anyone looking for a definition of economics is going to be disappointed because economics is not united enough to agree on one. i have heard professional Economists define economics as the Inada assumptions of the productoin function, I have heard it described as a tribe by historical economists, and I have heard it confused with a non-science like mathematics (which is not a science because it deals with absolute truths that arise from form. Some of economics is a science, i.e. the bit that tries to formulate and refine theories to fit observations as good as possible). Where economics is a fuzzy science is that none of our theories holds all of the time (to great frustration of economists, I might add) and hence we’re in a world of provisional and conditional ‘reasonable fit’.
The line that economics is an approach and does not have a core ‘empirical territory’ always struck me as false. There might be an imperialist element to economics, much like any group that is successful tries to grab a bit more territory, but for me, at core we primarily deal with materialistic aspects of life: that is where our own bread gets buttered. Having said this, I share Don’s implicit wish for a theory of everything and feel that social scientists should indeed try to formulate more general theories and stories that put these core territories in their perspective.

My own answer to what utility is was posted as a comment to Don’s blog about it.

If economics can clearly explain the situations where it will work and the situations where it wont work (for whatever reason) then it can still be very useful

Tel (38) – I posted my comment before I read yours. I agree.

And if we do find circumstances where the approach doesn’t work, then perhaps these can teach us something about market behaviour too. It’s hard to image that people’s behaviour is completely ‘economic’ in some domains and ‘non-economic’ in others.

As for Becker. I do find his arguments for household production persuasive. I find it difficult to accept that people derive satisfaction from washing machines in the same way they derive satisfaction from chocolate.

When it comes to improving the theory, I much prefer his strategy of applying the economic approach unflinchingly to all human behaviour than to that of an economist who relies on some vague sense that the approach isn’t appropriate in this or that circumstance but never explains why.

On point (1), you have made a commitment to take real-world measurements and all measurements have errors. To do the job properly requires some analysis and understanding of those errors, in order to judge what is theoretically achievable. It would be astounding to be able to get consistently accurate outputs from error prone input data.

On point (3) it is not a matter of “acceptance”, we have a well understood way of testing predictions and using those tests to evaluate the theory.

From Don:

So it seems to me that if the economic approach fails to explain human behaviour in areas like marriage and religion then it fails full stop.

Not true at all. If economics can clearly explain the situations where it will work and the situations where it won’t work (for whatever reason) then it can still be very useful (albeit not the ultimate tool for all jobs). The key is being able to establish a boundary (and nice to be able to establish a theoretical basis for that boundary, but an empirical construct would be adequate).

At the moment though, economists don’t seem to have any idea of where the boundaries are, let alone why.

For example, economists don’t seem to deliver a clear definition of what is trade. Let’s look at the ETS… on what basis is trading emission certificates any different in fundamental terms to just paying a carbon tax and trading regular dollars like we always do? In practical terms, emission certificates are much more difficult to audit (need to keep a chain of custody through the entire system to catch the cheats) and much more expensive to implement (need to build an entire new handling system in parallel with our existing financial system). In theoretical terms since a certificate is always exchangeable for dollars anyway, trading dollars should be theoretically equivalent.

You would think there would be some basic principle to fall back on in order to evaluate such a situation.

You’re right, I have appealed to economic reasoning. And I can’t see how we could dispense with it.

But I’m puzzled and disturbed by economics. I’m persuaded by Becker’s argument that economics is defined by its approach rather than its subject matter. And if it applies to market production it also should apply to household production.

So it seems to me that if the economic approach fails to explain human behaviour in areas like marriage and religion then it fails full stop.

I worry that a “rich and eclectic approach to economics” is just an excuse for ad hoc explanations and a crude and ignorant kind of instrumentalism (ie “We know the assumptions are wrong and the theory’s predictive power is fragile and context dependent, but we’re just not interested in finding out what actually explains human behaviour.”)

I assume that by ‘involves money’ you are speaking very loosely, and mean relating to the production, distribution and consumption of wealth; if so, why shouldn’t there be a branch of social science devoted to that spahere of activity?

You’ve written a couple of posts about house prices, separating supply and demand factors and noting the role of prices in balancuing these. What were you doing then if not economics?

I think that what really bothers you is not economics per se but the Chicago Scool’s narrow conception of behaviour, and in particular its attempts to extend the atomistic, maximising, choice model to social phenomena like marriage and religion that fall outside the scope of traditional economics. But plenty of economists dislike this approach too — you might enjoy Geoff Harcourt’s essay ‘The social science imperialists’ if you haven’t come across it.

It makes more sense to reject Becker’s reductionist dogmas and defend a rich and eclectic approach to economics, than to take his definition at face value and make it a basis for rejecting economics’ claim to scientific status.

