There is a new working paper out from the Bank of Japan, written by four authors from its international department, considering the prospects for China’s productivity growth over the long term. It’s pretty interesting and not just the usual stuff. One of the novel aspects is its analysis of food security. While that’s not a traditional topic for a productivity analysis, the logic is pretty straightforward.

Part of the growth in China’s labor productivity is coming not just from improving the productivity of specific industries, but from the shift of workers across industries (since China’s workforce stopped growing about a decade ago, labor productivity growth accounts for virtually all of its GDP growth: GDP growth = growth in output per worker + growth in number of workers). The structural transformation of the economy–moving workers from sectors where they produce less value-added, like agriculture, to those where they produce more value-added, like manufacturing–is one of the fundamental motors of economic development.

The authors estimate that out of China’s 10.7% average annual productivity growth from 2006-10, moving workers across sectors accounted for about 2 percentage points, and about 1.5 percentage points of the 7.6% growth from 2011-15 (these numbers are not given in the text, so I estimated them from a chart; apologies for any inaccuracy). But the reallocation of labor across sectors contributed less than 0.5 percentage points of the 6.9% growth over 2016-19, so a slowdown in structural change does seem to a factor in China’s overall growth slowdown.

The authors suggest that if the government’s concerns for food security lead it to try to keep workers in agriculture, that could impede structural transformation and therefore slow growth. They model a future growth trajectory for China that suggests that if its industries continue on the same path of productivity convergence and structural change as Japan, South Korea, Taiwan and Singapore, it can grow by an average of about 4.8% until 2035. But that trajectory of structural transformation would mean a continued sharp decline in the agricultural share of employment, and thus (since labor productivity growth in agriculture is slow) in agricultural output. So a continued rapid pace of structural transformation may not be compatible with food security. Here’s the relevant passage from the paper and the supporting charts:

The first issue concerns the balance between food security on the one hand and the shift of labor from agriculture to other industries on the other. In China, agriculture at present accounts for a larger share of GDP than in the East Asian 4 when they were at a similar per capita GDP level, reflecting the Chinese government’s policy of achieving a rate of food self-sufficiency of 95% or higher. However, in the baseline estimate, the employment share of agriculture and the share of agriculture in GDP will decrease significantly in the future due to a combination of the movement of labor across sectors and the effects of the aging of the population. As a result, real output in agriculture would drop to less than 40% of the current level, which means that China would have to effectively abandon food self-sufficiency.

However, in practice, it is unlikely that the Chinese government will tolerate such a change in industrial structure from the perspective of food security. Therefore, to consider a more realistic path, we assume that the shift of labor from agriculture will be limited to an extent that maintains the current level of real output in agriculture. In this case, GDP in 2035 would be about 10% lower than in the baseline estimate and only 1.87 times the current level.

It is possible to quibble with some of the details in this analysis. The Chinese government does not actually have a target of 95% self-sufficiency in all food supply. In the past, there had been an official target of 95% self-sufficiency of staple crops (liangshi 粮食, which is usually translated as grain but also includes beans and tubers). But this was difficult to enforce, and in practice large imports of soybeans have been tolerated. Xi Jinping proposed adjusting this policy early in his tenure to have a more realistic target (see my previous post from 2015 on the food security policy debate).

The white paper on food security published in 2019 mentions 95% self-sufficiency in cereals (rice, wheat, corn), but as an achievement rather than a strict target. The white paper does articulate an overall goal of self-sufficiency–“China makes sure it relies on itself for food supply”–but this is not given a strict quantitative definition, which allows the government flexibility (see this analysis from the excellent Dim Sums blog on Chinese agriculture). The government’s actual goal is probably to maintain certain levels of output of staple grains rather than to limit all food imports. It’s likely that could be accomplished even as the shares of agricultural employment and value-added continue to decline.

Nonetheless, I think the paper’s point that official food security concerns can act as a brake on structural change is correct. China’s actual agricultural policy has been fairly conservative, in the literal sense of trying to conserve an existing order. The trade war with the US reinforced the risks of relying on imported food, and Xi urged more focus on domestic production. While there may not be hard target on the acceptable level of food imports, there is a general push to maintain a large population of agricultural workers and slow rather than accelerate their shift into other sectors. Xi’s government for instance is encouraging rural residents who migrated to the cities for nonagricultural work to return to rural villages. The government has been clear that it wants to preserve the collective system of rural land ownership, which prevents farmers from being dispossessed of their land but also limits their freedom to leave it (for more on this point, see this Dim Sums post from March).

The paper is entitled “China’s Long-Term Growth Potential: Can Productivity Convergence Be Sustained?” and is worth a read.