Industrial Organization.
International Trade
Development
Labor economics
Health economics
Sports economics
Economic history

Non macro. Non finance. All economics.
Though I do share your distaste for a lot of “Applied Micro”, it’s really more because of method (let’s do some data mining then slap together some sloppy half baked theory that has the word “demand” in it somewhere, that wouldn’t last ten seconds if it had to be spelled out rigorously) rather than the subject matter – anything that involves choice or accounting is fair game for economics.

This is assuredly not what we were promised the internet would be like.

I would stand by a set of far weaker claims that I made in my original post:

1.) RCT tended to be invoked or referred to, and even used, by those who opposed government intervention in markets – it was a useful device for making deregulation seem intellectually respectable. It may well have been misused in particular instances, but nevertheless it was used in this way – and there is an interesting wider story to be told here about the shift from embedded liberalism in the 40s to neoliberalism after 1980 as the intellectual hegemony in much of the political and economic establishment – and the role that academics, including academic economists, played in this shift. I would be delighted to have any suggestions for further inquiry.

(I will certainly read Tom Slee’s book, thank you for the reference)

I can’t claim to have anything like decisive evidence for this claim, but I will look into it in more detail later this year.

2.) There is something very fishy about RCT. Very fishy indeed.

No doubt micro analysis can be turned to particular applications in which some element of a utility function is not a widget, but instead some other external good. I suppose I tend to think that anything that’s both of widespread importance and distinctively economic is going to be macro, and that RCT when used as (at least notional) microfoundation tends to jettison the compexities of the added realism that some of the micro literature deals with. On a vague-ish note, I’m thinking of how GDP is calculated, for example- though that’s more to do with materialism than with non-tuism.

But it may be that we don’t disagree that much after all – I’m probably guilty of a bit of overstatement, and you’re right that I’m not well-versed in the macro literature. But at least my criticisms of economics are a cause (I abandoned study of the subject in favour of straight philosophy) rather than just an effect of my comparative ignorance.

And I’m sympathetic to the point about needing to try even if imperfectly. It’s just that taken too far (and I think that it probably often is somewhat, as you suggest), that route eventually leads to pure superstition.

None of the claims, that is, not buried at the end of an overlong screed like #21: Arrow – whose ‘impossibility’ theorem [is] more of a ‘barely possible incompatibility of plausible sounding assumptions’ finding.

Arrow is a cheerful and unapologetic social democrat
gol (grins out loud – the ‘out loud’ is just acro-filler as in ‘lol’). Very wise. If you’re going to be unapologetic you do well to be cheerful about it.

With a cheerful enough demeanour and a bit of bustle, you can even say a loud and hearty ‘Sorry!’* while making a blithe exit, thus fulfilling a formal requirement while remaining clearly and disarmingly unapologetic. That is how I shall henceforth imagine our Ken. Perhaps surrounded by gaping, goggling neoliberals, ‘a la H E Bateman.

[*preferably with the final vowel an ‘eh’ in patrician style, and with intonation and timing roughly following the 1st 2 notes of Colonel Bogey/The Mono-orchid Dictator Song]

For just such an argument read Tom Slee’s wonderful “No-one Makes You Shop at Walmart”

Alex Abella again:

“Arrow’s theorem proved that whenever two or more people are asked to rank their pick of three choices, unless there is a dictator among them who forces his view to prevail, there is no free choice for no one will agree on the same results.”
http://www.huffingtonpost.com/alex-abella/rand-and-the-bush-tax-reb_b_101391.html

2.

Rational choice theory ‘proved’ that collective decisions were by definition dictatorial – it follows that they are unjust. No?

3.

The claim that RCT undermined attempts to regulate is an empirical one. I will go away and research the relationship between RCT and deregulation – to see where and how it was used to promote deregulation of financial markets in particular. It would be interesting if you could find some real world examples of it being used to promote regulation, whether in conjunction with assumptions about ignorance and externalities or on its own.

4.

“So your claim that RCT undermined any practical attempt to subordinate markets to anything at all wasn’t a claim about the real world.”

