Wednesday, February 4, 2015

Poor economics

If you care about economic and philosophical issues to do with poverty then you might be interested in this free online course. It's based on the book Poor Economics by Esther Duflo and Abjhit Banerjee. In the beginning of the book (pp. 4-5) they say that:
This book will not tell you whether aid is good or bad, but it will say whether particular instances of aid did some good or not. We cannot pronounce on the efficacy of democracy, but we do have something to say about whether democracy could be made more effective in rural Indonesia by changing the way it is organized on the ground, and so on.  
In other words, as I understand it, they reject the idea that economics or economists or anyone else can pronounce generally on whether democracy or aid or free markets or any other general solution to poverty works, but they do think that people can identify what works and what does not in particular cases. This sounds right, but it suggests a very modest role for economics as a science. (I say 'suggests' because I still have a lot to learn about all this.)

Two of the first readings required by the course are blog posts for and against the idea that Jeff Sachs' belief in giving away insecticide-treated mosquito nets (ITNs) for free has been vindicated by empirical evidence. The first says that:
new randomized experiment carried out by Jessica Cohen and Pascaline Dupas reaches striking and unambiguous results:
Taken together, our results suggest that cost-sharing ITN programs may have difficulty reaching a large fraction of the populations most vulnerable to malaria. [...] We also find that, for the full range of parameter values, the number of infant lives saved is highest when ITNs are distributed free.    
The second counters that:
Cohen and Dupas have confirmed that demand curves slope downward. More than that, they have found that there is residual demand for bed nets from mothers with new babies in a community that has benefited from extensive demand promotion activities in conjunction with a social marketing campaign for bed nets, which are very highly subsidized, but not free.
In other words, even "striking and unambiguous results" produce next to nothing that can be generalized with confidence. What is found to happen in one region in Kenya is found to happen in that region in Kenya, given the particular circumstances prevailing at the time. Things might be different even in that region at another time.

I don't mean that these findings are worthless. They are at least suggestive, and might eventually add up to more than that. But despite all the praise, my first impression (more impressions coming soon!) is that this cutting-edge work in economics really confirms a lot of the main points made by people generally regarded as anti-economics or anti-social science, especially the point that human life is too complicated and affected by too many variables for anyone to be able to identify universal laws of human behavior.

17 comments:

  1. Have you read A. R. Louch's Explanation and Human Action (published in 1966)? It deals with questions rather like this.

    ReplyDelete
    Replies
    1. That book may be out of print now though there seems to have been at least one subsequent edition. I first read it forty years ago and only recently rediscovered it on my book shelf and started reading it again. He relies heavily on Wittgenstein's insights re: mind and agency and how we talk about these things. Here are a few links I stumbled on yesterday addressing the book:

      https://books.google.com/books?id=Sya7zOeIZHEC&pg=PA275&lpg=PA275&dq=explanation+and+human+action+a.+r.+louch&source=bl&ots=IfTRdZSTHr&sig=MrU9CnBg1W63D3TPj9_CR8cpq7g&hl=en&sa=X&ei=kKbSVP2bIcGlNpWFgdAB&ved=0CDcQ6AEwBw#v=onepage&q=explanation%20and%20human%20action%20a.%20r.%20louch&f=false

      http://link.springer.com/article/10.1007/BF00368032#page-2

      https://books.google.com/books?id=IYcQjHXz7fwC&pg=PA241&lpg=PA241&dq=explanation+and+human+action+a.+r.+louch&source=bl&ots=QOKV6i0Hus&sig=4UyPv9e4SdLI643PTRDN0ACCFK0&hl=en&sa=X&ei=kKbSVP2bIcGlNpWFgdAB&ved=0CDoQ6AEwCA#v=onepage&q=explanation%20and%20human%20action%20a.%20r.%20louch&f=false

      Delete
    2. would second the Louch rec.
      " human life is too complicated and affected by too many variables for anyone to be able to identify universal laws of human behavior."
      I've long been in this camp but wonder what is left than to philosophy as relates to ethics?
      -dmf

      Delete
    3. Thanks for the links!

      Yes, ethics would have to be something else. Exactly what is something I'm still trying to figure out.

      Delete
  2. Louch argues that the social sciences are moral investigations rather than the sort of enterprise that hangs on formulating and testing general laws. But he distinguishes the social sciences as moral assessment from moralizing which he describes as cases of social scientists critiquing other cultures they're studying in light of their own mores and ethical standards.

