China’s Slowdown: More There than Meets the Eye

By Bob Barbera and Yingyao Hu.

If there isn’t light in the 21st century world – or at least not as much light as economic statistics would suggest there should be – then something may be amiss.

In China, that appears to be the case. And that suggests that there has been a sharp deterioration of growth there that began five years ago – even though it is only now beginning to show in official statistics.

Our work lends strong support to a notion that many have offered up concerning China’s economy over the past several years: a good deal of China’s gains in output, 2012-2017, reflected investment in unproductive assets. By downsizing the economic value of those investments we judge that net income flows in China, year-end 2017, were roughly 21% lower than one would judge them to be, based upon official China real GDP pronouncements.

Perhaps the most important implication of a downsized sense of China’s current aggregate income flows relates to its implications for China servicing its mushrooming debt. China’s debt-to-income ratio, according to official estimates, ended 2017 at roughly 300% of GDP. Our revised measure suggests that the debt ratio is dangerously higher.

Characterizing China’s aggregate economic performance, serious students of the Middle Kingdom know, is a daunting task. Seemingly paradoxically, forecasting growth rates for official Chinese real GDP is straightforward. There is, however, no paradox. China’s top line growth rate is meant to be met, and government actions are continually taken in an effort to do so. Furthermore, official economic reports are ultimately political documents. Accordingly, when the Chinese Politburo announces a 7% growth rate target for the upcoming year, the year-end official tally of output gain is destined to come very close to the goal.

Conversely, precisely because officials move both metal and statistics, to meet official goals, enlightened private sector handicappers of China’s evolving economic circumstances know they simply do not learn much about China’s economic health and well-being from official real GDP pronouncements.

A new paper, “Illuminating Economic Growth” offers a possible solution to this dilemma. The paper was written by Yingyao Hu, one of the writers of this blog post and an economics professor at Johns Hopkins, and Jiaxiong Yao, a JHU graduate student.

The paper describes a model that uses satellite photography to calibrate changes in evening light intensity. Using newfound techniques for measurement error models, the paper demonstrates that joint movement of official GDP measures and the evening light intensity allow one to infer the nonlinear relationship between the light intensity and the true GDP, together with the distribution of the latter.

The changes in the evening light intensity can be mathematically filtered in a way that, for most countries, results in a trajectory that tracks changes in output. For China however, that measure of real output growth bears little resemblance to official GDP statistics since the 2008-2009 Great Recession. China, according to the Hu/Yao model, suffered a much more substantial slide in economic performance in 2009. And growth appears to have been substantially slower from 2012 through 2017.

The data suggests three things. First, China suffered a very sharp slowdown in 2009, completely at odds with official Chinese data but perfectly logical given both the collapse for global trade that we know occurred and the extremely large share of China’s GDP that depended on trade in that period.

Second, the fade for net income gains over the past five years looks substantially more dramatic than official data suggests.

Third, directly as a consequence of the first two conclusions, cumulative growth for China’s net income looks to be roughly 21% smaller than China’s official tally of real GDP growth.

One naïve reaction to this disconnect would attribute all of the divergence to falsifying data. This is, however, hard to reconcile with other metrics. China employment, consumer spending and consumer confidence, all suggest that gains for jobs, wage income and spending held up over the period, along the lines captured in official data.

Nonetheless, official gains for overall national economic performance seem much stronger than what would be justified by the pace of nighttime illumination increases, identified in our model. Similarly, relatively steady global commodity prices and a persistently strong Chinese appetite for the buying of raw materials are at odds with the notion of severe Chinese output retrenchment.

We offer the following hypothesis. China, in the aftermath of the Great Recession, committed substantial sums for office, apartment and infrastructure construction. Rapidly rising vacancy rates, however, suggest these investments, over the period in question, delivered less and less economic value. In effect, buildings kept going up in increasing numbers, but fewer and fewer lights went on.1

We can, therefore, accept China’s official real GDP tallies. In turn, however, we calculate the rising rate of depreciation that allows the official real GDP growth rate to match the trajectory envisioned by our illumination model. China’s net domestic income, roughly output minus depreciation, thus grows much more slowly than output. Importantly, a debt-to-income ratio, using this net income metric, looks substantially more daunting than the official debt-to-GDP metric that most people use.

Obviously, severely underutilized investments are incapable of servicing their debts—the signature snapshot of an investment bubble. In a traditional capitalist economy, bubbles end when access to credit is cut off and bankruptcies soar. In China, where banks are recapitalized by the government without fanfare, market driven restraint can only arrive when Chinese officials decide they can no longer paper over losses. That said, the snapshot is unambiguously unsustainable, despite the fact that we cannot quite imagine what will precipitate an end to this dynamic.

Certainly shining light on Chinese debt excesses does not break new ground. Rapidly rising debt and deteriorating credit efficiency raise concerns about financial stability and a disruptive adjustment warned a paper published by two IMF economists, Sally Chen and Joong Shik Kang, nearly a year ago. They went on to note that “International experience suggests that such rapid credit growth is not sustainable and is typically associated with a financial crisis and/or a sharp growth slowdown.” They warned that action was needed to stem the rise in Chinese debt before it grew too large.

Discussing that paper this past summer, Martin Wolf, the Financial Times columnist, conceded that China’s government and central bank could protect the financial system from collapse if an immediate crisis arose. But, he added, “…if gross debt were to rise above 400 per cent of GDP over the next decade, even that would be less certain.”

Less certain, and much sooner. Using our revised measure of real income flows, and accepting official Chinese tallies of the gains for prices from 2007 to 2017, we estimate that net nominal income grew around 10.5% per year, in contrast to the 13.2% pace of gross domestic income. The 2017 debt-to-income ratio, using our income measure stood at 380% of GDP.

Moreover, the pace of debt growth over the past 10 years, has been 1.8 times faster than our estimate of the pace of nominal income growth. If that ratio holds true in 2018, then a rise of 6.5% for net income would put debt-to-income at 400%.

Thus Wolf’s worry about the arrival of a truly worrisome debt ratio, sometime “over the next decade,” seems mistimed. Using our measure of Chinese income streams, we would submit that China’s debt ratio is touching 400% just about now.

China Nominal Income and Debt Ratio Calculations

2007 – 2017
(Annualized)
Official real GDP 8.2%
Official nominal GDP 13.2%
Official deflator 4.6%
Hu/Yao real net income 5.6%
Hu/Yao nominal net income 10.5%
2007 2017
China real GDP 100 220.3
China nominal GDP 3560 12283
Hu/Yao real net income 100 173.2
Hu/Yao nominal net income 3560 9659
China debt 6230 36726
Official debt/income 1.75 2.99
Hu/Yao debt/income 1.75 3.80

Notes:
1. Michael Pettis, Peking University in Beijing, is a long-standing proponent of this view, arguing that China, increasingly over the past several years, has committed to nonproductive investment projects—projects whose value is less than the cost of investment.