On your points:

(1)As I understand it, the set of phenomena characterised as economic means all human behaviour that is (i) goal directed and (ii) requires scarce means with alternative uses.

Lionel Robbins writes:

Economics is the science which studies human behaviour as a relationship between ends and scarce means which have alternative uses (pdf).

Gary Becker takes a similarly broad approach. He writes:

I believe that what most distinguishes economics as a discipline from other disciplines in the social sciences is not its subject matter but its approach.

(2)There’s a great deal of overlap between economics and other social sciences. For example, recent debates about asset price bubbles are as much part of the domain of psychology as of economics. And of course public choice theory, the economics of the family and other excursions have taken economics into many areas occupied by other social sciences.

On the issue of utility maximisation, Becker writes:

Everyone recognizes that the economic approach assumes maximizing behavior more explicitly and extensively than other approaches do …”

On Mill – I’m inclined to treat Mill’s political economy as an ancestor of modern economics — related but not the same thing.

I understand that there’s an English tradition where the philosophy of utilitarianism (more Benthamite than Millian) and economics were combined. FY Edgeworth is a good example.

But the attack on cardinal notions of utility broke the link between the two and economics headed in a different direction.

Don, as long as you accept that (1) there’s a set of inter-related phenomena that that can be usefully characterised as economic, (2) no other science tries to explain these systematically, and (3) the phenomena are amenable to rational inquiry — if you accept these, I can’t see the point of making an issue of whether economics is a science.

I don’t think many economists, even neoclassicals, would agree that economics as a science ‘starts with the assumption of utility maximisation’. They would, on the other hand, go along with the general thrust of Mill’s essay on the scope and method of economics, which tries to locate the boundaries, and to didtinguish between an art and a science. If you’ve read it, you’ll know there’s nothing about utility maximisation in there, nor any concern with ‘making problems tractable to the math’ (don’t we say ‘maths’ in this country, by the way?).

I’d also be interested to see the average & standard deviation of the adjustments being made. By the same law, if I understand it correctly, those adjustments should average to somewhere in the vicinity of zero.

… I would stand up to say that economics has been very succesful and should not be depicted as the lame brother of more succesful sciences.

Is economics a science?

And what is economics anyway? It starts with a set of assumptions — people maximise utility, their behaviour takes place within markets (actual or metaphorical), their behaviour is rational (completeness, transitivity etc)and they have stable preferences.

If you deviate from these assumptions (eg behavioural economics) then you stop doing economics and start doing some other kind of social science.

The assumptions seem to exist in order to make problems tractable for the math. And that’s what most economists seem to want to talk about. Not whether the assumptions represent the bones of a testable theory, not whether the assumptions are realistic or refer to entities that actually exist, not whether the assumptions lead to predictions which are demonstrably false (eg gift giving at Christmas).

Take ‘utility’ for example. Most economists refuse to say anything about it — they’ll tell you it’s whatever people maximise. A few will say that it’s ‘happiness’ or ‘pleasure’ minus ‘pain’ (as we’d learned nothing in psychology or philosophy since Bentham). So the answers are either vacuous or demonstrably false.

Many of the arguments for economics’ status as a science appeal to ideas about what physics is that were current in the 1930s (eg operationalism, instrumentalism). Did we mention the physics envy?

I’d be the last one to argue that the success of economis is based on its data, but I would stand up to say that economics has been very succesful and should not be depicted as the lame brother of ‘more succesful’ sciences. Many of our institutions, like the ACCC, the Reserve Bank, the various regulators, auction mechanisms, trade agreements, tax systems, etc., reflect mainstream economic thinking. At least part of the reason why the last recession was so mild was because the government almost blindly followed the advice of the economists in the Treasury and the RBA, who quickly instigated a stimulus package (go early, go hard, go consumption) and, together with the other governments who were given the same advise by their economists, resisted the temptation to go protectionist. Economics is a very inexact science, yes, and we learn more from institutional experimentation and basic theory than our supposed ‘data’, but we learn nevertheless and our societal structures reflect the learning from economists.

Now, as to the data of other, I think you underestimate the degree of fudge and carpet sweeping that goes on in things like medical laboratories. It has been rumoured that more deaths occur due to medical mistakes than on the road, but more to the point: an awful lot of ‘abberant human and animal subjects’ get weeded out of the data before publishing. The notion that you truly get to see all that goes on in labs and that all that is verifiable and can be blindly trusted is a myth.

I fear that if you demand pristine data to deserve the accolade of science that you wont be left with many fields. Do you for instance think the data coming out of supercolliders needs no cleaning? Dont get me started on medical data.