    He does not, of course, indulge in metaethics, an inquiry that looks to me to be a uniquely philosophical endeavor (being the inquiry into how our moral judgments may -- or may not be -- justified). To the extent that metaethics touches on normative ethics (for what is normative necessarily stands on whatever rationale ethics can claim as its own) there is room for philosophical inquiry, too.

    But as to the making of normative judgments, it seems to me that that is not a philosophical matter at all for if it were then only philosophers (and perhaps only very high level philosophers at that) could ever hope to be genuinely ethical (as in getting normative judgments right). And that would make ethics pointless for the rest of us. So any moral account at a metaethical level must somehow show how we can and do make at least some reliable normative judgments and how and why they can be shown to work.

    ReplyDelete
    Replies
    1. I'm not sure that what you describe in your last sentence can be done, but the rest all sounds true. Not that I necessarily agree with Louch, but I think I need to read that book. Thanks.

      Delete
    2. I think the most interesting stuff in Explanation and Human Action is Louch's handling of the mental. He departs from a quasi behaviorist interpretation of Wittgenstein to make the point that to operate on a moral level (i.e., to make use of reason-based explanations for action) one has to presume a mental life (my words here, not his, since I came to the same conclusion -- albeit forty years later). He also seems to have predated Dennett's account of intentions (as being a function of our taking an "intentional stance" towards some entities) with his emphasis on, and clarification of, the distinction between causal and purposive explanations.

      Where Dennett speaks of three stances (the mechanical, the design and the intentional), Louch invokes two (causal and purposive) and makes the interesting point that in some important sense the purposive is more primitive than the causal with us. That is, we learn the purposive uses in our explanations as children before learning the causal. I'm not sure I agree with that (and it seems to run contra to Dennett's hierarchy) but it caught my attention and is certainly worth thinking further about.

      Delete
    3. There's probably a Wittgensteinian origin of all these ideas, I would think. Not that Wittgenstein himself would necessarily agree with them, but Dennett acknowledges a debt to Wittgenstein and I imagine Louch would/might too.

      Delete
    4. Yes, absolutely. Louch is avowedly indebted to Wittgenstein and Dennett certainly credits Wittgenstein. Both also owe something to Ryle though Louch is somewhat critical of him, thinking that he lacked Wittgenstein's subtlety. Louch explicitly argues that the quasi behaviorist interpretation of Wittgenstein is mistaken (in the middle of the book, I believe, where he deals with intentions, purpose and consciousness).

      Delete
  3. This is interesting. In this context, I can say of you what Wittgenstein said of Cantor: a clever man got caught in this net, so it must be an interesting net!

    In other words, as I understand it, they reject the idea that economics or economists or anyone else can pronounce generally on whether democracy or aid or free markets or any other general solution to poverty works, but they do think that people can identify what works and what does not in particular cases. This sounds right, but it suggests a very modest role for economics as a science.

    It's not just "they" who reject this idea, but contemporary economics on the whole. I haven't read this book, but it seems to be a completely innocuous mainstream book of mainstream economics, done in its mainstream self-image.

    (Cf.: "He says that philosophy leaves everything as it is, and that what philosophy amounts to is a battle against the bewitchment of our intelligence by means of language. This sounds right, but it suggests a very modest role for philosophy as a science.")

    this cutting-edge work in economics really confirms a lot of the main points made by people generally regarded as anti-economics or anti-social science, especially the point that human life is too complicated and affected by too many variables for anyone to be able to identify universal laws of human behavior.

    Contrary to what these "anti-economics people" have managed to convince themselves of, they were simply never promised any "universal laws of human behaviour" in the first place.

    First: Economics is divided into macroeconomics and microeconomics. Of these two, only macroeconomics studies even so large an entity as some particular economy as a whole. It does not make any pretence of studying humankind as a whole, much less its "behaviour".

    Second: There is such a thing as behavioral economics, but it's a small subset of economics (and an intersection of macroeconomics and microeconomics).

    Somebody who thinks that economics is about "universal laws of human behaviour" seems to think that: 1) macroeconomics is the only kind of economics there is; 2) behavioral macroeconomics is the only kind of macroeconomics there is; 3) the behavioral science that behavioral macroeconomics makes use of is embedded in some kind of stereotypical nineteenth-century positivism, according to which science is done to discover "universal laws" of some sort.

