John Quiggin Scores Again
Robert Waldmann
I am very glad that John Quiggin has found a group even less worthy or respect than us macroeconomists. He writes
It seemed for a little while as if the delusionists had scored another win, when Phil Jones, the scientist who has been most viciously target by the hackers/harassers gave an honest answer to a deliberately loaded question prepared by them and put to him in a BBC interview
BBC: Do you agree that from 1995 to the present there has been no statistically-significant global warming?
[kip]
Looking at the responses of ’sceptics’ to this episode we can distinguish four or five sets (depending on your views about set theory)
[skip]
5. Those genuine sceptics who pointed out the dishonesty of the claim, and called out those on their own side of the debate who promoted it. Obviously, members of this set deserve some serious respect and attention in the future. Unfortunately, the intersection between this set and the set of “sceptics” in the currently prevailing sense appears to be the empty set[2]
That’s 5 sets Professor Quiggin. It’s also statistically significant evidence that global warming skeptics are, on average, either more clueless, more dishonest or both than the general population. Also the data set does not contain statistically significant evidence against the hypothesis that all global warming skeptics are incapable of understanding basic statistics, recklessly irresponsible and total liars.
Quiggin assumes that most are only one or two of these things, but there is no evidence for his belief in the data set he analyses.
Thin comfort for macroeconomics. What Quiggin has shown is that climate science is science, just as he has shown that macroeconomic science is not science in “Zombie Economics.”
“…just as he has shown that macroeconomic science is not science in “Zombie Economics.””
Um, no. He has shown that many mainstream practitioners of macroeconomics endulged in mistaken assumptions, group think and the like. That is not evidence that macro is not science. I understand there is an eagerness to make this claim, but that’s very sort of eagerness to make claims that led macroeconomists down the wrong path.
In general, it’s best to try to say things that are true, as well as personally gratifying.
First of all, the link to what Quiggin actually said http://johnquiggin.com/index.php/archives/2010/03/01/four-lies-and-an-empty-set/ At least I hope that is the correct link.
As far as I can see, Quiggin restricted himself to a very narrow field — a single BBC interview and a resulting Daily Mail article from the same vein of yellow journalism that we Americans are exposed to daily on all subjects from Fox News. Unless I am missing something, he did not take any pot shots at macroeconomics in this article, although if I recall correctly he is not especially a fan of current economic thought. His position seems to be that taking a simplistic economic system that we can understand then trying to fudge it toward reality isn’t working. See http://crookedtimber.org/2009/11/19/bookblogging-what-next-for-macroeconomics/
I think that Quiggin is right that the handling of the story here is preposterous, but no more so than the constant drumbeat of climate change prose that is IMO equally ill found. Torturing data until it confesses works about as well as torturing prisoners. It’ll tell you whatever you want to hear.
I don’t know why so many eminent economists who don’t actually know anything much about climate are so anxious to believe one side of the story and to disparage anyone who points out that most of the claims made by “climate scientists” are not themselves especially credible. It’s not like their data is pristine, their statistics unambiguous, or any of their models have ever actually been validated. And it’s never been that big a secret that their spokesorganization the IPCC is highly politicized and does a mediocre at best job of peer review. All that can be determined in a few hours of Wikipedia reading and Googling.
Codger, couldn’t have said it better.
Robert, JQ sets up a personal and unsupported premise then argues on that premise. Which part of that kind of “arguing with yourself” debate do you find convincing?
There is MUCH, much, even more growing evidence regarding the quality and the handling of the actual temperature data sets. Are you aware that the UK Met has admitted there may be problems with their historical temperature calculations and are proposing a complete do over. One that will use independent approaches.
Furthermore are you aware that much of the raw data is being reviewed in a very detailed fashion? This set of efforts should yield an understanding of how in the future to better process these data.
I will commend you on using the more correct term, skeptic, versus the pejorative of denialist.
This argument is very nearly over. Once the EPA ruling is fought and EPA loses in the courts there won’t be any more stakes to put in the heart of this misplaced issue. But do not despair, the next issue is being forwarded as we speak. The only remaining issue in that is which gets the most traction. Then we can repeat this unending cycle of misinformation and hyperbolic reporting.
