Eric the murdering economist
July 7th, 2009 at 1:04 am by David FarrarI have previously covered the damning critique done my Eric Cramption and Matt Burgess of the BERL study which found the cost of alcohol consumption was around $5 billion a year. Crampton and Burgess cited multiple errors in the BERL study (including no counting of benefits) and concluded it was actualyl less than 5% of that.
NBR reported last week a response (finally) from BERL:
Adrian Slack says Berl was only commissioned by the Ministry of Health and ACC to look at the social costs and not the benefits of alcohol, and would have needed an additional $135,000 were it to extend its remit to examining the benefits and policy implications. …
A pity that it was not made clear at the time it was costs only. I wonder why the Government would see value in commissioning a paper that looks at costs without benefits. Anyway onto the next comment my Mr Slack:
He accused Dr Crampton & Mr Burgess’s critique as being based on strong assumptions about perfect markets, perfect information, and individual rationality.
“So for example someone who murders someone, from the individual’s point of view, Eric would be, I presume, quite comfortable with that. The person who decides to murder someone else makes an evaluation of what are the benefits and costs to me of this action? Society says ‘well some people do murder other people’, but society says ‘that’s not good.’”
Now that was not a type. He just said that Eric Crampton would be comfortable with someone murdering someone (from an economic perspective). This is BERL’s response instead of a detailed point by point response to the 30 to 40 errors cited in the report?
Paul Walker responds with disbelief – not just from the sillyness of the analogy, but the repeating of economic mistakes:
If the only costs of murder were the internal cost to the murderer then we may not be too concerned with murder. BUT, there are some obvious, to most people if not Adrian Slack, external costs to murder, that is, the loss of life of the victim. The victim is the victim because they have not willingly agreed to be murdered, that is what makes murder, … well … murder.
I have no doubt that both Eric and Matt are opposed to murder, and for the very good reason that it violates the victim’s property right in themselves. Murder is not a market transaction in the sense that it is not a voluntarily agreed to trade resulting in both parties being made better-off.
One of the major points that Matt and Eric made about the BERL report is that BERL didn’t seem to know the difference between internal and external costs. The Slack quote above only reinforces that point.
Indeed an own goal. Eric Crampton also responds:
Economists tend to think that murder is a bad thing. Why? Well, despite the murderer presumably enjoying the act, his gain comes at a cost that he doesn’t personally bear: the death of his victim. That’s the kind of cost that economists tend to call an externality. And so economists tend to support laws against murder. We similarly tend to support laws against theft: while the thief tends to think taking other folks’ stuff is a good idea, the thief’s victims tend to be hurt by it and the thief won’t weigh those folks’ losses against his gains. In these kinds of cases, individuals’ rational calculation of their own costs and benefits lead to socially bad outcomes because of the substantial external costs.
Eric goes on to say that having BERL paint him as pro-murder (economically) is gettign close to a version of Godwin’s Law where you should concede defeat if that is the best you can do.
Blaise Drinkwater also comments on the costs vs benefits issue:
But just because I buy that the BERL report is a costs curvey only, I’m not obligated to buy the report, which bungled the costs badly. Remember, the BERL report said that alcohol costs New Zealand’s society the equivalent of $4,794m, using an “international framework” that seems to have as its main justification the academic equivalent of a circle-jerk. Burgess and Crampton, employing more mundane economics, came up with a figure of $662m. BERL is yet to explain satisfactorily why their headline figure seems to out by a factor of seven.
That is what I am most interested in. I do hope BERL does a more robust and detailed response than they have to date, so people can then judge with confidence which figure is most useful.
Tags: Adrian Slack, alcohol, BERL, Blaise Drinkwater, Eric Crampton, Matt Burgess, Paul Walker
July 7th, 2009 at 1:58 am
BERL has just put out a public response here, which is excellent news. Eric and I are writing a response.
Treasury has also issued an apology which BERL has put on their home page.
Eric has published a couple of adjustments to the working paper here. Cost adjustments add $36 million, but thanks to a pointer from Treasury we also found an additional $197M excise taxes not previously accounted for. These offset costs. So alcohol is now showing a small overall benefit after excise of $37 million.
However, as Eric points out, that number is so small compared to margins of error that zero is probably a good approximation.
Vote:July 7th, 2009 at 7:33 am
> That’s the kind of cost that economists tend to call an externality. And so economists tend to support laws against murder.
Oh, right, so economists don’t support laws against murder because, well, murder is wrong and illegal, they support laws against murder because it creates an “externality”. Haha, you’ve got to love economists. It reminds me of the joke about the 3 guys adrift on a liferaft with lots of canned food but no can-opener. While the other two guys are debating how they’re going to opne the cans, the economist says, “Don’t worry guys, let’s just assume we have a can opener”.
If there were no external costs, would Eric support murder? Bizarrely, his answer seems to be yes.
