China wrestles with data-accuracy conundrum amid global spotlight
The veracity of official data from China has long been questioned due to inconsistencies, missing key indicators or massaging of the numbers for political reasons. Now the stakes seem higher: the rise in the shadow banking system and question marks over the accuracy of economic and industry data add to fears that calculating systemic leverage in China’s economy is mission impossible.
For bears, China’s abrupt suspension of industry-related data end-June from its monthly survey of manufacturing, as reported by Bloomberg – meaning data on steel orders and certain class of inventories disappeared – confirms an oft-cited fear: opacity rules the roost in Beijing’s treatment of official data.
However, there are signs the quality of the statistics is set to improve. As with most change in China, it will take time.
As it becomes more integrated into the global economy, China needs to boost its credibility among markets and international investors by demonstrating that its data are more fact than fiction and provide a reasonably accurate picture of the state of its economy.
There is also expectation that now an era of slower GDP growth has been endorsed, China will engage in less data air-brushing as it focuses on re-balancing its economy away from an investment and export-led model towards domestic consumption.
“Everybody knows the figures are manipulated,” says Anastasia Nesvetailova, director of City Political Economy Research Centre at City University London. “A lot of it is for local or national politics so a lot of data we see as real in fact are forecasts by civil servants.
“Because they operate on targets, similar to the old soviet mentality, their own internal reporting within departments are not on real life but what they need to achieve, what they plan to perform, what they budget for next year or for the next five years. So these are forecasts being taken as de facto for what has happened.”
It is not a straight-forward open-and-shut case of lies, damned lies and statistics. Chinese GDP growth figures, for example, while they are widely discredited, are thought by experts to provide only an approximation that under-states growth as often as it exaggerates it in a bid to smooth out volatility.
Unthinkable in a developed economy, no proper unemployment data exists. The 4% rate, unchanged in more than a decade, includes only urban unemployed with residents’ permits, completely overlooking millions of migrant workers from the countryside.
China’s data are also distorted by speculative capital inflows and a shadow banking-financed investment boom which might contribute to spectacular yet phantom top-line growth numbers.
Many investments are off-balance sheet in non-performing infrastructure or real estate projects. These bad debts are being rolled-over which eats into credit expansion, meaning less real GDP growth from each additional unit of credit poured into the economy.
The central government is cracking down on shadow banking by tightening regulation, reporting and trying to restrict funding to the sector.
Nesvetailova argues China’s need to control the sector means it has more reliable data on (legal) shadow banking than on general economic statistics thanks to bodies such as China Central Depository and Clearing, China Securities Depository and Clearing, and the China Trustee Association.
The complexity of the production, income and expenditures approaches the National Bureau of Statistics (NBS) uses to calculate growth can result in a GDP measure in which shadow banking is over- or understated, depending on whether holdings are on- or off-balance sheet.
Sara Hsu, assistant economics professor at the State University of New York, comments: “Balance sheet, income and expenditures statistics may all be reported to the government inaccurately by corporations so there is inaccuracy at the very root of GDP calculation.
“To make matters worse, the use of different price indices may result in entirely different GDP figures. We must therefore take China’s statistics with a grain of salt.”
On a practical level, there are huge administrative and logistical obstacles to gathering quality data in a country and economy that are so large. Much of the data is collected, collated and supplied to the NBS in Beijing by local governments.
These local governments might have agendas that incentivize them to overstate or understate the performance of the local economy. A looming election might prompt a governor to exaggerate how the economy has done under his leadership, while another might paint a bleak picture to gain more financial assistance from central government.
To tackle this problem, the NBS last year adopted a unified system to collect output, retail sales and investment data from 700,000 firms directly in a bid to improve accuracy.
The NBS has also been working on the collection of other statistics and its data methodology in an overall effort to provide more reliable figures for policymakers. New methods of measuring unemployment that are more in line with international standards are also being piloted.
Both language and actions signal a change in emphasis from simply churning out more data, at which China has been quite successful, towards better-quality data – and it’s clearly coming from the top.
The bureau’s official organ China Information News recently quoted NBS chief Ma Jiantang calling for fraudulent data submission to be tackled as a matter of urgency. The paper went on to state that those submitting inaccurate information for statistics might be “severely punished”.
For Nesvetailova, of City University, it is telling that key bureaucrats who have spoken out about the poor quality of data, far from being purged, have been promoted.
Li Keqiang, who was appointed premier this year as part of the Communist Party’s once-a-decade power transition, has been quoted as far back as 2007 as saying that GDP figures were “man-made” and “for reference only”.
When the export figures for May showed an improbable rise, the General Administration of Customs openly blamed the anomaly on firms over-reporting overseas sales or invoicing for non-existent orders – another data-warping problem.
Customs and the State Administration of Foreign Exchange have since launched a campaign to crack down on fake export reporting.
“As a part of the global economy, China is extremely interested in attracting foreign capital, it’s interested in rankings, it’s interested in world indices and they’ve learned that to gain membership of the club you need data,” says Nesvetailova.
“Chinese policymakers have historically proved themselves extremely pragmatic, whatever the corruption or presumed ideology; they are very flexible and tuned-in to the market, to the economic context. So there is more receptiveness to tuning the model.”
She adds: “They also have the infrastructure, the coordination between bodies, the relationships, the sort of capital you can use if you really decide to go for a change – incremental or deep structural.
“So we shouldn’t see deterioration of data unless there are further warning signs of a crisis, in which case they could start cooking the books even more than they do now to misrepresent the picture.”
Standard Life’s head of global strategy Andrew Milligan says although official figures are for public consumption, one of the difficulties facing the authorities in their economic reform effort is a lack of detail about what’s going on in the economy.
“We’ve seen several statements from the premier and president along the lines of the quality of growth is more important than the quantity,” says Milligan. “Local officials have been encouraged not to make up the figures and to instead produce more credible data; not numbers manufactured because they think Beijing wants to see high growth rates.
“There probably are other statistics which are not published, say on the state of the labour market. Given the level of detail for Chinese economic statistics, which is understandable for a fast-changing, large, low-to-middle income country, it’s not surprising that the authorities find it difficult to know precisely how to steer the economy.”
Milligan adds: “Economists find it difficult in the UK, Europe and US to work out what’s really going on – the whole UK productivity conundrum, for example – and we have masses more data in such countries than we do in emerging nations, such as China.”