Big Ideas
Aine Seitz McCarthy
  • Blog
  • About

Why randomized control trials are good for economics

10/21/2013

2 Comments

 
Because apparently many people don't believe economics is a science. Raj Chetty defends the field in the face of disagreement here in the NYT. First of all, economics is a social science. Secondly, as with most scientific research, testing theory through hypotheses and empirical analysis makes the field scientifically and policy relevant. Thus, the macroeconomists are maybe a little jealous of the development economists' RCTs:

As is the case with epidemiologists, the fundamental challenge faced by economists — and a root cause of many disagreements in the field — is our limited ability to run experiments. If we could randomize policy decisions and then observe what happens to the economy and people’s lives, we would be able to get a precise understanding of how the economy works and how to improve policy. But the practical and ethical costs of such experiments preclude this sort of approach. (Surely we don’t want to create more financial crises just to understand how they work.)

Hat tip: COCO (a real scientist)
2 Comments
David Wiczer link
10/21/2013 06:14:41 am

It's weird to define stuff as "science" just like it's weird to define stuff as "art." Is it pretty? Are we quantitative? One can create something that's a "yes" and is it then necessarily art or science? Anyway, that's really philosophical.... but my basic point is that words are a bit silly and we should leave stuff pretty open.
OK, other side: I'm going to be a bit nasty. I think that one of the crucial things econ lets us do is exclude theorizing as silly and not serious. I'll take Chetty's opening point that Shiller & Fama have "opposing" "views," would seem problematic for a "science." (too many scare-quotes there). But they're both economists that do "serious" work, in that they use "serious" methods, even if some of the conclusions are inconsistent. For each of them there's a thousand non-serious assholes who just make up opinions and state them as fact. Financial commentary is rife with it.
So now I've got to confront what "serious" methods are. The problem is that they're not super "serious" in the scientific sense. Macro doesn't use the scientific method, with control groups and non-parametrically identified, testable hypotheses, like real scientists (yeah Coco) or people with cool experiments would do. Instead we work in the world of Hanson (see! I worked in all of this year's Laureates!) where we get little theories, those imply a finite number of moments of the distribution of data (instead of the infinite moments that potentially exist) and then we test those finite number of moments. When I get super techy I twist the data in ways that let me test implicit functions of moments, but that's just letting me pose more complicated theories to the data, rather than be more certain of a particular hypothesis. Anyway, I think it's useful work for us to do, because there are complicated theories out there that should be confronted in some way with data, but it's not "answering" whether they're true or not.
That bugged me a lot at first, but then I learned to stop worrying and love the bomb ;)

Reply
Joseph Lemien link
11/11/2013 06:03:15 am

I'm happy to see economic RCT getting more press. I think that it is one of the major things that can gain more respect for a field, such as how use of the scientific method differentiated psychology from philosophy due to attempts to measure and quantify.

Thanks for sharing.

Reply



Leave a Reply.

    Aine Seitz McCarthy

    International development, economics and some pretty ambitious ideas from a stubborn graduate student clinging to her sense of adventure.


    Categories

    All
    Agriculture
    Amusements
    Books
    Camp
    Demography
    Economics
    Education
    Family Planning
    Fieldwork
    Futball
    Gender
    Grad School
    International Development
    News
    Public Policy
    Research
    Travel

    RSS Feed

    Enter your email address:

    Delivered by FeedBurner


Powered by Create your own unique website with customizable templates.