ravallion4febfig1.jpg
VoxEU Column Poverty and Income Inequality

A historical perspective on China’s success against poverty

The extraordinary reduction in poverty that China underwent after 1980 is often attributed to the pro-market reforms of Deng Xiaoping. This column uses counterfactual analysis – comparing China’s development to neighbouring countries with similar cultures and strong historical ties – to propose a new perspective. When judged against the development of South Korea and Taiwan, the bulk of China’s progress since the reforms began seems mostly a matter of making up for the failures of the preceding 30 years, when Maoist policies left an extra quarter of the Chinese population in poverty.

There is no doubt that China has seen a large reduction in poverty since 1980. Judged by the World Bank’s $1.90 a day poverty line (in 2011 prices at purchasing power parity), the national poverty rate fell from almost 90% in 1981 to under 4% in 2016, implying 800 million fewer people living in poverty (using the World Bank’s PovcalNet site). Many explanations have been offered for China’s accomplishment in reducing extreme poverty, including frequent emphasis on the pro-market reforms initiated in 1978 by Deng Xiaoping, who ruled China until 1990. Deng’s initial focus on agrarian reforms – notably, de-collectivisation and restoring market-based incentives for farming – has been identified as a major factor in output gains (Lin 1992) and the country’s progress against poverty in the 1980s (Ravallion and Chen 2007). 

However, any evaluative interpretation of a measure of social or economic outcomes over time requires consideration of a relevant counterfactual trajectory. That is especially important when drawing lessons for policy. If prior policy mistakes left a high initial level of poverty at the outset of a reform period, then we would be inclined to evaluate the overall performance less favourably, even while giving credit for subsequent corrections to those mistakes. When we applaud post-reform success, we cannot ignore pre-reform failure. In that light, there may not be much left to explain. 

In the present context, the fact that Deng’s policies are universally seen as ‘reforms’ begs the question of whether their success was simply the undoing of the prior mistakes that called for those reforms. For almost 30 years, the pre-reform policy regime in mainland China followed a path laid out by Mao Zedong from the early 1950s. To some extent, Deng’s reforms were enacted to correct Mao’s failures; otherwise, why were the reforms necessary? Quantification is crucial to a more thorough evaluation: Just how much of the subsequent decline in poverty in China can we attribute to undoing past mistakes versus creating a new trajectory of progress? The literature has been silent on this question.

My new research paper (Ravallion 2021) adopts an explicit counterfactual perspective on China’s progress against poverty since 1950. This requires a defensible counterfactual, in the sense that it can be argued that another option was available. The historical record suggests that two relevant counterfactuals for China in the Maoist period were geographically and culturally close at hand around 1950, namely South Korea and Taiwan (SKT). The peoples of China, Taiwan, and South Korea share many common cultural features, with Confucian philosophical roots. Notable similarities include their work ethic, their respect for harmonious integration in society (rather than individual gratification), the reverence attached to learning, and the importance attached to the family, including in production and especially in peasant farming. These strong historical ties extend back to ancient times. 

The armed conflicts, political realignments, and economic setbacks within Northeast Asia during the first half of the 20th century left China, Taiwan, and South Korea very poor by 1950, though China was clearly poorer. A number of observers described the extreme poverty of rural China in the decades prior to WWII. Virtually all China’s rural peasantry faced regular hunger in what Mallory (1926) dubbed the “land of famine”. During the ten years after the end of WWII, two relatively new but very different economic models emerged out of the poverty of Northeast Asia. Like mainland China, both South Korea and Taiwan had historically been supportive of governmental intervention in production, and this did not change after 1950. The point of departure was in whether the state should actually own the means of production. The difference was not just between ‘communism’ and ‘capitalism’ but between the Maoist version of communism and a type of capitalism that emerged in SKT and in post-Mao China – what Milanovic (2019) dubs “political capitalism”. My new paper argues that, based on the historical record, a version of political capitalism was a viable option for China around 1950. The country took another path, reflecting the personality and power of Chairman Mao, and his desire to outdo his rival Joseph Stalin in implementing rapid socialist industrialisation (Li 2006). This came in no small measure at the expense of the poor rural economy. There was much debate, at the highest levels, about this path at the time. Mao prevailed. 

