Press play to listen
Chinese GDP growth (seasonally-adjusted) was 11.5% qoq in Q2. This was stronger than consensus forecast (+9.6% qoq) and more than reversed Q1 contraction of 9.8% qoq. This record-high growth reflects both a post-lockdown bounce in economic activity and of course extremely “favourable” base effects.
This is pertinent for China but also its key major trading partners and global economy. China accounts for about 20% of world GDP in PPP terms and thus growth single-handedly contributed about 2.3 percentage points to quarter-on-quarter global GDP growth in Q2.
We estimate that in constant terms China’s GDP hit a record high in Q2 2020, was 0.6% larger than in Q4 2019 and only 2.2% lower than if GDP growth had continued to trend at 1.4% qoq. For GDP to return to “trend” in Q3 GDP growth would have to be about 3.7% qoq.
In this scenario the shape of China’s GDP level would approximate that of a square-root (a broadly symmetric recovery). In comparison we expect the recovery in global GDP to resemble that of a hockey-stick (an asymmetric recovery) as most major economies likely contracted sharply in Q2 (Shape of Recovery: Square Root & Hockey Stick, 5 June 2020).
Over the years analysts have queried the accuracy of official Chinese data, with the consensus view seemingly that they overstate the “real” pace of domestic economic growth. This remains difficult to prove (or disprove) in our view.
In the past decade quarterly Chinese GDP growth has been steady and less volatile than in major developed economies, including the United States and United Kingdom, which has led to suggestions that Chinese GDP data are being “smoothed”.
Chinese Policy makers could counter that they have more policy-levers at their disposal than their US and British counterparts and that GDP growth in China was far more volatile in Q1 2020 and likely Q2 than in the United States and United Kingdom.
Our analysis suggests that official GDP data do capture the broad changes in the direction of Chinese economic growth. Notably unofficial manufacturing PMI data have historically correlated more closely with the official GDP than official PMI data have.
Chinese GDP rebounded 11.5% qoq in Q2, cancelling out 9.8% qoq contraction in Q1
GDP in the world’s second largest economy rebounded faster than expected in Q2, with the shape of the recovery V-shaped in level terms. However, questions remain about the accuracy of official Chinese macro data.
Quarter-on-quarter seasonally adjusted GDP growth was 11.5% according to National Bureau of Statistics data (see Figure 1). This was stronger than the consensus forecast (+9.6% qoq) and more than reversed the 9.8% contraction recorded in Q1 2020. This was the fastest pace of quarterly growth on record based on NBS and World Bank data going bank to 1995 (the previous high was 4.5% qoq in Q1 2007, according to our estimates). This rapid rebound reflects both a post-lockdown bounce in economic activity and of course extremely “favourable” base effects. This translated into year-on-year growth of 3.2% in Q2 2020, up from -6.8% in Q1 and stronger than the consensus forecast of +2.5% yoy (see Figure 1).
This is pertinent for China but also its key major trading partners and the global economy. China accounts for about 20% of world GDP in purchasing power parity terms, according to the IMF. Therefore Chinese growth of 11.5% qoq in Q2 single-handedly contributed about 2.3 percentage points to global GDP growth according to our estimates. This will of course not have stopped global GDP from contracting sharply in Q2 as very few major economies are likely to have recorded positive growth in Q2 (as we will explore in a forthcoming Insight). However, stronger-than-expected Chinese growth will have partly alleviated the collapse in world growth (had Chinese growth been in line with the consensus forecast of 9.6% qoq it would have contributed “only” 1.9 percentage points to global growth).
Trend growth and shape of Chinese GDP recovery
We estimate, using an index constructed from official quarter-on-quarter seasonally-adjusted growth rates, that in constant terms China’s GDP hit a record high in Q2 2020 and was 0.6% larger than in Q4 2019 – the previous quarterly record high (see Figure 2). Bloomberg states that China’s “economy is still 1.6% smaller than at the end of the first half in 2019” but does not make explicit whether it is referring to GDP in current prices (nominal GDP) or in constant prices (nominal GDP adjusted for inflation, i.e. real GDP).
However, we also estimate that China’s real GDP (level) was about 2.2% lower in Q2 2020 than it would have been had GDP growth continued to trend at 1.4% qoq in both Q1 and Q2 2020 (see Figure 2). In order for Chinese GDP (level) to return to “trend” in Q3 (the red dotted line in Figure 2) we estimate that GDP growth would have to be about 3.7% qoq (the dotted blue line).
