Chang-Tai Hsieh on the Inaccuracy of Chinese Economic Growth Statistics

Chang-Tai Hsieh conducts research on growth and development. Hsieh has published several papers in top economic journals, including “The Life-Cycle of Plants in India and Mexico,” in the Quarterly Journal of Economics; "Misallocation and Manufacturing TFP in China and India," in the Quarterly Journal of Economics; "Relative Prices and Relative Prosperity," in the American Economic Review; "Can Free Entry be Inefficient? Fixed Commissions and Social Waste in the Real Estate Industry," in the Journal of Political Economy; and "What Explains the Industrial Revolution in East Asia? Evidence from the Factor Markets," in the American Economic Review.

Hsieh has been a visiting scholar at the Federal Reserve Banks of San Francisco, New York, and Minneapolis, as well as the World Bank's Development Economics Group and the Economic Planning Agency in Japan. He is a Research Associate for the National Bureau of Economic Research, a Senior Fellow at the Bureau for Research in Economic Analysis of Development, and a member of the Steering Group of the International Growth Center in London. 

He is the recipient of an Alfred P. Sloan Foundation Research Fellowship, an Elected Member of Academia Sinica, and the recipient of the Sun Ye-Fang award for research on the Chinese economy. 

Hank Snowdon CMC 21' interviewed  Chang-Tai Hsieh on April 10, 2019.

China has long been accused of manipulating its economic statistics, and your research confirms that China’s reported growth numbers are overstated. Can you explain exactly what your research found?

I wouldn't quite put it that way — I don't think that it's correct to say that the Chinese manipulate the GDP statistics. Fundamentally, the underlying data is collected by local governments, and the National Statistical Bureau has very little control over what they do. The national bureau takes that data and is forced to make adjustments to the numbers. It's not manipulation of the data, but it's rather about the statistical agency not having enough resources or the authority to do the job they're supposed to do. This is where the mistakes come in, due to this underlying institutional structure. Coming up with this number is a very difficult task. You need to have lots of resources and a lot of authority. The underlying structure of the Chinese administrative system is such that the body that's responsible for putting together these numbers doesn't have the necessary authority or the budget.

When we did checks on the data we found that the GDP growth rates for the last 10 years have likely been overstated. But before that the numbers were roughly right. And it probably was the case that before, in the 1990s, the growth rates might have been understated.


Can you briefly explain your approach to estimating China’s real GDP numbers?

What we did was check the underlying data that was used. We looked at the underlying microeconomic data, and we checked to see which parts were reliable and which parts of the data were unreliable. After that, we tried to understand where exactly the National Statistical Bureau made adjustments to the numbers provided by the local authorities. Then, in the places where the data didn't seem to match, we looked at the data for tax revenue and we used that to back-out what the underlying true numbers must have been.

So it's really these three things. It's first about checking the underlying data — the data is based on something, so we went to the same sources, and we checked which parts of the data were reliable and which were not. Then we tried to figure out exactly where the adjustments were made, and in the parts where the data didn’t check out, we tried to come up with numbers that were better.

Even though we do see that the data is likely overstated by 2% per year since about 2008, most of the components of that seem to be roughly right — there were basically two main components that seemed to be off. On the expenditure side, the investment data seems off, and on the production side, it's the data on manufacturing output that is incorrect.


Why has the gap between China’s reported and actual GDP growth increased significantly in the past decade?

I don't know, I’m only going by what the data says. It’s still the case in the last ten years that they’ve been making large adjustments to the data, so it’s likely that the magnitude of the adjustments should have been much larger recently. But why haven’t they been making the bigger adjustments? I’m not sure.


What are the implications of China’s output being lower than reported? Are they in even worse shape during the current slowdown than has previously been assumed?

There's no question that the slowdown is worse than how it has been reported. But what is also true is that the investment rate is much lower than what the official data says it is as well. The Chinese surplus, with respect to the rest of the world, used to be 10 percent of Chinese GDP, but now it’s down to about zero. Chinese growth does not seem to be driven by investment, nor does it seem to be driven by exports. The flip side of this decline in investment is that the consumption rate has gone up.


If you were to recommend a fix to the Chinese government, what would it be?

When you dive into the details of any national accounts, what you find is that there's a lot of judgment calls. The issue that plagues the Chinese data is not that unusual relative to the issues that face any country in the world. It’s a very difficult task to come up with a number that captures the complexity of an economy that size.

The answer to the question is that it's about providing more resources and authority to the statistical authorities, but given the Chinese institutional structure, it’s very hard. In general, the statistical apparatus doesn't seem to be any better or any worse than what I see in in the typical country. If you carry out a similar exercise for any country, you are going to find periods where probably the truth is understated or overstated.


Top Chinese leaders themselves don’t trust official data.  Prime Minister Li Keqiang famously said that GDP data are for “reference only”.  How do Chinese leaders make key economic decisions when confronted with data they know are faulty if not fraudulent?    

If you're a smart policymaker, you are going to know that there are a lot of things that are not being captured perfectly in the data. So if you're smart, you look for alternative sources of data as well. The quote by Li Keqiang is not about the national numbers, it’s about the regional numbers, which are indeed notoriously bad.

Hank Snowdon CMC'21Student Journalist

Picrazy2 [CC BY-SA 4.0 (]

Share this:

Leave a Reply

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