Government statistics – still not fit for purpose

November 21, 2013 by Ken Davies

Although China’s statistical system has improved enormously in the past three decades, it remains distorted by political imperatives and vested interests. It needs to be reformed and allowed genuine independence so that policy-makers and the public can know what is really happening in the Chinese economy.

Following double destruction in the Great Leap Forward (1958-1959) and the Cultural Revolution (1966-69), China’s statistical system has been rebuilt and greatly expanded since economic reform started at the end of 1978. From the early 1980s, the National Bureau of Statistics (NBS) has published a fat statistical yearbook packed with figures, as have provincial administrations, ministries and major institutions. All of these also maintain more or less frequently updated websites full of figures.

The NBS is proud to be part of the process of improving national statistical systems led by the IMF, under whose guidance it has promised to publish its data more frequently and use modern techniques such as statistical sampling to supplement the total counts that are a legacy of the former central planning system.

In the 2000s, a more sophisticated array of economic statistics, based on those in countries like the United States, began to be developed, including a purchasing manager index and opinion polls of producers and consumers. So does China now have economic statistics up to international quality standards? Can the leaders rely on them for managing the economy? Can citizens, producers and consumers trust them? The answer to all those questions is: not yet. There are serious problems with all the major economic indicators.

In the early 1990s, economists urged the Chinese government to move from publishing production-side gross domestic product (GDP) figures (which record the origin of output as being from industry, agriculture and services) to adding standard expenditure-based GDP figures (which divide output into such categories as investment and consumption), as used in developed economies.

By the turn of the century, expenditure-side GDP was included in the China Statistical Yearbook, but it now looked as though the figures were exaggerated. The economist Thomas Rawski showed that the 1998 and 1999 figures were far higher than was indicated by other published information, and another economist, Ken Davies, in a paper entitled “Lies, damned lies and Chinese statistics”, demonstrated that this had been the case since the 1980s.

Have things improved since then? No. Just as in previous decades, national GDP and GDP growth figures for 2011 and 2012 were below the provincial total and average, indicating over-reporting at provincial level. The sum of provincial GDPs in 2012 was CNY 57.6 trillion, CNY 5.7 trillion more than the CNY 51.9 trillion national figure. In 2011 there had been a similar discrepancy of CNY 4.6 trillion.

Another typical case of almost certain over-reporting was the announcement that GDP had grown by 7.5% year-on-year in the second quarter of 2013, which included at least one month (April) when exports had been massively exaggerated (see below), and was inconsistent with other figures suggesting a more rapid slowdown.

Provincial over-reporting of GDP is the result of what some consider to be the evolution from a party-state oligarchy to a meritocratic civil service: promotion of provincial officials to national positions of power is nowadays less based on political loyalty and more on performance. Unfortunately, the measurement of performance is in the hands of those in power at provincial level.

China’s leaders occasionally complain that the books are cooked. When he was Party leader in Liaoning in 2007, the current Prime Minister, Li Keqiang, told the then US Ambassador to China (according to Wikileaks) that he didn’t rely on official GDP statistics, which were “for reference only” (ie useless), but instead looked at electricity consumption, rail cargo volume and bank lending – rather like many bank economists outside China.

The NBS occasionally rebukes local officials for forcing businesses to report false numbers to inflate figures. Yet many economists, both within and outside China, seem to take it for granted that official expenditure-side GDP figures are genuine, despite such evidence of manipulation. Anecdotal evidence suggests that the real (though still exaggerated) figures that form the basis for GDP growth pronouncements are those collected for production-side GDP, based on a reporting system that was developed for Soviet-style net material product (NMP) compilation of GDP.

For example, one independent provider of statistics (whose board of directors included people who were also on the board of the NBS) some years ago offered to provide annual GDP figures for every county in China. One potential customer asked how this was possible, given the difficulties of calculating GDP for just a few, much larger metropolitan areas in developed countries. “You could only,”said the naïve customer,”perhaps do that if you just added up the gross value of industrial and agricultural output and discounted it by the GDP deflator.” “Of course,” was the answer,”How else would you calculate GDP?”

There is also circumstantial evidence that the central authorities, like most Chinese companies, maintain more than one set of books.

In 2013, it is no longer just GDP statistics that are subject to widespread scepticism. The General Administration of Customs reported that in April this year exports grew 14.7% year-on-year, which looked extremely unlikely in view of the sharp decreases in imports from China experienced by the country’s trading partners. Why? Massive fake export invoicing had been used to conceal inflows of hot money generated by the Fed’s loose money policy. A crackdown on this practice resulted in a sharp fall in exports reported in May.

Another important data point for policy-makers, total lending, has also been cited as being wrong. In the first quarter of 2013, total social financing (a new measure created in 2011) was reported to have increased by CNY 1 trillion to reach CNY 1.6 trillion, despite many indicators that growth was slowing. Unnamed sources in the People’s Bank of China (the central bank) suggest that double-counting was occurring.

Inflation figures are not trusted by consumers. This happens in other countries, but in China’s case the mistrust is also shared by economists who consider that underestimating inflation contributes to bloated GDP figures in the form of an understated GDP deflator.

The statistical system needs comprehensive house-cleaning. One of the yardsticks by which to judge the outcome of the Third Plenum of the 18th Central Committee of the Chinese Communist Party will be the effectiveness of measures to render the country’s biased statistical system objective and independent so that it can play a real role in policy-making.

Leave a Comment

Your email address will not be published. Required fields are marked *

Thankyou for visiting China Outlook

You have read one of your two free articles on China Outlook. To read your second free article, please sign up to our mailing list below:

Submit info and continue reading

If you want to receive full access to all China Outlook articles, plus our archive, please subscribe using the box below.


Thankyou for visiting China Outlook

You have now reached your limit of two free articles.. To continue reading please subscribe.

Subscribe Back to home