I don’t know why you think this follows from what I wrote.

RCT undermined attempts to subordinate markets to anything at all (a claim about the implications of RCT and their impact on policy-making – a claim about RCT’s inadequacies and their consequences for policy-making).

Markets need governments, and market actors love subsidies. Again, a claim about the real world.

Why do these two claims contradict one another?

Now you can prove me wrong if you can find any examples of RCT being used to assert the sovereignty of anything at all over markets. (being used to formulate arguments in favour of regulation, for example).

I believe that RCT was only used in support of moves to deregulate, and that it was used only to undermine regulation as a legitimate function of government. These are very strong claims and I will qualify them if you can show me an example of RCT being used to support regulation in particular instances, or of RCT being used to promote the legitimacy of regulation as a state function.

5.

RCT begins from the assumption that individuals order their preferences and seek outcomes that maximise their utility by satisfying as many of their preferences as possible. Now this starting assumption is wrong and wrong in highly consequential ways. As a device for analysing market behaviour it might well have some uses, but as a description of human agency it is terrible – it rules out the possibility that humans can freely and rationally decide on a common course of action in the light of collectively agreed principles when individuals have conflicting preferences.

The rules of chess are useful when you are predicting what move your opponent will make, they are less good at predicting what humans will do in other circumstances.

“I also want to know what evidence, if any, could convince you that RCT didn’t provide useful weight to attempts to deregulate or to fend off regulation.”

If you could show RCT being used to promote regulate and fend off deregulation, then I would certainly revise my views.

Arrow claimed to have established mathematically that attempts to establish public goods collectively (to pursue a collective utility) could only be resolved by force or fraud –

To quote Alex Abella:

At its most basic, Arrow’s work demonstrated in formal terms – that is, in mathematical expression – that collective rational group decisions are logically impossible.

Kenneth Arrow showed that group decisions made by voting will display some irrational elements, even if all the voters are rational. If your voting system tries to reflect the views of everyone, then some examples of irrationality include things like a voter might increase the preference they have for good x, and yet the voting process will thereby result in a lesser weight on good x, or say an alternative is added which no voter prefers to the existing goods, and yet the voting system will produce a different result just because of the irrelevant alternative. (Different voting systems can produce different irrational effects). However if everyone agrees to abide by the results of the voting, at the price of some irrationality, and on the reasonably meta-rational basis that dictatorships have their own massive problems, there would be no need for force or fraud. (And, even if voting systems produced perfectly rational results, if some voters refused to abide by said results, then there is still an argument for the use of force – I think that most people would agree that would-be rapists can be justly restrained by force).

2. I am not sure what point you are making about public goods – I argued that it “followed from rational choice theory that only markets can discover and deliver public goods justly”. If collective decision-making is irrational by definition – as RCT claims – then only the market is left to deliver anything at all in a just way.

I don’t recall RCT saying anything about justice. Sorry for not picking up on your use of the word “just” before. How does RCT imply anything about justice? I provided you with a Wikipedia link earlier describing rational choice theory, and just is not mentioned in there, and I can’t recall ever hearing anyone before claim that RCT was a theory about justice, as opposed to decision-making. I think that’s why I missed your original use of the word and thought you were making a statement about what RCT predicts for the supply of public goods, not a statement about the justice or otherwise. I mean, gosh, what you think RCT is, is even further off the target than what I thought your original claim was. It’s one thing to get the prediction of RCT for the provision of public goods wrong, it’s another whole level to think it’s a theory of justice as well as one of decision-making.

3. I am still rather unconvinced that RCT undermined efforts to regulate. There are plenty of RCT theories that can be used to justify regulation, as I said before, just add in some assumptions about ignorance, or externalities. There are also other theoretical basises for failing to regulate, even if you think that people are irrational, once you also apply that to the people making up the government the case for regulating market players becomes a lot weaker. You haven’t addressed these points.

4. So your claim that RCT undermined any practical attempt to subordinate markets to anything at all wasn’t a claim about the real world? Then why did you use the word “practical”? Isn’t practical the opposite of theoretical?