VtCodger,
I don’t know why so many eminent economists who don’t actually know anything much about climate are so anxious to believe one side of the story and to disparage anyone who points out that most of the claims made by “climate scientists” are not themselves especially credible.
As far as I know, the only economist that has entered into the “science” part of the climate science debate is McKitrick, and he is on the skeptic side of the fence. And I’m sorry, but his econometrics is mediocre at best. He made a valid criticism concerning an error in the weighting of variances in a Principal Components Analysis model…and that’s fair enough…but then he made an even more egregious error when he “corrected” the data set. For the most part economists believe in the division of labor and assume for the sake of economic analysis that the consensus view on global warming is right, and then economists try to estimate the effect on GDP and discounted utility. What most economists disparage is the goofy idea that global warming is a good thing. What economists bring to the debate is the right way to think about risk under conditions of extreme uncertainty.
But Quiggin has a point here. When Phil Jones was talking about statistical significance he had a very specific meaning. The Daily Mail ran a stupid story and all of the usual suspects (e.g., WSJ) picked up on the story. I think this is another example of intelligent but malevolent thinkers on the right basically manipulating the stupid teabagger crowd. Look at those two pictures of teabagger rallies that I posted yesterday and ask yourself if anyone in that picture understands what is meant by statistical significance. That’s why the deafening silence from McKitrick and McIntyre. They played folks on their side for fools.
CoRev,
The intellectual argument is very nearly over. And your side lost. And the proof is in the fact that on this site you have been all over the ballpark as to whether or not the earth is warming. At times you have agreed that the earth is warming, albeit due to natural causes. Okay, but if that’s the case, then isn’t the debate about the accuracy of temperature data irrelevant? What are you trying to tell us? That the earth is warming but if you correct for data errors the earth is cooling??? That seems to be the two stories you’re trying to sell at the same time. Sorry, it doesn’t fly.
The UK Met has agreed to redo their calculations, but contrary to your claim the UK Met has not admitted that there might be problems with their historical data. The UK Met has said that after the recalculations they are quite confident that their basic results will stand. If there are any differences, those differences will be at the margin.
CoRev,
Go read what the UK Met said. The UK Met was very clear. They said that they believed the recalculation would sustain their original claim, but wanted to go through the exercise in order to satisfy critics. So when the UK Met comes back with similar numbers, what will be your excuse?
I noticed that you tried to ignore my observation that you’ve been all over the ballpark with inconsistent claims. In this case simultaneously claiming that the earth is warming and that data showing the earth is warming is wrong because it’s really cooling.
2slugs, you’re not serious? You just restated my opinions then argued on that restatement. Wait! I just said that’s what JQ did. Wow, have I found a trend?
Lessee, the UK met decides to recalc and it’s NOT because they admit there might be errors? OK, only in 2lugs-land that makes sense.
Those are definitely worth a SHEESH!
CoRev,
Lessee, the UK met decides to recalc and it’s NOT because they admit there might be errors? OK, only in 2lugs-land that makes sense.
The Met Office is confident that its analysis will eventually be shown to be correct. However, it says it wants to create a new and fully open method of analysing temperature data.
http://www.timesonline.co.uk/tol/news/environment/article6945445.ece
You lose. Again.
2slugs, I have read the UK Met announcement. It will be interesting after those food fights are finished to see the highs and lows in the trend lines. Will they correspond with the current set? Dunno. Depends on how they process the data. What if they pick a different base period? All bets are off on the anomalies. I presume even you will agree with that?
As to my views, I have stated them several times on this blog and once even in response to you. Yours on the other hand were just intriguigingly ignorant. Your views were we should stabilize today’s climate? If we did is this the optimum? If we could and this is NOT the optimum, why are you recommending it?
***What most economists disparage is the goofy idea that global warming is a good thing. What economists bring to the debate is the right way to think about risk under conditions of extreme uncertainty.***
No, I don’t think so. First of all, there is nothing especially goofy about the idea that global warming could be a good thing. If, for example, it stabilized rainfall patterns — and it could — that would be quite a good thing as dams and aquaducts would be built where they were needed and would be properly sized, we’d give up agriculture in areas where it is doomed to fail, etc. Is global warming going to do that? Who the hell knows? Certainly not either the economists or the climate scientists. It would be a good thing if the Sahara bloomed again. Not so good a thing if the rains quit coming to the American midwest. And who cares if Barstow gets 4 inches of rain a year or three? Neither is enough to grow anything but sage, 16 kinds of cactus, mesquite, gorgeous wildflowers that last about three weeks and the biggest jackrabbits you can possibly imagine.
