The persistent impact of autocratic rule on social capital

Social capital is the capital that is embodied in inter-personal relationships. It encompasses networks, alliances, shared norms and values, reciprocity, and trust. The level and forms of social capital are determined by people's attitudes and beliefs. So, if political or economic institutions alter those attitudes or beliefs, then the level or form of social capital might also be altered.

That is the thesis underlying this working paper by Melanie Xue (Northwestern University) and Mark Koyama (George Mason University), which looks at the impact of autocratic rule in Qing dynasty China on subsequent measures of social capital. Specifically, Xue and Koyama look at the impact of literary inquisitions. As they explain:

Following the Manchu occupation of China in 1644, and the establishment of the Qing dynasty, imperial China saw a sharp increase in political repression and an entrenchment of autocratic rule. Intellectuals, the most influential figures in local society, saw new restrictions imposed on them. One watershed event was the intensification and routinization of literary inquisitions - investigations which targeted the speech and writings of intellectuals.

Essentially, Xue and Koyama compare prefectures that had had at least one literary inquisitions (the 'treatment' group in this quasi-experimental research) with prefectures that had not (or had not yet) had any literary inquisitions (the 'control' group). They look at both historical measures of social capital (e.g. the number of local charities) and more modern measures (e.g. generalised trust).

This is a very detailed paper. Their first step is to demonstrate that political repression through the literary inquisitions affected attitudes and beliefs. Looking at the number of 'reputable individuals', which is based on a compendium that listed people who were:

...well known for reasons that included prominence in science and technology, medicine and healthcare, education, classical and literary scholarship, history, art, or poetry.

Comparing matched treatment and control prefectures over the period 1640-1819, they found that:

...literary inquisitions led to fewer reputable individuals and that this decline was more pronounced among individuals who came of age in the decade of a literary inquisition... This concurs with historical accounts that individuals withdrew from public life and sought to evade attention in order to keep a low profile...

Next, they looked at the contemporaneous impact on social capital. Charities are a good measure of social capital in this context because:

...the level of charity provision reflects the degree to which individuals are willing to volunteer time and resources to help other members of society...

They found that:

...after a prefecture was first exposed to a literary inquisition case, the number of local charities in that prefecture fell by an average of 38% in the following decades, relative to prefectures that never had a literary inquisition, or prefectures that had not yet experienced a literary inquisition. Mapping out the full dynamic response of charity formation, we characterize the evolution of social capital after exposure to literary inquisitions, and show that the “charity gap” between prefectures which had and which had not been affected, kept widening for the next four decades before stabilizing. This effect did not go away towards the end of the charity data series in the early 20th century.

So, the impact of political repression on social capital was relatively large. It also endured for some time. Looking at data from the Chinese General Social Survey (CGSS), they found that:

...Qing persecutions are associated with lower generalized trust. This effect is economically significant: in our main specification, political repression is associated with a decline of 0.179 in trust, which is 16.7% of its standard deviation...

In contrast, there is no effect on trust in family members, so the impact of political repression is on trust outside of the family. Xue and Koyama also demonstrate negative effects on literacy levels among those educated in the early 20th Century, when education was decentralised and so relied more on social capital for its delivery. They also show that those effects don't depend on other political and social events such as the Taiping Rebellion, the exodus to Taiwan in 1949, and the Cultural Revolution (the latter two events might affect the earlier results because they change which people would be observed in the CGSS sample). They also provide evidence using instrumental variables analysis that their results are causal - the literary inquisitions in Qing China caused the differences in literacy rates in 20th Century China.

Finally, they also show impacts on modern political attitudes and behaviour:

Starting with attitudinal questions, we find that individuals in prefectures with a legacy of literary inquisitions are less likely to say that people like themselves can have an impact on decisions made by government... and less likely to believe that their suggestions to the government will be adopted... reflecting greater political apathy in affected prefectures... Next, we examine participatory behavior... Our main findings are that survey respondents from affected prefectures are less likely either to volunteer on local committees... or to make suggestions to local committees...

We find that individuals in prefectures with a legacy of literary inquisitions are less likely to agree with the following statements: “Western-style multi-party systems are not suitable for China”... “Free speech is ‘Western’ and will only lead to chaos”... and “Modern China needs to be guided by wisdom of Confucius”... For other questions in the survey, such as those regarding social issues, there is no systematic difference in individual responses between prefectures with a legacy of literary inquisitions and those with no such past.

There is a lot to digest in this paper. The authors conclude that:

Both the results from the historical panel and those from post-Qing cross sections suggest that political repression permanently reduced social capital. The effect of literary inquisitions on social capital has survived, even after local institutions were transformed by the modern socialist state...

We establish social capital as a missing link to understanding the dynamics of state-society relations. Autocracies can provide order and public goods when social capital is low. For this reason, autocracy may appeal to individuals in societies with low social capital. Hence, by reducing social capital, autocratic rule can introduce a self-reinforcing cycle that favors its survival and persistence.

This self-reinforcing cycle has implications for many modern societies that have tried unsuccessfully to adopt more democratic norms. It demonstrates the persistence of autocratic rule, and the norms and attitudes that sustain it.