit 12 shows a nonlinear relationship where relatively mild year-over-year PPI inflation (i.e., between -5% and +5%) appears to have very little impact on CPI non-food goods inflation.
But even at more extreme levels of PPI inflation, the magnitude of the pass-through remains modest: an additional 1pp increase in PPI inflation from its currently elevated levels boosts CPI non-food goods inflation by 0.25pp which translates into 0.1pp for headline CPI inflation.
Given the continued slowdown in credit growth – the year-over year growth in the stock of total social financing (TSF) dropped to 10.0% in September from 13.5% a year ago – and the “just do enough” approach of policymakers, we revise down our credit growth forecast to 10.5% for 2021 (previously 11.5%). This still implies a modest pick-up in sequential credit growth in Q4. In addition, we recently changed our monetary policy forecast and no longer expect a RRR cut in Q4. This is not a call on the broader monetary policy stance. Rather, recent communications by the PBOC suggest that the central bank is likely to use targeted liquidity instruments (e.g., SME and green financing relending programs) instead of broad-based RRR cut to replace the large amounts of maturing MLF loans.
Meh. It’s pretty obvious that the base case is that Q4 growth is not going to rebound as property and infrastructure both get worse.
Q1, 22 is a chance to lift as the energy crunch ends with collapsing coal prices and swathes of industrial production can resume, including metals processing.
But that, too, will be a head-fake and only serve to delay action on the property market where the real growth downdraft is coming from.
If China does not back off property, which is rightly Goldman’s base case, then its economy will struggle to grow at all. But GS can’t say that given it is still trying to polish its commodities bubble.
The base case remains that growth gets so bad that policymakers will eventually panic and juice credit, probably as the global economy slows, commodities crash and supply-side bottlenecks clear.
That looks more Q2, 22 to me.