    1) is completely false. 2) is simply not true. 3) just isn't true.

    Here, as when I earlier assessed Winch's 1950s criticisms of sociology, it helps to look at a fresh issue of a leading journal in the field. This month's issue of the American Economic Review has articles on such topics as: the mortality of African Americans who migrated away from the rural South; the effect of toxic plant openings and closings on house prices nearby; the treatment of malaria; the relationship of work experience to wages and how it has changed over time.

    Is dying a behaviour? Or if you were a public health official dealing with malaria treatment, wouldn't you be more interested in a paper with "malaria treatment" in the title than one with "universal laws of human behaviour"?

    Inasmuch as economics is a science, it's an empirical science, just like medicine, or law, or engineering. And what these sciences examine is of course some particular thing or another. Ironically, for economists this goes without saying. They can say it if you like; it's just that they never imagined someone wanting it said separately.

    ReplyDelete
    Replies
    1. Thanks, Tommi. I haven't read your next comment yet, but here's a quick response to a few of your points here:

      1. I'm no Cantor and quite capable of stupid mistakes
      2. I really am only recording an initial impression, nothing more
      3. the book is presented not as innocuous or mainstream but as cutting-edge, radical, new, etc. Maybe that is all meaningless hype, but that presentation helps create the initial impression I had
      4. I am almost certainly too influenced in my thinking about economics by a friend of mine who is an economist and propagandizes constantly on its behalf, sometimes making claims to the effect that it is the science of human behavior. He, at least, seems to think that economics can and does show what the right response to an economic depression is, whether minimum wages work, and so on. In other words, that it provides large-scale solutions to large-scale problems and that these solutions are essentially the same for all places at all times. The odd thing is that he admires Duflo and Banerjee even though, as I see it (again, having nothing but a first impression to report), they repudiate many of the kinds of things he wants to claim for economics. Perhaps I am misunderstanding him.

      Delete
  4. That said, the mistake is by no means a stupid mistake. There are some individual schools and research programmes in economics, such as the Chicago school, public choice theory, or law and economics, which have – especially at their worst – made many embarrassingly extravagant claims of roughly the sort that the "anti-economics people" unfairly associate with economics as a whole.

    This is sometimes known as economics imperialism, and it's significant that the phenomenon is so widely remarked upon as to have a Wikipedia entry to itself. But it's not the whole of economics, any more than (say) logical positivism, deconstruction, or Thomism is the whole of philosophy, or even the whole of Western philosophy.

    It's just that these schools I mentioned have long had influential backers in politics and the media, which have enabled them to punch above their weight and give the impression that they are more paramount in economics an an academic discipline than they have ever actually been.

    I myself know some academic economists personally; have written repeatedly on economics, in both my books and my newspaper column; and have received actual fan letters from economists I don't know, telling me how gratified they feel that a layperson like myself writes on economics without conflating it with economics imperialism, for a refreshing and totally overdue change.

    They feel as offended and bewildered by the conflation as you would if you introduced yourself as a philosopher and found yourself being held to account for some one particular piece of philosophy which you don't just disown, but which you feel so alien to as not to see the need separately to even disown it – Aristotle's defence of slavery, say – merely on the grounds that it's philosophy. It is my experience that the statistically average academic economist is more likely to want to dissociate economics from economics imperialism than to associate it with it.

    (Even one of the papers in the February issue of the AER, "Overconfidence in Political Behavior", examines the very phenomenon of social psychology that leads to economics imperialism, among other things – and not because the authors are thrilled about the phenomenon!)

    ReplyDelete
    Replies
    1. Thanks again. I know what you mean about the kind of conflation you describe. Someone once gave me an article about Heidegger's Nazism as if to say that this proved the depravity of philosophy.

      By the way, I read and very much enjoyed Ha-Joon Chang's book. I'm trying to overcome my ignorance. It's a slow process.

      Delete
  5. https://varoufakis.files.wordpress.com/2013/10/yv-on-bbcradio-4-seismology-vs-economics-cut-version.mp3

    ReplyDelete
    Replies
    1. Thanks! Presumably he doesn't think economics is a complete waste of time. Figuring out what it can and should be (how it should be done and to what end) seems like an interesting question.

      Delete