Second, the thesis that econmists can properly assess risk certainly seems suspect in view of events of the past few years. The profession did not exactly cover itself with glory in this debacle.
And finally, while I agree with Quiggin, why is no one pointing out loudly and repeatedly that the underlying temperature data that is being wrangled over almost certainly contains a significant heat island bias that is building over time as humans alter the environment around the measuring points? That in itself could easily account for much or of all the observed temperature rise, significant or not. Here’s picture of a particularly egregious example, the weather station at Rome’s Clampino airport. Note that the weather gear is surrounded by tarmac and parking areas that surely weren’t there when the station was put into operation. And it’s hard to believe that jet exhaust streams are going to do anything other than bias the measured temperatures upward. http://wattsupwiththat.com/2009/03/28/how-not-to-measure-temperature-part-86-when-in-rome-dont-do-as-the-romans-do/
Maybe the bias isn’t large? One could hope that. Surely they either correct the data (how?) or don’t use it to compute global temperatures? Well, no — as far as I can tell, they use it as is. This is an outlier. Most data is much better? It’s probably an outlier. But folks that are seeking out weather measuring sites and assessing them say that most are probably not the same as they were when they were built and that in almost all cases the environment modifications will bias the measured temperature in the same direction … up. And keep in mind that we are worrying about small fractions of a degree here, so even small biases are likely to be significant.
The following is for Robert and 2slugs. Reading 2slugs’ above Met PR we see them wishing to create a new and open approach for calculating the average temps. One of the major issues in calculating them is the current adjustments for Urban Heat Island (UHI) effects. Some even conjecture that much of the observed temp increase is due to UHI alone.
Dr Roy Spencer of the Univ. of AL, Huntsville (UAH),. alos one of the key contributors to the satellite temp calculations, has started a new study analyzing US temps related to population densities (a proxy for UHI.) See here:
Spencer: Direct Evidence that Most U.S. Warming Since 1973 Could Be Spurious
His preliminary findings: “This is a very significant result. It suggests the possibility that there has been essentially no warming in the U.S. since the 1970s.
…
Taken together, I believe these results provide powerful and direct evidence that the GHCN data still has a substantial spurious warming component, at least for the period (since 1973) and region (U.S.) addressed here.
There is a clear need for new, independent analyses of the global temperature data…the raw data, that is. As I have mentioned before, we need independent groups doing new and independent global temperature analyses — not international committees of Nobel laureates passing down opinions on tablets of stone. (Spencer’s emphasis)“
So a new and open approach with better adjustments is needed. Will it show the temp trends to be significantly different? Dunno. It will surely add some level of confidence to the reults from this truly horrid raw data.
Robert, Spencer does compare his results with the CRUTEM (Jones’ data set estimates). Does this prove anything about Jones and Quiggin’s defense of his actions? Nope, but it does show that the defense was not needed (or seriously needed depending on your view point), as Jones has already shown, via his emails, his processes and actions to be less than scientifically ethical.
Robert,
Your rigorous truth in reporting standard is admirable. However, it seems rather selective. Where is the outrage over the data-challenged hyperbolic pro-AGW alarmism that has so scared 2slugbaits? If skeptics have misrepresented Jones’ quote, what about Al Gore’s misrepresentation of facts? http://scienceandpublicpolicy.org/monckton/goreerrors.html
VtCodger,
And keep in mind that we are worrying about small fractions of a degree here, so even small biases are likely to be significant.
