Skip to navigationSkip to content
A worker is pictured in front of a billboard of local real estate advertisement in Chongli county of Zhangjiakou, which is jointly bidding with the capital Beijing to host the 2022 Winter Olympic Games, July 30, 2015. Beijing, which hosted the 2008 Summer Olympics Games, hopes to beat Almaty, Kazakhstan, and win the 2022 Winter Olympics, when a decision is made on Friday in the Malaysian capital, Kuala Lumpur. REUTERS/Jason Lee
Reuters/Jason Lee
Tougher to find work than it seems.

China’s unemployment rate is much higher than the government says it is

By Gwynn Guilford

Of the many reams of iffy Chinese economic data, the iffiest of all must be the urban unemployment rate.

Behold how remarkably flat it is:

That’s far stabler than in other countries (even in Japan, where policies make it hard to lay off workers):

Though most economists have a pretty tough time putting their faith in China’s unemployment numbers, few alternatives exist.

However, a new working paper (registration required) published in the National Bureau of Economic Research has taken a stab at calculating China’s actual unemployment rate. And surprise, surprise—it’s a lot different from what the government reports. Take, for example, the researchers’ estimate that between 2002 and 2009, the urban unemployment rate averaged a staggering 10.9%.

"Long-run trends in unemployment and labor force participation in China," Feng et al. 2015

The paper, which is based on a 22-year-old national household survey that is administered by the Chinese government’s statistics bureau, breaks its analysis into three periods:

  • The SOE heyday: From 1988-1995, state-owned enterprises (SOEs) still dominated the economy. The average unemployment rate of 3.9% in those years was only slightly higher than the official figure.
  • The SOE reform era: From 1995 to 2002, as the government spurred market reforms that led to mass layoffs of SOE workers, the rate climbed to an average of 6.6%. During those years, the share of workers employed by state companies fell from 60% to 30%.
  • The manufacturing boom: You’d think that things would improve during 2002-2009, when China’s World Trade Organization entry sparked a labor-intensive manufacturing boom, and the expansion of college enrollment improved labor force quality and absorbed young would-be workers. However, as the 10.9% average suggests, these weren’t enough to offset the lingering effects of the SOE layoffs. The global financial crisis didn’t help.

This, say the authors, suggests that as of 2009, China’s labor market still hadn’t recovered from the period when SOEs dominated the economy. They also argue that China’s unemployment rate is higher than it should be given how wealthy the nation as a whole has become.

There are some big caveats to the findings. First off, the way employment is calculated varies by country, making it hard to make apples-to-apples comparisons.

Of a bigger concern is that the dataset doesn’t reflect rural migrant workers, even though they have been a huge source of urban labor in the last decade, making up as much as half of China’s urban workforce today (pdf, p.11). That could mean that the actual unemployment rate is lower than the authors calculated. On the other hand, nearly 60% of rural migrant workers work in construction and manufacturing (link in Chinese). Given China’s slowing economy and sluggish export demand, they might soon be pushing the true unemployment rate up.

Of course, the dataset ends in 2009, making it even harder to guess the current job market conditions. Perhaps the only thing certain about China’s changing labor market is the official unemployment rate trend: don’t count on it to budge.