The historical data for quantifying the impacts on poverty are far from ideal. The World Bank’s PovcalNet does not include surveys prior to 1980. In going back to 1950 (and before), the single best source is a 2002 paper by Bourguignon and Morrisson (BM). Their paper does not provide country-level estimates of poverty measures, but the authors kindly provided their income shares by decile (with the richest decile split into two ventiles) and GDP per capita for the 11 years spanning 1820–1992 (at irregular intervals). The dataset includes China and SKT. I have used these data to estimate poverty measures over time using parameterised Lorenz Curves and BM’s “extreme poverty line”. I then use a difference-in-difference (DID) estimator of the impacts of the Maoist policy path from 1950–1980, assessed against the SKT counterfactual. My paper also provides various checks and tests of the estimates, including placebo tests using the data prior to Mao taking control of mainland China in 1949; the available historical data go back to 1820. There is no sign of anything one could call a significant impact of the Maoist path prior to Mao taking over (Ravallion 2021). 

Using aggregate SKT as the counterfactual, my paper argues that the bulk of China’s progress against poverty since the reforms began was indeed making up for the country’s lack of progress against poverty over the preceding 30 years. The following table summarizes the main results. The counterfactual poverty rate for China is 14.5%, as distinct from 41.6% based on the actual data available for China.

Table 1 Estimates of the poverty impact of the Maoist path for China 

 

Source: Author’s estimates; impact estimates are for each year relative to 1950. See Ravallion (2021) for further details on estimation methods.

When judged against the development paths of South Korea and Taiwan, my calculations suggest that the Maoist path meant that an extra quarter of the Chinese population were living in poverty by the time Deng’s reforms began. Considering the possible biases in the available data, it seems more likely that this is an underestimate than an overestimate of the poverty impact of the Maoist path (Ravallion 2021). While data uncertainties remain, it is clear that a large share of China’s post-reform reduction in the incidence of poverty can be thought of as the country’s success in correcting past failures in its economic policies. 

How long did it take for post-reform China to catch up to its counterfactual standing? The Figure below provides the time series for my estimates of the post-1980 poverty rate in China using two poverty lines that can be considered roughly comparable to the historical lines the new study used, based on BM but using 2011 prices. Those lines are $0.90 a day and $1.40 a day. The Figure also indicates my counterfactual poverty rate for 1980 of 14.5%. By 1990, just after Deng resigned as leader, the post-reform trajectory of (rapid) poverty reduction had fully made up for the estimated ‘lost ground’ attributed to the Maoist regime. However, by the higher $1.40 line, this did not happen until 2003, though 50% of the gap between the 1981 rate and the counterfactual was reached by the end of Deng’s rule.

Figure 1 Poverty rates for China and counterfactuals for 1980

 

Source: Author’s estimates using PovcalNet (accessed 3 January 2021). 2011 prices use the World Bank’s purchasing power parity rates for consumption. Nearest-neighbour smoothed scatter plot for the poverty rates.

Two further insights emerge from the paper’s analysis of the historical data. First, a closer alignment of poverty reduction trajectories between China and SKT had already begun by the time Deng’s reforms commenced. Indeed, virtually the entire impact of the Maoist regime on poverty occurred in its first 20 years. This is consistent with the greater emphasis given to rural development in the late Maoist period (Bramall 2009). Second, a large share of the extra poverty in 1980 attributed to the prior Maoist regime was eliminated less than ten years into Deng’s reform period, when the primary focus was agrarian reforms. 

References

Bourguignon, F and C Morrisson (2002), “Inequality among World Citizens: 1820–1992”, American Economic Review 92(4): 727–744.

Bramall, C (2009), Chinese Economic Development, London: Routledge.

Li, H Y (2006), Mao and the Economic Stalinization of China, 1948–1953, Maryland: Rowman and Littlefield.

Lin, J Y (1992), “Rural Reforms and Agricultural Growth in China”, American Economic Review 82: 34–51.

Mallory, W H (1926), China: Land of Famine, New York: American Geographical Society. 

Milanovic, B (2019), Capitalism Alone, Cambridge, Mass.: Harvard University Press.

Ravallion, M (2021), “Poverty in China Since 1950: A Counterfactual Perspective”, NBER WP 28370.

Ravallion, M and S Chen (2007), “China’s (Uneven) Progress Against Poverty”, Journal of Development Economics 82(1): 1–42.

5,564 Reads