RCT doesn’t describe the real world, which is the point I am trying to make
RCT doesn’t perfectly describe the real world. And for some situations it really sucks at describing the real world – eg decision-making under uncertainty. But in some cases it provides a reasonable model of decision-making in the real world, which is a good start. A map doesn’t perfectly describe the real world, but that doesn’t mean maps are useless.

5. Before I answer this, I want to know what you mean by “governing assumptions about human agency.” I mean, I can defend RCT as a model of human decision-making in many circumstances on the basis that all models are wrong but some are useful. But I just don’t know what you mean by “human agency” or “governing assumptions.”
I also want to know what evidence, if any, could convince you that RCT didn’t provide useful weight to attempts to deregulate or to fend off regulation.

Similarly, constant accusations of mental infirmity and ignorance might undermine my self-confidence but they won’t necessarily cause me lose faith in myself altogether.

I would much prefer it if they lead you to look at what RCT actually says before criticising it.

You asked for a reference for my claims about Kenneth Arrow –
“At its most basic, Arrow’s work demonstrated in formal terms – that is, in mathematical expression – that collective rational group decisions are logically impossible. Arrow’s paradox, or Arrow’s impossibility theorem, as it came to be called, presented an unshakeable mathematical argument that destroyed the academic validity of most kinds of social contract.”

Alex Abella, Soldier of Reason: The RAND Corporation and the Rise of the American Empire (New York: Harcourt), 2008, p.51. I spent some time in the British Library looking at the subject a couple of months back and didn’t find anything to contradict Abella’s description. We don’t rank our preferences in the manner that RCT claims, which makes RCT moonshine from the ground up.

2.

I am not sure what point you are making about public goods – I argued that it “followed from rational choice theory that only markets can discover and deliver public goods justly”. If collective decision-making is irrational by definition – as RCT claims – then only the market is left to deliver anything at all in a just way.

3.

To repeat – I said that RCT undermined efforts to regulate. To undermine something is not necessarily to destroy it – an undermined wall doesn’t always fall down.

Similarly, constant accusations of mental infirmity and ignorance might undermine my self-confidence but they won’t necessarily cause me lose faith in myself altogether.

4.

“You have now made two contradictory claims about markets:
In the real world markets require governments.
Furthermore, rational choice equated human motivation with self-interest in a way that undermined any and all practical attempts to subordinate markets to anything at all”

My claims are not contradictory – I was making a claim about the world and another claim about RCT - RCT doesn’t describe the real world, which is the point I am trying to make.

5.

Here are two claims –
1. RCT provided useful intellectual weight to attempts to deregulate or to fend off new regulation, contributing directly to the current financial crisis.

2. Its governing assumptions about human agency are indefensible.

Do you disagree with both claims?

So when politicians find a hint that RCT means people DO act rationally, they don’t read that as meaning we should try and understand the internal ‘rationality’ of an action, they read that as meaning all decisions are made for the common good of the community that makes them.

Which politicians? I’ve never run across a case of a politician taking that view. The 1984-1990 Labour Government in NZ for example introduced individual transferrable fishing quotas (ITQs) with the intent of reducing overfishing, which is hardly compatible with the view that all decisions are made for the common good of the community that makes them. The NZ National Government from 1990 to 1993 introduced the Resource Management Act, which amongst other things required developers, including state agencies, to have regard for environmental impacts and Maori and heritage values. Those were the two governments in NZ that I think are agreed are the closest to neoclassical policy, and I don’t see them carrying out your assertion.

And also in my experience politicians love to attack politicians in different political parties. About the only thing that overweighs this is appealing to very commonly-held values – eg kissing babies, favouring democracy. If one party of politicians adopted the view you outlined, I would be very surprised if the other parties didn’t oppose the view just for the political benefits of attacking it.

Incidentally, the introduction of ITQ is another example of an application of RCT to policy advice. Why are fish stocks collapsing while sheep and cow stocks aren’t, whatever else you might say about the environmental damage caused by farming? Because the lack of property rights to fish meant it’s rational for each individual fisher to overfish.