All measured data has errors, but never the less some is better, some is worse. My question is whether those in various disciplines have been realistic regarding the analysis of those errors. Have they even done any error analysis?

When it comes to cleaning, it also comes down to reasonable justification for the cleaning and careful documentation of the process involved. Not a matter of black and white, but deeply a matter of opinion. Why does the data need cleaning? Does this process make more accurate data or merely more consistent?

The particle smashers have the luxury of a great many repeated tests, allowing them to build up a statistical profile. Economists and climate scientists don’t live long enough to be able to rerun their theories thousands of times over.

Medical data is suffering partly from the problem of lack of repeatability (although not so much), partly from the problem of vested interests getting involved (likewise Economics and Climate science both, but the Economists at least understand a few guiding principles explaining how vested interests operate, generally choosing to ignore those principles for the obvious reasons). The medical industry also operates to a much tighter budget than either Economics or Climate Science (after all, the doctors are encouraged to produce real results for their money, and then do research in whatever time is left over).

One of the biggest problems in the medical industry right now is lack of actually looking. For example, a great many “Swine Flu” confirmed cases don’t go through rigorous DNA testing, they just seem like similar symptoms. I’ve heard there are cases of pneumonia and TB getting grouped in the same bucket as swine flu which makes you wonder why they bother to keep stats at all. I could also point out the conflict of interests that if I discovered that doing some simple thing would make everyone healthy, getting a profit out of that discovery would be impossible — so naturally I always discover something that is expensive, difficult to obtain and makes people feel healthy for a short time only.

However, when it comes to genuine deliverable outcomes, let’s be honest… Medicine (imperfect though it is) has delivered more than Economics, Climate science and the various particle smashers (no doubt some government paper-shuffler is still hoping that the Large Hadron Collider is going to pop out a recipe for a bigger and better bomb, and no doubt the researchers don’t feel inclined to explain otherwise).

But you know, when was the last time you tried to phone America and ended up talking to Mongolia instead? Pretty reliable system isn’t it? Hugely complex, but yet predictable. To make that happen required a lot of people to check things very closely, again and again. Turn your lights on, and they come on. Open your fridge, and it is indeed cold. Nothing natural about that, these thing are hard work in crystallised form.

How much software goes into an Airbus? Admittedly they do crash now and then (because of tin whiskers — blame the Greenies, not the Engineers) but if that software was delivering the haphazard results that Economists get away with then we would not need airport security, because we would not have any airports. Fair scope for comparison in my opinion. I still hold that there are some solidly established sciences that may not deliver pristine data, but they do deliver a substantially higher standard.

fair question. When I think about the likely candidates, I would think about the countries most likely to fear submergence under a rising sea. The small Island nations, together with Bangladesh, the Netherlands, perhas even with some individual states of larger countries, could form a separate coalition. Several of the schemes would indeed seem far out of reach of that type of coalition, but chucking large amounts of sulphure oxide into the atmosphere is not an impossibility for such a sub-set, nor is experimentation of chucking certain chemicals into the oceans.
The point of an ETS depends on whether you believe it will work. if you believe it will work, it sounds like an economic solution. If not, the point is to pretend to be doing something whilst you keep exporting coal.

For peoplewho think that an effective ETS will not happen., geo-engineering seems the only realistic way forward, i.e. some kind of technological fix that can be implemented by a single worried country or a sub-set of countries desperate enough to try unproven technology to cool the planet down.

Which worried country would do this and why? A rapidly submerging Fiji? Not likely. A rich country or bunch of countries that stand to lose by climate change, like the US? Do you really think that the US would come up with the money to put reflectors at L1 for such a huge price, let alone be allowed to tinker with our planet unilaterally by the rest of the world.

Surely, the whole point of an ETS is to make solutions become economic, perhaps grand expensive schemes like the ones you describe but, more likely, hundreds of smaller technological improvements.

I agree that the temptation you sketch is certainly there. The more power and influence that is on offer for the ‘winner’ of a peer review process, the more the peer review system will be actively corrupted. The same forces are at work in economics too, and it will only get worse as the outcomes of the peer review process are more and more used as indicators by governments and universities to determine rewards.

As to these particular emails, I did not see true falsification discussed. What was discussed was selective representation, a bit of politically motivated reluctance to truly explain the data, and active networking to smear other scientists. Its the last thing that worried me most, but nothing in the exchange really put the basic data in doubt.

