Just How Believable Are Economic Numbers Coming Out of China?

The man set to become Chinese prime minister in early 2013, Li Keqiang, became prematurely famous last month when he was reported as admitting that China’s gross domestic product figures were “for reference only.” This comment, from the future leader of the world’s second-largest economy, will be used to dismiss Chinese government data for years to come.

This was not Li’s only comment, however.

Li was also quoted as saying something far more valuable: He cited rates of bank lending, rail cargo and electricity consumption as signaling the true health of the Chinese economy.

Although Li’s method, in keeping with his background, places too much weight on industry, it also yields interesting and reasonable insights.

In particular, China’s economic performance looks stronger than official figures for 2010 indicate.

Economic performance during 2008 and 2009, however, may have been much weaker than official numbers suggest.

This greater instability, and the desire to hide it, shows that China is playing with a weaker hand than it first appears.

According to China’s State Statistical Bureau, GDP growth during 2010 was 10.3 percent, bringing the economy to just over $6 trillion.

Perhaps surprising, most other official figures imply that the SSB actually understates Chinese growth for 2010.

For instance, the consumer price index is reported to have climbed 3.3 percent. However, this is the smallest gain among China’s many price measurements.

A better price measurement is the GDP deflator — the difference between arithmetic GDP growth and announced real growth — which for 2010 was 6.4 percent.

Either true inflation is notably faster than the consumer index shows, or the GDP deflator is too large and real GDP growth is higher than announced. Either way, the economy ran hotter than the SSB claims.

Components of GDP reveal another problem.

Fixed-asset investment and retail sales are benchmark measures for investment and consumption, respectively, but add up to more than China’s GDP by themselves.

Along with the trade surplus, which is the world’s largest, the three main components of GDP add up to far more than GDP.

This is because fixed investment and retail sales do not correspond to investment and consumption as understood outside of China. The SSB nonetheless insists on reporting them.

Fixed investment is an absurd 70 percent of GDP, twice the proportion of just eight years ago.

Perhaps when fixed investment exceeds 100 percent of GDP, Beijing will acknowledge that it is a useless measure.

A weakness in the official figures is that GDP growth consistently outpaces personal income growth. It is possible that rural incomes may have, at times, grown faster than GDP in 2010, depending on the price measure used. Urban incomes, however, rose considerably more slowly. China’s rapid official growth does not provide equally rapid gains for its people.

Finally, the official report on 2010 will be revised.

Every Chinese GDP revision to this point has been the same in two crucial ways: (1) It is always higher, and (2) it is always incomplete.

GDP growth is always said to be higher than initially estimated, but revisions for many figures are not published, so the numbers are rendered incomparable.

China’s revisions reduce transparency and credibility and make Li seem wise.

One view of Chinese economic figures is that the Chinese Communist Party is obsessed with stability, and this obsession leaks into data reporting to the extent that the highs and lows of China’s economic performance are dampened.

This hypothesis is supported by examining, as Li suggested, rail and electricity.

Growth in rail traffic is generally slower and bounces around more than GDP, but the size of the gap between the two was a maximum of 4.4 percentage points from 2001 to 2007.

In 2008, the gap reached 5.0 points.

In 2009, it was 8.4 points, as rail traffic slowed a great deal but GDP was said to slow only slightly.

This is simply not credible. In 2010, rail traffic seems to have leaped from far slower than GDP to faster than GDP for the first time in more than a decade. Li might find this implausible—and rightly so.

Officials measure electricity production rather than consumption, so that all figures stem from a single source (electricity production is also more stable than consumption).

Chinese data shows that growth in electricity production was faster than GDP from 2001 to 2007.

In 2008 and 2009 it was slower, then in 2010 it was faster again.

This pattern can be traced back even further: in difficult years, electricity production is slower than GDP; in all other years, it is faster.

However, it is highly unlikely that China rotates between splurging on electricity and energy-efficiency.

It is also unlikely that rail freight growth can collapse while GDP growth barely budges.

Li’s theory argues, in contrast, that Chinese GDP slowed far more than the SSB admitted during 2008 and 2009, and accelerated faster than it reported in 2010. The last element of Li’s triumvirate of indicators, bank lending, is more complex.

It is difficult for large amounts of rail traffic or power production to escape Beijing’s watch.

But it is much easier for large amounts of lending and borrowing to occur outside of formal banking channels, especially when cities and provinces are encouraged to borrow.

In addition, commercial bank lending accelerates as the economy accelerates. Policy-driven bank lending, in contrast, rises to bolster a weak economy.

Bank lending, therefore, does not have a clear connection to the true speed of growth.

Instead, it is linked tightly to central government objectives. It is the main tool for the Chinese government to encourage growth with conventional fiscal policy involving far smaller sums of money.

In that sense, Li was right, again, to view bank loans as vital to assessing the health of the economy.

Over the past 10 years, if official figures are accurate, China averaged 17 percent loan growth and 10 percent GDP growth.

This level of loan growth is unsustainable, but it is distorted by the panicked 2009 stimulus.

The 2010 performance is more worrisome. Official 2010 GDP growth is average, but the lending needed to achieve it is well above average.

Compared to 2004 to 2007, unpleasantly faster lending is required for the same growth.

This is an economy noticeably weaker than five years ago.

The obvious issue is that growth rates in rail traffic and electricity production indicate that the GDP figures are not accurate.

In particular, 2008 and 2009 were weaker than acknowledged, and 2010 was stronger.

It is difficult to draw conclusions from these findings, but if the economy was very weak in 2008 and 2009, hyper-aggressive lending might have been appropriate.

However, the continuation of rapid lending in a strong 2010 constitutes a major policy mistake.

From 1998 to 2007, China worked to address problems in banking.

The work has now been undone by forced policy lending.

Officials estimate hidden local government debt at 10 trillion yuan ($1.5 trillion), a number that rises every time the government mentions it.

Yet local borrowing continues, as do other errors — tightening to this point is a myth.

Interest rate increases have not kept up with inflation, so real interest rates have actually fallen.

If GDP growth is accurate, China’s economy is structurally weakening.

If, as is more likely, GDP growth is artificially stable, a damaging policy error was made last year.
By Derek Scissors, PhD, research fellow in Asian economic policy in the Asian Studies Center at The Heritage Foundation. Jakarta Globe