Another quarter of remarkably precise China GDP growth data

October 19, 2015

????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????China GDP releases are starting to look like near-perfect landings each and every time, in all kinds of weather conditions and visibility.

Yet another quarter has just gone by – literally less than three weeks ago – and already statisticians have reported that growth slowed a tiny sliver from Beijing’s 2015 target of 7 percent recorded for the first half of the year. Now it’s 6.9 percent, slightly above the Reuters consensus forecast from 50 economists of 6.8 percent.

It is difficult to understate just how precise such figures are in the grand scheme of economic data reporting.

It is also difficult to ignore just how remarkable this stability is considering the Chinese authorities are trying to rebalance the entire economy away from reliance on exporting manufacturing goods toward domestic consumer spending.

And that worry about a Chinese growth slowdown was one of the main reasons cited by the U.S. Federal Reserve for holding off last month on its first rate rise in nearly a decade.

That’s also not to mention that China growth concerns dominated the International Monetary Fund and World Bank’s latest meetings in Lima, Peru.

In the past three years, Chinese GDP data as reported have only missed the Reuters Polls consensus three times, and on each occasion it was because the reported growth figure beat by just 0.1 percentage point.

For the periods of Q4 2013 through to Q1 of this year, the reported figure was exactly on forecast.

China GDP

Other large and important global economies are nowhere near as accurate. U.S. growth data have taken even the most pessimistic forecaster completely off guard on several occasions since the financial crisis, most recently in the first quarter of last year. The initial report for Q1 GDP this year also matched the lowest forecast. Initial U.S. growth data have only actually been reported exactly in line with expectations three times in the last half decade.

It seems implausible that economists, who are often widely panned as a group for failing to predict economic turning points, are uncannily able to nail Chinese GDP within a few tiny slivers of a percentage point each and every time.

Many analysts and investors, including those with plenty of experience covering China, increasingly are challenging the growth data.

Magnus on China

The most obvious difficulty most people have is the speed at which the official numbers are published. The U.S., for example, won’t publish its first take of Q3 GDP data for more than another week. Those figures will be revised twice over the following two months, and then again, perhaps substantially, in future benchmark revisions.

Lubomir Mitov, UniCredit’s chief CEE economist, wrote in a weekend note to clients before the latest release:

“I have my doubts that a country of 1.5 billion people with that complicated administrative structure can really produce a GDP number in a month upon the expiration of the quarter and rarely revise it much. I also become somewhat suspicious when the growth number always closely matches the official target. (This suspicion probably stems from my background, having grown up in a centrally planned economy where the five-year plan targets were always met despite ample evidence to the contrary).

— Graphic and analysis by Sarmista Sen

No comments so far

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see