Vote:July 7th, 2009 at 7:57 am
When was the study commissioned? Who drew up the terms of reference? My Google-Fu is weak today….
Vote:July 7th, 2009 at 8:13 am
I think Ross, in the context of a Cost-Benefit Analysis- which is where the murder debate was being discussed- the externality argument is perectly fine. This does not preclude there being other reasons why economists would be opposed to murder.
To put in bluntly, all disciplines that deal with complex systems use assumptions. I’ve done extinction-modelling for endangered species and these models are riddled with assumptions about the drivers of extinction (and often lack necessary feedbacks). The trick is that in ecology, you make predictions about the probability of persistence over 100 or 1000 years. So, no one is going to know how good these models are until the year 3000.
I might observe that the NZ economist Bil Philips did in fact build a secret radio at great personal risk to himself as a prisoner of the Japanese during WW2. I’m willing to gamble that this was done with real parts, not parts that were assumed to exist.
Vote:July 7th, 2009 at 8:55 am
Chthoniid
That was an admirable response, making your point firmly without resorting to name-calling.
I was typing a response to Ross that was both longer and less clear -I can now safely consign it to oblivion.
Vote:July 7th, 2009 at 9:21 am
Right choice seeing MIT. It was an ailing institution revived by a Kiwi.
Vote:July 7th, 2009 at 9:32 am
I sincerely hope someone in the Law Commission did not want to commission a report saying what some people on the Commission wanted to hear just to suit their own viewpoints and to please Helen Clark. There is a member of the Commission who is an expert at writing Bill of Rights Act Section 7 reports for Parliment saying that the most repressive legislation passes muster with the Act.
If the report was only supposed to look at costs rather than benefits, then it should have clearly said so.
Vote:July 7th, 2009 at 10:02 am
For a second I forgot that scumbag Clayton Weatherson’s first name, thought it was another post about him given the even worse revelations which came out yesterday ( just when you think it cant get worse ) – another murdering economist (although a pretty big failure as an economist by all accounts ) I can see him actually using some of the above arguments in his defense, and thinking they will work – thats how out of touch with reality he is.
Vote:July 7th, 2009 at 10:21 am
Godwin indeed, these guys are getting paid money for an highly skilled technical analysis and the best they can do in their defence is come up with arguments that would embarass fourteen-year-olds on a Star Trek forum…
Vote:July 7th, 2009 at 10:28 am
All this discussion about the BERL report estimating the harmful cost of alcohol at $5.4 billion completely ignores earlier research by Brian Easton which put the harm at over $16 billion in a publication titled:
THE SOCIAL COSTS OF TOBACCO USE AND ALCOHOL MISUSE
http://www.eastonbh.ac.nz/?p=59
To a non economist like myself, it seems to me that Crampton and Burgess are living in an ivory tower in La La land.
Vote:July 7th, 2009 at 11:21 am
ross,
>If there were no external costs, would Eric support murder? Bizarrely, his answer seems to be yes.
If there were no external costs (ie no negative consequences to anyone at all) then it would not be called murder (supporters of euthanasia dont call it murder for example)
You also have missed that it was the BERL boffins who started this inane argument about the economics of murder
Vote:July 7th, 2009 at 11:34 am
@yesbut
Note that Easton does not consider any benefits of alcohol consumption, and the paper’s methodology extends to all risky behaviour. Risky behaviour that leads to increased incidence of accidents or death would also create ‘like’ costs (reduction in GDP).
Hence, according to this logic we would not want people to become firemen, ride bicycles, go horse-riding or play rugby because of its social cost.
Vote:Hence, the methodology has not been duplicated since 1997.
July 7th, 2009 at 11:53 am
peterwn – in BERL’s defence they do say their’s is a cost only study. But then they confuse things by considering benefits throughout their report, usually but not always deciding they are zero. BERL repeatedly refers to their study as an analysis of welfare, which requires an assessment of benefits, and they frequently refer to “net social costs” which means something other than the obvious (to us) meaning of costs less benefits. It’s easy to see why the Law Commission and others, including us, have had a hard time understanding what’s going on. When the Law Commission did make its mistake, BERL refused to step in and sort it out.
As they point out in yesterday’s response, BERL was only asked to consider costs. But for some reason they went the extra step and looked at gross benefits too, and decided quite inexplicably they are zero (almost) across the board for anyone drinking a bit less than two pints per day on average. Why they didn’t just ignore benefits entirely we don’t know.
We put a section in our critique titled “Is this a cost/benefit study?” which points out all of this and notes the difficulty of counting costs but not benefits when the line between the two becomes blurred.
Vote:July 7th, 2009 at 2:10 pm
Matt: I’d put it a bit differently. The only reason they’re able to count private costs as social costs, in their framework, is that they define there as being no compensating benefits. Without that assumption, private costs cannot be social costs.