Tim Wilkinson:
and all the caveats and fine talk about not assuming a simplistic and deeply unattractive kind of material self-interest are largely ignored when it’s time to produce some figures

Firstly, rational choice theory doesn’t assume that people only act self-interestedly. It assumes that people maximise their utility, whatever that is. If you value something that is not in your material self-interest, and act rationally in order to get more of whatever it is that you value, then RCT has it covered. Building a church (temple, mosque, etc) may not be in someone’s material self-interest, but they can still go about building a church rationally.

As for ignoring this when it’s time to produce some figures – how do you explain therefore the use of surveys in valuing environmental assets? This is when the general public is surveyed about how much they value some environmental asset, not in terms of pure material self-interest, but to get at the values placed on things like sheer existance. More generally, cost-benefit analysis, for all its faults, including particularly its openness to manipulation, does generally include benefits to the wider community. See http://en.wikipedia.org/wiki/Contingent_valuation for some more information on valuing environmental benefits.

The arguments for Pigovian taxes or cap-and-trade systems can be based on knowledge, not on material self-interest. Take for example the case for placing taxes on greenhouse gas emissions. We want to reduce greenhouse gas emissions, but we also want to do so at the lowest possible cost, as we have many other claims on our resources, such as supplying an aging population. However, we don’t know the lowest possible cost. Consider for example the debates over food-cost miles. Buying your food from nearby sounds at first stab like a plausible way of reducing greenhouse gas emissions from transporting food, but it’s foolish if to grow the food nearby takes more energy than growing it further away in better climatic and geographical conditions. But it’s awfully difficult to calculate when food is better grown nearby versus far away. A tax, or a cap-and-trade, on all greenhouse gas emissions would mean that the cost of those emissions will be reflected in the price, avoiding the need for all those arguments. An economy made of altruists would need prices as much as an economy made of the most simplistic, selfish materialists. See http://en.wikipedia.org/wiki/Pigovian_tax for more on this.

As far as I can tell, the microfoundations of macroeconomics are, well, shaky at best, so I would like to see the papers in macroeconomics that convinced you of your statement above with respect to macroeconomics.

So I am rather unconvinced that your claim about what happens is accurate.

It would be better to admit that you don’t know, or that you are never likely to know, but instead it seems to be assumed that findings must be produced (that’s what economists do, you see), and however divorced from reality, some results are better than none.
Yes, that is an important force in applied economics. Governments, well at least democratic governments, want some advice on what to do, and are very reluctant to accept the laissez-faire answer of “do nothing”. And why do democratic governments want this sort of advice? Because much of the voting population does want governments to intervene and do something. And of course economists who don’t want to compromise themselves by providing practical advice generally don’t apply for government jobs doing that sort of work. And for those who do get the jobs, using a model at least provides some protection against claims of individual bias and arbitrary decison-making.

Well yes, rational choice in the minimal sense you describe is a default heuristic for interpreting behaviour that everyone uses – historians and sociologists included. Even those wedded to abstruse theories and vocab about various kinds of social forces can’t help using the standard categories of the intentional stance when their guard drops (or their quoting fingers get tired). Even the egregious Freud can be seen as describing subconscious rational choice (in the minimal sense of means-end rationality) on the part of multiple sub-selves acting under jejune (not irrational by RCT standards) beliefs and desires.

But this has nothing to do with RCT as a pervasive economic model stuff about Israeli students is, topically enough, not even economics as we know it. A sociologist, historian, psychologist might provide the same analysis. The point is, it’s explanation. In explaining an event or fairly discrete phenomenon we can happily start from a presumption of practical rationality and modify, complicate extend or qualify that account in the light of how well the explanation fits the observed facts (including facts established after we start explaining and testing hypotheses).

Economics is not primarily about explanations, and especially not about historical (singular) explanations (leave those to speculative evolutionary theorists who do such a good job of replacing Kipling’s efforts with a close substitute.)

Predictions are the thing, and while it’s true that many defects of RCT (in its specifically economic sense) are well-documented and widely acknowledged by economists, it’s mostly lip-service as far as I can see – the underlying conception still holds sway over the maths, and all the caveats and fine talk about not assuming a simplistic and deeply unattractive kind of material self-interest are largely ignored when it’s time to produce some figures. Humility then goes out the window and instead airy assurances abound – the model is inaccurate, but it’s probably good enough. In fact, it’s not even necessary to say the second part. Presenting the results implies it. I’m reminded of TV or radio shows where they say ‘well let’s have a show of hands. Well, according to our unscientific straw poll…’

It would be better to admit that you don’t know, or that you are never likely to know, but instead it seems to be assumed that findings must be produced (that’s what economists do, you see), and however divorced from reality, some results are better than none.

What I mean is if RCT can help us know that most decisions are ‘rational’ but are nonetheless constrained by ‘stuff’ isn’t that robbing RCT of any normative or analytical value? What in RCT allows us to know which ‘constraints’ are the ones we should look to? And also how does RCT allow us to understand whether a decision was a good or bad one?

I come at this from the angle of being a Political Scientist or Historian (so, no doubt lacking in the scientific rigour required to comment about these things). But, whilst I’d agree that LBJ’s decision to escalate the Vietnam War was completely rational and I’d be inclined to think George W. Bush’s decision to Invade Iraq was also rational (given allowable and understood constraints), I’d nonetheless also argue they were both terrible decisions. The irrationality or rationality of the decisions doesn’t change my judgement, or effect my explanations for why the events happened. Whether they can be called rational or irrational seems to miss the point.

Anyway back to the original discussion, what I think that perhaps what you are missing is more often than not the words ‘rational’ and ‘irrational’ are synonyms, especially in politics and policy making, for ‘good’ and ‘bad’. So when politicians find a hint that RCT means people DO act rationally, they don’t read that as meaning we should try and understand the internal ‘rationality’ of an action, they read that as meaning all decisions are made for the common good of the community that makes them. So, bankers make ‘rational’ decisions (as a basic reading of RCT tells us) and to policy makers this means the decisions will be guaranteed to be ‘good’, in this case make lots of money (which RCT does not conclude – it can’t). So, this isn’t the fault of RCT then, just the fault of whoever exposed RCT to policymakers and didn’t explain that it doesn’t mean what they think it did. As a Brit I’d therefore smugly suggest American academics are to blame for too easily wanting to be heard by policy-makers. This is clearly not something British academics do. Anthony Giddens being a prime example of how we keep well out of things!

Hang on isn’t Giddens a ‘Sociologist’? I’ll get my coat…

RCT is the idea that individuals choose the best action available to them given stable preferences and their constraints. RCT encourages respect of other individual’s decision-making processes. When a rational choice theorist sees another individual making a decision that seems irrational, RCT implies that the theorist should be asking “what are the constraints that that individual is under”? And this question often leads to deeper understanding, take for example “Sect, Subsidy and Sacrifice: An Economist’s View of Ultra-Orthodox Jews”, Berman, August 2000. In this article the author investigates why Israeli Ultra-Orthodox men study full time in yeshiva until age 40 on average, while similar religious groups in the USA only study to their early twenties. The explanation he comes up with is that the men are signalling their commitment to their religious group by making a costly signal of commitment. But Israeli law exempts religious students from the draft, so to prove their commitment to the group, and not merely to draft-dodging, the Israeli men have to study for considerably longer than the period for which they are exempt from the draft. If Berman had just said “well what do you expect, people aren’t rational”, he would have missed out on this possibility to gain a deeper understanding. The guidance is not perfect of course, but there is some pressure to be internally consistent, and this has some value.

There is a certain arrogance I find in many advocates of irrational models of decision-making, to take a really gregarious example, Freudian psychoanalysis assumes that the psychologist knows far more about other people’s motivations than those people do. Freudian pyschologists were notorious for interpreting all arguments against their theories as simply evidence of psychological flaws in their critics. No humbleness there. See for example http://www.human-nature.com/freud/tallis.html for an outline of what I am talking about. Once we start assuming that other people are irrational, and particularly the more easily we make that assumption, what can cause us to stop?