As to why we should still put any trust in what most climate scientists seem to think, i refer to my answer under #10: if there is anything i do trust, it is the outcome of competition between ambitious young scientists. if there was a serious known alternative contender, we would have heard about it. It is still possible that the current dominant story is puffed up and wrong, but the longer it takes for the hidden alternative to emerge, the less likely the alternative exists.
As to temperature variability, the climate scientists themselves are very aware of this and seem to be able to discount the most obvious sources of ‘natural’ variation (such as solar flares). The failure of the temperature data to show an increasing trend in the last 10 years is of course not good for the AGW story, but dont forget that this still means the last 10 years is a lot warmer than previous periods, i.e. the lack of continuing trend does not mean we are back where we were in 1800. the possibility that some as yet un-understood process is keeping temperatures increases down is of course quite possible and may well be the source of the next big climate story. We can only speculate at this point.

“Paul, Whilst I take your points about incomplete and fuzzy data and its anlysis and interpretaion, that was always going to be a problem proving AGW once some bright spark came up with the notion and the idea took off that it was worth investigation and resources being allocated for that purpose. I’m an agnostic on AGW and even I can see the obvious safer road of say Plimer’s stance. Not hard to see the huge variability in climate over time and appreciate our limited and somewhat feeble attempts to explain it all. With a decent temperature record of only 150 yrs or so and given the time scales of the earth this was always going to be a tough nut to crack and we all need to acknowledge that.

Therein lay a great temptation to see things in the record and data if you came to the problem with preconceived notions that this was important green work you were embarking upon. Basically it attracted a certain type of scientist with a feeling that mankind was being a bit unkind to Gaia and if AGW could be proven, there for all to see would be the defining proof of that. IMO that temptation proved fatal as the inner circle tightened and outsiders were increasingly excluded and eventually frozen out. Then when they ran into the inevitable data problems and the overarching difficulty of the task facing them, they began unwittingly at first to see things that weren’t there and inexorably to drink their own bathwater more and more as they faced the obvious and easier skeptic questioning. You can see it clearly when they collectively began to believe it was their precious data and noone elses despite the obvious.(its our preciousses eh Smeagol)

But there was worse to come as they increasingly engaged in the political process and science turned to advocacy. Oh it starts out as this is important green work on behalf of all mankind with more resources needed and slowly but surely degenerates into advocacy. They attached themselves to the UN and became shills for global ETS and right there was the danger Crichton warned so poignantly about in 2003. To actively connive to conflate critics of a global ETS with AGW deniers was to commit them fully to the political hack road. Well if you become a political hack you can’t complain when you’re hacked down by the political process.

The emails between the inner circle show how corrupted they’d all become and after talking about lunching the raw data if an FOI was successful, that’s what they either did or else were deliberately lying in response to reasonable and mandated FOI requests. Indefensible by any stretch of the scientific imagination and they’ll rightly reap the whirlwind for it. However their greater crime will ultimately be to smear all AGW science and dry up any research resources for it in the forseeable future. When this scandal has finally run its course, you’ll have buckleys of getting AGW research grants out of the pollies for a long time, mark my words.
”

Note by Paul: For some reason this comment was sent to me as a subscriber to this thread, but did not survive some other unknown process (perhaps a crass in the night, or something like that), hence I am reposting it.

”
When we burn coal we make lots of money, Only problem is CO2 is released. Now we have to recapture the co2 and put it back in the ground at a cost of lots of money.
Seems to me it would be more logical to leave the co2 (coal) in the ground in the first place and just build more safe, modern nuclear power stations.

”

Note by Paul: For some reason this comment was sent to me as a subscriber to this thread, but did not survive some other unknown process (perhaps a crass in the night, or something like that), hence I am reposting it.

the link to the Washington times you post brought me to a hyperbolic editorial. In between stories of cocaine in chicken and the obligatory ridiculing of Abama’s health plan, I saw nothing but tendentious reporting. Forgive me for not taking this outlet as the decider as to what is official and what is not.

_Tel,

I fear that if you demand pristine data to deserve the accolade of ‘science’ that you wont be left with many fields. Do you for instance think the data coming out of supercolliders needs no ‘cleaning’? Dont get me started on medical data.

To bag the climate data simply because it is not pristine says more about the ignorance of the commentators on how science actually operates than that it is a reflection on the validity of what was done.

I hate to be the one to break this but there are people who don’t see Economics as the archetypal standard in how science operates.

To be fair though, Economists and Climate Scientists have similar track records on the predictive capabilities of their theories, so Paul, maybe you have unearthed a deeper theme running through both disciplines that deserves closer attention.

Simple. The dog ate the homework, but he didnt know the dog ate it until recently. Dogs are sneaky that way. Even though he refused requests to provide the homework, which he didnt have, but didnt know he didnt have, he cant really be faulted for not providing something he didnt have, even if he thought he did. Thats clear, isnt it?

Sadly, back in high school I was too stupid to deliver such humdinger excuses and was consequently forced to learn mathematics, physics, etc.

Second prize is better than nothing I guess… *SIGH*