On how economists view murder: David Friedman has a very nice argument about it in Law’s Order.
Vote:July 7th, 2009 at 4:36 pm
I tend to side with BERL on ths issue. The Tresury’s Peter Bushnell criticised the BERL report even though he hadn’t even read it. Hmmm, that’s professional. And now he’s been forced to apologise.
“My comments were in response to second hand information and as a result I made criticisms of the BERL research that, with hindsight, I would not have otherwise made. I accept that the BERL research is not a full cost/benefit study and regret that the media coverage didn’t adequately convey that BERL worked to a specific brief to research only the costs of harmful alcohol and other drug use. I recognise that this new media attention has placed BERL in a difficult position and apologise for the part that my reported comments played in this situation.”
So the media did a shoddy job of reporting the BERL study. What a surprise. The title of BERL’s study was a dead give away: “Costs of Harmful Alcohol and Other Drug Use”. So it was pretty obvious that BERL did not take account of any benefits. To be sure, the study’s introduction makes mention on several occasions that the study is all about measuring the costs of (ab)using alcohol and other drugs. It’s the media and Treasury that have been negligent here.
Vote:July 7th, 2009 at 4:48 pm
Indeed, on page 8 of the BERL study, the authors write:
“This study concentrates on the economic costs of harmful use, and does not explicity estimate the the social impacts of benign or beneficial use. Literature on beneficial use was not specifically reviewed, and estimated impacts of beneficial drug use were not included in this study. However, there are intangible benefits…from non-harmful consumption of alcohol or other drugs. As these impacts are benefits, however, they do not fall within the scope of of this study on harmful drug use.”
That seems pretty clear to me. Which part of that did the media and Treasury find so difficult to understand? Or did neither get to page 8 before they mouthed off?
Vote:July 7th, 2009 at 5:26 pm
Ross, I went a bit beyond page 8. In fact, I got all the way to the end. And look what’s buried at page 173.
In short: the only reason BERL got to count private costs as social costs is because they decided that private benefits equal zero. This is somewhat different from taking no position on benefits or deeming benefits outside of remit.
Vote:July 7th, 2009 at 9:15 pm
“Oh, right, so economists don’t support laws against murder because, well, murder is wrong and illegal, they support laws against murder because it creates an “externality”. Haha, you’ve got to love economists. [...]
If there were no external costs, would Eric support murder? Bizarrely, his answer seems to be yes.”
Ross, you hit it spot on. I (speaking for myself, not Eric) would be fine with murder if there were no external costs. The thing you (and I) don’t like about murder is that it kills someone who didn’t consent to it – that’s the external cost, and the reason it is wrong. It’s not an internal cost because it’s not borne by the actor (the murderer).
If you think there is another reason murder is wrong (other than the externality), I would be interested in hearing it.
Note in suicide, the cost is borne by the actor. However, do note that there are other external costs put on friends and family of the victim/actor. This is usually the reason people give for thinking suicide is wrong (other than the obvious waste of a life).
Vote:July 8th, 2009 at 6:56 am
Eric,
Well, it’s clear that BERL think that “harmful alcohol use has zero private benefit”. But alcohol is not just consumed in harmful quanities, and it appears that BERL is aware of this. So I don’t agree with your point. Even if I did agree with it, it’s irrelevant in the context of the BERL study because the study is all about costs. I can see why it’s about costs. I would not expect emergency staff called out to a car accident where the driver, who’s intoxicated, has wrapped himself around a power pole and is on death’s door, asking the guy about the benefits of drinking.
We know that drinking has benefits but we also know that it incurs significant social costs. ACC and the Ministry of Health wanted to quanitfy those costs which is why they commissioned BERL.
Vote:July 8th, 2009 at 7:01 am
Having said that, I would take issue with BERL’s estimate that 50% of drinking is harmful. That seems high. But then you are always going to have disagreements about the nature of assumptions made by economists. That’s why economics is and always will be a social science and not a hard science.
Vote:July 8th, 2009 at 8:51 pm
“Well, it’s clear that BERL think that “harmful alcohol use has zero private benefit”. But alcohol is not just consumed in harmful quanities, and it appears that BERL is aware of this. So I don’t agree with your point. Even if I did agree with it, it’s irrelevant in the context of the BERL study because the study is all about costs.”
But Ross the study we are told is all about costs, so you have to ask, Why are they discussing benefits? Its either a cost study and so no benefits get a mention or its a cost-benefit study in which case you deal with both, properly. Why, if it is out of scope, did BERL assume zero benefits? BERL’s scope argument seems somewhat irrelevant. Note that BERL says, ” The project brief was not to assess benefits…” But then BERL took a position on them, saying benefits are zero. Regardless of the report’s purpose, such a position seems strange and Matt and Eric try to explain why.
Vote: