Chicken and egg economics
After my recent post, which dealt with issues of equity, I have been invited to consider how appropriate it is to bring normative issues in to a discussion of economics. I was referred to a paper which discusses the distinction between economics as a science and political economy:
For Robbins, the science of economics and the entire field of economics were quite different… He saw [the] pure science of economics as only a small sub branch of economics — a branch, which in his view, had nothing to do with policy. He saw another branch of what economists do — political economy, as the branch primarily concerned with applied policy, not with science. Here, he wanted value judgments to have free reign, and to be allowed into the analysis.
I like the distinction, but I feel that the normative judgments should inform our science, not just the other way around.
If the aim of economics is to study the allocation of scarce resources then surely there is little point in studying allocative mechanisms which are not practical or feasible. For an economist to study the science of economics without considering the constraints of what is politically possible is to render his science irrelevant before the ink has dried. Is the science of economics to be a tool for signalling intelligence, or do we actually seek to improve the world by doing it? If it is the latter then developing the science without first considering what tools the political economists need is pointless.
Academic economists should be mindful of the normative judgments that political economists make when they choose their research topics. If the tools that they develop do not aid in making these normative policy judgments then they are not helping the cause of economics, only the academic’s own career. While a separation between the science and political economy is important, one must not make the mistake of thinking that the science can be effectively done without reference to the normative judgments that are being made.
You and Matt both seem to have much bigger hang-ups about the objective/normative split than I do. I think it is fairly evident that there is no solution to evaluating the normative value of policies: we can’t know utility curves, or how those utility curves interact, and we probably can’t even agree on the maximisation criteria to be applied. So no matter how great any policy appears in theory, we can almost never rule out the possibility that it makes society “worse off”. Some areas of economics (happiness research, neuroeconomics) may make inroads to these issues, but substantial resolution seems unlikely/a long way away.
But I don’t see this as a huge problem. I think plenty of policy is so poor, and deviates so substantially from economic principles, that there is plenty of scope to recommend policies that are likely to improve welfare in all but the most bizarre sets of utility schedules. I also think that a lot of policy stems from beliefs that are either demonstratably wrong, or demonstratably founded on false principles. Economics can go some way to correcting these things.
“For an economist to study the science of economics without considering the constraints of what is politically possible is to render his science irrelevant”.
I’m unconvinced that this is much of a constraint. NZ has gone from strongly socialist to a paragon of free-market economics (and some of the way back again) in under 30 years. Marx and Friedman both espoused radical economic philosophies long before either was put into place. Throw out the assumption that we’ll remain in a democracy, and the constraints seem even less binding.
But anyway, public choice economics seems to deal quite substantially with what kind of results the political process is apt to produce.
I do like economics that seeks to test theoretical results empirically (although there are plenty of important results to come out of pure theory). But economics is hardly lacking in unresolved problems, so I think there is plenty of room for economists to derive results outside of a normative framework. Just thinking of macroeconomics, most of the policy framework seems to be consensus these days, but there is plenty of scope for research that improves our forecasts or our knowledge of macroeconomic interactions.
None of this abstracts from the fact that our normative views color our research agendas (and thus potentially our research results). Economics always faces the challenge of proving that results were derived in an objective framework rather than a normative one. Unfortunately that kind of criticism normally comes from sources poorly placed to assess the objectivity of the assumptions that underly economic modeling.
To call you on your belief in the importance of normative judgments, what specific steps could economics/economists take to improve the situation?
“I have been invited to consider how appropriate it is to bring normative issues in to a discussion of economics”
Huh, I just sent you the paper because I thought it was interesting 😉
“but I feel that the normative judgments should inform our science, not just the other way around”
None of this rules out that economists choose what to study based on what issues are important on the day. The purpose of the positive – normative split is to define what part of our analysis is descriptive, and what part is value-laden.
Robin believed that ‘economic science’ should be solely positivist, as he was one of the positivists around during the positivist revolution in the sciences. However, he realised that we have to include value judgments before we can provide policy advise in most cases.
The distinction is useful insofar as its allows economists to agree on a set of fundamental ‘facts’ before applying value judgments to make policy – in other words it is useful because the value judgments we make are transparent.
No-one thinks it is possible to analyse a problem without value judgments – hell even physics has given this up. However, the distinction is useful insofar as it guides our method and provides transparency.
Perhaps I can answer both your queries at once by describing the motivations for my rant/ramble. Academic economists tend to lean towards practicing economics science rather than political economy, and this is what tends to get published in journals. How many of those journal articles are useful to the political economist or policy-maker, and how many are obscure, abstract results that serve little purpose other than to make the author look smart? I contend that many journal articles fall into the latter category. The authors do not consider what the purpose of publishing the article is: publishing is an end in itself (and not unreasonably so, given the incentives they face in most universities).
As I am repeatedly told, there are three types of articles: there are theoretical articles which advance our theoretical understanding of a topic; there are practical articles which are directly useful to the political economist; and, there are theoretically practical articles, which apply known theory to problems which are of no practical use to anyone. The last category is useless, but very popular among economists because it is by far the easiest sort of paper to write.
I didn’t have a query, I was just stating what I think the distinction between positivist and normative is in economics, and why it is a useful one – I think the question of ‘what to study’ is separate from the question of ‘how to study it’. The positive – normative split Robbins discusses is in the how category, not the what category.
Ultimately I understand that the reward structure in academia may not get the optimal set of ‘economic knowledge’ out, but I don’t really know much about that so I wouldn’t want to guess either way. Your argument for it sounds convincing though, so I’d be apt to agree with you in the light of no more information.
Also I was stating that I didn’t send you the article in any sign of attack (in fact I enjoyed your discussion in the post about hotels and the disabled), I just thought you might enjoy it 😉