Whether one feels positive or negative about China’s economic prospects, everyone can agree that its indicators are a mess. But the fact that the data are flawed does not mean that they are deliberately manipulated to yield a particular outcome.

These distortions affect two key policy concerns that dominate public attention. One has to do with purportedly overstated gross domestic product growth rates. The other involves perceptions about China’s unbalanced growth as reflected in its extremely low consumption share of GDP and high investment share.

Yukon Huang
Huang is a senior fellow in the Carnegie Asia Program, where his research focuses on China’s economy and its regional and global impact.
More >

Many China watchers believe that officials have deliberately overstated economic growth figures to meet plan targets, citing for example that the sum of provincial growth rates exceeds the reported national rate. Local officials probably do exaggerate but other factors such as double-counting of cross-provincial activities and differing methodologies play a larger role which the centre then reconciles through separate surveys.

The growth rate debate is often driven by technical assumptions about price deflators and input-output relationships. A Conference Board expert will use an alternative methodology to argue that annual GDP growth rates over the past three decades were closer to 7.5 per cent than the reported historic 10 per cent. But another expert then rebuts that this comes from assuming no improvement in the quality of China’s industrial production, which defies logic.

Some observers have used proxies such as electricity sales to suggest that GDP growth was exaggerated since the growth rate of electricity sales fell much more than that of GDP during the global financial crisis. But others realise that energy sales normally decline more than GDP during downturns but increase more rapidly during expansions because of differing intensities in energy use across industries. Overall, growth in electricity production has actually been much faster than GDP growth over the past decade.

Of more significance are the distortions originating from China’s central planning days when a “material balance approach” was the basis for deriving GDP rather than the standard UN system of national accounts which China adopted but is still fine-tuning. China’s statistics have not been fully cleansed of Beijing’s former bias in favour of industrial production relative to services and neglect of informal private activities. This means that the size of China’s economy has been systematically underestimated rather than deliberately biased upwards.

This has been borne out by periodic “benchmark revisions” undertaken with international experts to revise the magnitude and composition of GDP. The last two rebasing exercises were done in 1993, which increased the nominal size of the service sector by 32 per cent and GDP by 10 per cent, and in 2004 which led to another large increase in the share of services in GDP and upped the annual economic growth rate from 9.2 per cent to 9.9 per cent for the previous decade. Ironically, the government was not enthused about these upward adjustments since it weakened China’s case for favourable treatment in international trade and aid negotiations.

Many people see the extremely low share of consumption to GDP as indicative of repressed spending and the high share of investment as waste. Yet few pause to wonder why until the recent slowdown, the news was all about double-digit sales growth rates in the luxury goods, fast-food outlet and home furnishing sectors. How can one reconcile retail sales that were growing at 15-20 per cent a year for decades with GDP numbers that suggest lacklustre growth in personal consumption? This raises suspicions that something is amiss, quite possibly that consumption is seriously understated. A 2008 Morgan Stanley study estimated that Chinese GDP was about 30 per cent higher than official figures, and per capita consumption was as much as 80 per cent higher. Some research surveys show that household income has been understated by some 20-30 per cent of GDP.

There is some official support for such views. The head of the national accounts department at China’s bureau of statistics acknowledged in 2009 that its household consumption figures are deficient. The statistics are based on an obsolete, 30-year-old sample survey; do not take full account of cash transactions; do not include fully non-cash provision of social services; and have not been adjusted to reflect the market values for owner-occupied housing. Moreover, as a result of high sales taxes, businesses often do not issue receipts leading to household purchases being underreported.

These conclusions have been given more credence by two Shanghai-based professors, Jun Zhang and Tian Zhu, who concluded that personal consumption was significantly underestimated by as much as 15 percentage points of GDP. This is due to underreporting of housing expenditures; company-provided consumption-related benefits; and informal household income.

By implication, estimates of the share of investment in GDP are too high. This stems partly from the practice among many companies and government entities to package consumption-type expenditures as investment because of the lingering belief that consumption is bad and investment is good. In addition, GDP estimates of investment do not adequately adjust for the rising cost of land which has soared in recent years.

Overall, the personal consumption-to-GDP ratio might be closer to 45 per cent rather the reported 35 per cent and the investment ratio about 38 per cent instead of 48 per cent. If so, then China’s consumption and investment ratios are in line with its Asian peers such as Japan, South Korea and Taiwan during their comparable stage of development.

China’s statistics are flawed but there is probably no systematic intention to bias the results in one direction or another. Contrary to popular beliefs, the existing distortions may actually be giving a more negative impression than the reality.

This article was originally published in the Financial Times.