Dani Rodrik: Give China a break

Congress and Wall Street and Main Street are for once in agreement: enough of this market manipulation, China needs to let its currency appreciate and balance its trade; US manufacturing is bleeding on the operating table and China is hoarding the blood supply.

In an AEA talk today, Dani Rodrik tells us to give China a break. His argument: the ‘Depreciate Now!’ view downplays the extent to which China’s currency and trade policy has driven its astonishing growth and social stability.

It also downplays the fact that China’s imbalance is also quite a recent phenomenon – since just 2001—the year that China joined the WTO. (WTO membership meant abandoning many of its traditional tools of industrial policy. So the currency strategy is, at minimum, a means of managing the transition.)

Even if currency manipulation is no temporary move, maybe we should simply let China be. For a bit. China has more poor people than any country but India, and Chinese trade and industrialization has done more to reduce world poverty than any action in the history of humankind. Letting that machine run for a few more years may be the greatest kindness of strangers in human history as well.

I could also see a fairness argument to be made. The West (if I remember my economic history) enjoyed a long period of trade surpluses that helped speed its own growth. Is it time to give other nations a chance?

Dani makes a slightly different point, however: if we take the long view, a growing China with a labor surplus employed is a safer China (and, by extension, a better safer world). Our short run pain is everyone’s long term gain.

I tend to agree, but with one reservation. Dani’s idea seems founded on another conventional wisdom: the idea that the risk of social unrest rises with the number poor and unemployed young men. Is that true?

It’s surprising how entrenched is this view, in spite of an absence of good evidence, especially at the micro level. The little evidence in political science and some economics suggests the instigators, terrorists and guerillas of the world seem to be drawn from elites, students, and middle classes at least as often as they come from the poorest and unemployed.

Apologies to Dani if I misrepresent his view–I blogged after hearing the 10-minute talk rather than  reading the full paper. When I’m mistaken, though, my readers are very good at pointing it out. Fire away…

6 thoughts on “Dani Rodrik: Give China a break

  1. I’d be a little less forceful with your “there isn’t enough micro evidence” claim about the connection between poverty and violence. Certainly, it’s a complex issue, with contervailing effects on the outcome, but there’s general support, in my view, for the fact that the unemployed (and disenfranchised) are more easily recruited than the rich by instigators of violence. Whether the instigators themselves are rich or not is immaterial.

    You can see this in Scacco’s “Who Riots?” paper, where the interaction of economic grievances and social networks leads to participation in riots. Also in Dube/Vargas, where falls in coffee (and other agricultural commodity) prices–a lowering of the opportunity cost of violence–increase incidence of guerrilla attacks. That’s pretty good micro evidence.

  2. Poverty as one of the causes of civil war – isn’t that one of the points that Fearon and Laitin make in their paper in American Pol Science Review?
    It seems to make sense, since the ‘leaders’/’instigators’ of rival factions need loyal and committed armies, without which, the chances of success would be low, meaning lower incentives to challenge the state in the first place.

  3. Tristan and Suvojit,

    Though there are several papers showing that income affects negativelly the probability of civil war in some places (just to cite some of them…Fearon and Laitin’s paper on American Pol. Sci Review; Miguel, Satyanath and Sergenti on the JPE), there are several others showing evidence right the other way, showing how unemployment, education and income might motivate participation in rebels/terrorist movements (Krueger and Maleckova on the Journal of Economic Perspectives; Abadie (2006) on the AER P&P; Benmelech and Berrebi on the Journal of Economic Perspectives, recently I’ve seen some empirical justification for the thesis that, with low unemployment, it becomes harder for the government to buy information from people who are sympathetic to rebels…unfortunately, I don’t remember the reference). So, just as professor Blattman was saying: it is ambiguous whether income correlates positively or negatively with conflict. Regarding Dube and Vargas paper, I’d argue that their exercise raises a question: is it income or distribution of income which may generate conflict? Miguel et al. might be read as the impact of shocks on agricultural income on conflict, Dube and Vargas might be read as the shocks of some commodity sectors income on conflict, both leave open the question: is this shock decreasing the opportunity cost of conflict, or generating a higher political conflict of interest between classes/sectors?

    Changing topics…I’d only “disagree” (I still don’t have a really strong opinion on that) with the post regarding the exchange rate in China having the ability to drive people out of poverty. In my experience as a Brazillian, when we used to do exchange rate policy, the result was rising prices, not necessarily higher exports/lower imports, decreases in competitivity and, for the lack of generalized access to financial markets by that time, increased prices translated into higher real income inequality. Surely, that was not a development policy in Brazil, and mostly, our exchange rate is now appreciated (and has been getting more and more appreciated for the last 6 years), and we haven’t seen any drops in the share of our industrial production on our GDP. Anyway…maybe Brazil and China are two different countries and I can’t take Brazil’s lessons to China. But, for my origins, I’m quite skeptical regarding this thesis that we need to use currency depreciation as a development policy. Sorry for the long message….

  4. Michel summarizes many of my feelings very well. The evidence is very thin and contradictory, and people publish and cite selectively. Economists have a tendency to take one or two good empirical results and generalize quickly. My worry is that studies that see no relationship between unemployment and violence do not get published. We only see papers with coefficients statistically significantly different than zero. When we do get a significant result, there is seldom replication. Finally, qualitative evidence to the contrary (of which there is loads) gets ignored because it is not ‘rigorous’.

  5. I’ve been working on a chapter on governance for a food policy textbook where I discuss this problem in some detail. Let me paste a few scattered portions from the section on conflict and see what your reactions are.

    P. Collier et al. (2003) declare that low and unequally distributed income per capita income is “the key root cause of conflict.”
    A 2003 UN Security Council report identified two main sources of the current conflicts in the central African states: lack of good governance and poverty. The World Bank (2004) similarly indicates that poor countries are three times as likely to experience civil war as middle income countries. As seen in Table 3, a comparison of 67 countries that experienced conflict between 1980 and 2005 with 44 developing countries that did not reveals that the countries in conflict had, in the period before conflict began, on average half the GNI per capita and negative economic growth, twice the poverty, less education, greater food insecurity, greater inequality between men and women, less access to improved water sources, and twice the under-five mortality rate (Pinstrup-Andersen, 2006; also Pinstrup-Andersen and Shimokawa, 2008).
    In Case 9-7, Avalos-Sartorio demonstrates that pockets of poverty within middle income countries are significant as well because “although poverty itself does not cause violent conflict, it does provide a basis for it.”
    None of this is meant to imply that impoverished, marginalized, or hungry people are terrorists. The cross-country research and case studies demonstrate that exclusion, poverty and related grievances provide a “moral and political foundation” for those who propagate violence (Pinstrup-Andersen and Shimokawa, 2008). Miguel (2008) argues that the desperation of poverty makes violence or crime more attractive. Reducing poverty, hunger, and inequality between groups of people can significantly reduce the probability of conflict and terrorism (Pinstrup-Andersen, 2003).

  6. I am intrigued by this statement: “The West (if I remember my economic history) enjoyed a long period of trade surpluses that helped speed its own growth.”
    This would mean that the West exported capital to its colonies in the colonial period.
    The economic history I remember (at least as it was taught to us in India) is that Britain drained capital from India and naturally had a trade deficit (I have never seen current account data so assume that the trade deficit is good enought in the basic accounting identity). But perhaps I am wrong.