If China’s GDP level indeed returns to trend in Q3 its shape would approximate that of a square-root (see Figure 2). In comparison we expect the recovery in the global GDP level to resemble that of a hockey-stick because global GDP likely contracted sharply in both Q1 and Q2 (unlike China where GDP recovered rapidly in Q2) and will likely recover quite slowly in Q3 and Q4 in our view (see Shape of Recovery: Square Root & Hockey Stick, 5th June 2020). Put differently, we expect global GDP’s path to be asymmetric whereas China’s recovery will be more symmetric.
Questions about actual Chinese growth rates unanswered but concerns likely overdone
It remains difficult to prove (or disprove) that official Chinese data over-state the actual rate of GDP growth, in our view. However, we conclude that while quarterly GDP data may have been “smoothed” in recent years they do broadly reflect the general trend (if not level) in economic growth.
Volatility in GDP growth in past decade has been lower in China than in US and UK
Over the years analysts have queried the accuracy of official Chinese data, including quarterly GDP numbers and monthly retail sales, industrial output, fixed investment and external trade figures (June data were released this week). Various methods have been applied to establish whether official Chinese data paint an accurate picture of domestic economic activity. These include using unofficial macro data (e.g. Caixin PMIs) or alternative indicators of economic activity (e.g. electricity consumption, traffic levels). The consensus view is seemingly that official Chinese data overstate the “real” pace of economic growth while understating its volatility.
- Figure 3 shows that between 1998 and 2007 the “volatility” in quarterly GDP growth was generally higher in China than it was in the United States and United Kingdom. It is not atypical for GDP growth in a large, open and transitioning economy to be more volatile than in a mature and developed economy. Moreover Chinese GDP growth was volatile during and after the Asian financial crisis, which started in July 1997 and extended through to 1998, and the SARS epidemic (late-2012 to July 2013). In contrast GDP growth in the United States and United Kingdom was reasonably robust and thus less volatile during these two periods.
- During the 2008-2009 financial crisis, volatility in Chinese GDP growth did not rise much, whereas it did in the United States and United Kingdom because in relative terms growth in China slowed less sharply than in the United States and United Kingdom. It was arguably easier for policy-makers in China, effectively still a command-economy, to manage growth by for example boosting government spending on infrastructure, than it was in the United States and United Kingdom where consumer spending accounts for a large share of domestic GDP. Moreover, external trade was a key driver of Chinese GDP growth and global trade did not slow much in 2008-2009 (and certainly slowed far less than it has done in recent months).
- Since 2010 the volatility in GDP growth in China has collapsed and consistently been below GDP volatility in the United States and United Kingdom – and that is arguably harder to explain. Actually the volatility in the volatility of Chinese GDP growth has been near zero since 2015 – another way of saying that GPD growth has trended in a very narrow range around 1.6% qoq (the red line in Figure 3 has been near zero and flat).
- One common explanation put forward is that official Chinese GDP data are being “massaged” to smooth out the peaks and troughs in growth. Chinese policy makers would counter that the domestic economy has matured while at the same time they have more policy-levers at their disposal than their US and British counterparts (for example they can more readily control the Renminbi as the People’s Bank of China fixes the daily USD/CNY central parity rate). Moreover, they could point out that, based on official data, GDP growth in China was far more volatile in Q1 2020 and likely Q2 than in the United States and United Kingdom because Chinese GDP i) contracted far more sharply in Q1 and ii) recovered rapidly in Q2 whereas GDP in the United States and United Kingdom likely contracted further (see Figure 4).
Official GDP data capturing broad changes in direction of Chinese economic growth
There has historically been a reasonably strong positive correlation between the change in the non-official manufacturing PMI (sponsored by Caixin Media since June 2015 and previously by HSBC) and the change in year-on-year Chinese GDP growth (see Figure 5). Note that we intentionally omit Q1 and Q2 2020 data as they would distort the scales.
Notably the historical correlation between the official manufacturing PMI and Chinese GDP growth since mid-2010 has been weaker. This runs counter to the idea that official macro data are tweaked to all point in the same direction and that they understate the true extent of the slowdown in Chinese growth (see Figures 6 and 7).