by liberal japonicus
More China stuff. As you’ve probably noticed, I’m skeptical about narratives that argue that China is collapsing because [insert reason here]. One of those reasons has often been that China is corrupt from stem to stern. A few links to get an idea of the arguments, here, here and here
There was something about these arguments that seemed to be off to me. So this interview with Dr. Yuen Yuen Ang was something that I thought was quite interesting. Ang points out a couple of interesting things, which include 1)how corruption is not neccessarily a drag on economic growth, 2)the judgementalism of Americans and 3)how overall our ideas of corruption are often constrained by particular viewpoints. Ang says:
…it’s much easier to condemn corruption among the poor. Very difficult to talk about influence politics among the rich. It’s a topic that people do not generally like to touch upon.
Here is a review of Ang’s book and this China Focus page gives a short print version of Dr Ang’s thesis. The page is interesting because it suggests that anti-corruption campaigns have not worked because there has been little change in global indices of corruption, but Ang’s whole thesis argues that indices that try to summarize corruption as a single number so one can rank countries as more or less corrupt misses a lot of nuance, suggesting that an ‘unbundled corruption index’ or UCI would be better at understanding the reality of the situation.
Another interesting point that Ang gets into at the end was how her research and arguments have not been accepted by many economists and social scientists.
Stephen Dubner: So, in the current book, you critique some of the literature on corruption. In your first book, you critiqued the poverty reduction analysis of quite esteemed economists like Jeff Sachs and Daron Acemoglu. So why should we be more persuaded by your analysis of China than theirs? Is there something fundamental that they are missing because they don’t understand China the way you do?
Ang: In one professional letter describing my work, it was said that this person has the nerves to challenge luminaries in the field. I think it’s meant to be a compliment, but I didn’t quite see it that way. I think that is a statement about the structural inequality in the profession because if we lived in a world where every academic is truly equal, then it doesn’t matter if I’m challenging Sachs or Acemoglu or any person in particular, it’s just about the findings. I would hope that in an ideal world readers would just look at the argument itself. And my critique is that many social scientists reduce the process of development into a mechanical outcome. Everyone wants to give kind of a short secret recipe that is like either one thing or the other. And I wanted to tell a different story that does not dumb down the reality. Conventional social science has a fundamental assumption which is that you can take development and break it down into discrete variables and you can apply an intervention and get a predictable outcome. That’s a very core assumption. And it’s an assumption about the nature of things. And it is so fundamental that nobody talks about it. It’s like assuming that water is wet. And so what I did in my work and it is the philosophical foundation for all of my work is that I reject this paradigm. I reject this mechanical worldview because it’s artificial. That’s not how social realities function. Social realities are not like machines. They are more like forest ecosystems. They are multi-dimensional, constantly changing, adapting to one another. So, we need to have a different set of methodological tools. You know, people were so angry at the unbundled corruption index. […] I can’t get the unbundled corruption index published as a journal article. It could appear in the book because a book is peer-reviewed as a whole and not in parts. But the reviewers were absolutely livid about the unbundled corruption index. And we know that reviewers are critical. So that’s very normal. But they were more than critical. They were personally angry and they tried to throw out every reason thinkable to block it. So when I see that I knew that oh I’m doing something that you know impinges on something personal to them and perhaps they have been using these conventional measures perhaps they have made arguments on the basis of these measures and of course they do not want this to be challenged.
Given that Sachs was the author of the so-called ‘Shock Therapy’ approach that was foisted on Russia after the breakup of the USSR, I don’t know why anyone thinks any other analysis by him would be worth the paper it was printed on, but that might just be me. Ang concludes with an example that I wish every academic would consider
The analogy I would use is have you heard of the term machine friendly crops […] there are certain crops that are easily harvested by machine and so farmers would choose these crops simply because they can be easily mechanized. And I think that in the knowledge industry we sometimes or maybe often see a similar dynamic and I would call it a publication friendly agenda. And so the incentives of the profession will lead people to overwhelmingly and disproportionately study certain kinds of topics in certain ways at the expense of truly important questions that frankly very few people want to touch.
My biggest complaint about the assessment culture that has set in across academia (and the overall rise of big quant that coincides with the monetization of big data) is that there are a lot more people with the tools to gather and measure the data than there are people who have the understanding, expertise, and rigor to tease out when the things we measure actually measure the right things.
I’m interested in reading Limits of the Numerical one of these days: https://press.uchicago.edu/ucp/books/book/chicago/L/bo146791774.html
It seems the sort of book that can take on the technocratic push for quantitative over qualitative data gathering and analysis. I find that the voices that most often get amplified in management meetings dealing in quantitative assessment are the voices that are on the wrong side of Einstein’s admonition that everything should be made as simple as possible, but not simpler.
I’d say that whether something is usefully measurable depends enormously on the topic.
For engineering it’s closer to critic — “if you can’t measure it, you can’t manage it.” For the physical (including biological) sciences it’s important when testing out new theories. But useless for coming up with those theories. For the social sciences, it ought to be important, again for testing theories (but again not useful creating them.) But currently, so much of it is poorly done that it isn’t. At least not yet.
For the humanities, I’d say it’s totally useless. Doesn’t keep fools from trying to do it anyway. But it doesn’t work because it can’t work.
Yes, on that last paragraph! In academia, but also in business, where we get a “measurement-friendly agenda”. To me it feels particularly pernicious when work and rewards are steered to things that can be most easily justified by measurement, without much work to tie those measurements to actual good outcomes.
See also Goodhart’s Law, which states that “when a measure becomes a target, it ceases to be a good measure”, or McNamara on body count. Although it’s pretty common to lament that since publication count became a target it’s become a poor measure, I don’t know that I’ve seen any effective action to work around it – although I never experienced the British system that limited academics to something like reporting only your five “best” papers, regardless of how many you’d published?
[…] Obsidian Wings** – China’s Gilded Age (now with extra corruption!) […]
I’d be interested to see the corruption index. My sense is that Russia, for example, is at a whole different level from China.
I think it’s possible to have a growing economy dispite widespread corruption. But much harder than without that corruption. Also, I’d argue that there is a point where corruption gets so bad that economic growth becomes impossible. Clearly China isn’t there. But I’m not sure how close they might be to the limit.