We might be talking about small fractions of a degree as long as all we’re talking about is CO2 concentrations below 400ppm and as long as trapped methane remains trapped. But that’s not what most economists are talking about. They use economic models to show that small increases in temperatures, which are something that we can understand intuitively, generally have negative effects globally. Sure, some areas do well, but most do not. We can measure the effects on exports. We can measure the effects of warm nights on crop yields. We have very good data on income levels and temperature with diverse instrumental variables so that we can isolate how much of the change in income is due to temperature rather than some other factor. And there are plenty of studies that show the overall effect of global warming is mildly negative as long as warming is contained to something like 2.5 degrees C. But when you start getting much above 2.5 degrees C, then things really start to go to hell fast. At 5 degrees C and we’re not just talking about stabilizing rainfall, we’re talking about alligators swimming in the Arctic (they did the last time temps reached that level). At that point the economic destruction and dislocation simply swamps any plausible claim of global warming being a good thing. Now there is a remote chance that some on this blog might see a 2.5 degree warming. Very little chance that readers today will see a 5 degree warming. So almost all of the bad news about global warming is stuff far out in the future…like the 22nd century. The problem is that CO2 is very slow to decay, so our failure to severely constrict CO2 emissions means that there is a very decent chance of catastrophic temperatures one or two centuries out. Remember, it takes hundreds of years to see any significant decay in CO2 even if we cut CO2 emissions to zero right now. It’s when you start thinking in terms of long time lags and in terms of catastrophic events that are simply orders of magnitude beyond our experience, that’s when you need to rely on uncertainty models that correctly gauge risk.
As to the recent experience of economists assessing risks, you are forgetting that it was economists who were warning that many of the financial models were relying upon “most likely” events and were ignoring “fat tails”. That is exactly the mistake that CoRev is making. When you are dealing with extremen uncertainty you have to quit thinking in terms of “most likely” and start worrying about plausible “worst case.”
CoRev,
Your views were we should stabilize today’s climate? If we did is this the optimum? If we could and this is NOT the optimum, why are you recommending it?
The welfare losses of assuming global warming is true when it is actually false are positive but relatively small. Think in terms of Type 1 errors and Type 2 errors.
http://en.wikipedia.org/wiki/Type_I_and_type_II_errors
As Wiki vividly describes it, a Type 1 error is an error of “excess credulity.” In this case it would be an error due to my wrongly believing that global warming is true when in fact it’s actually false. In your case we risk committing a Type 2 error, which Wiki describes as an error of “excess skepticism.” In this case we assume global warming is a hoax when it’s actually true. In this cas a Type 1 error is far and away the more serious because a false alarm can always be corrected at a later date. A Type 2 error is irreversible.
VtCodger,
Second, the thesis that econmists can properly assess risk certainly seems suspect in view of events of the past few years. The profession did not exactly cover itself with glory in this debacle.
I should also point out that the same Chicago School style economists who were assuring us that there was no cause for alarm about housing bubbles and wild financial instruments were also the same economists who are the most skeptics of global warming. It was left wing economists who were sounding the alarms about the risks of investment bankers runamok and it’s those same left wing economists who are most vocal about the “fat tail” risks of global warming.
2lugs, “key word” plausible.
Read my latrest on Dr Roy spencer’s current efforts. Or go to EM Smiths/Chiefio’s blog and look at his latest. http://chiefio.wordpress.com/
also, 2slugs, saying this: “When you are dealing with extremen uncertainty you have to quit thinking in terms of “most likely” and start worrying about plausible “worst case.”” doesn’t make a whole lot of sense. Why, because you are imputing certainty where little exists. Worst case in an uncertain scenario means there is much less likelihood any event may happen let alone the worst case.
You’re reaching because you believe the worst case is even plausible. Do a quick search and see if and when the temp were 2.5 to 5C higher. Better still, do a search and see where the temps are that much higher than your current locale. Then tell us just how bad those temps will be for you and your family.
I’m going to guess you will find your equivalent locale somewhere south let’s say Atlanta, Tallahassee, or Tampa. Scary!!!!
CoRev,
You’re reaching because you believe the worst case is even plausible.
An event with a 1 in 20 chance of happening is plausible.
Do a quick search and see if and when the temp were 2.5 to 5C higher.
There is a huge difference between 2.5 degrees warming and 5 degrees warming. The last time the earth was 5 degrees warmer was thirty to fifty million years ago, when, as I said, alligators were swimming in the around the north pole.
Then tell us just how bad those temps will be for you and your family.
This question shows what’s wrong with your analysis. You are very myopic. Global warming will, in all likelihood, have very little effect on my personally, perhaps some minor inconvenience for my kids, major headaches for their kids, and calamity for their kids. I’ve noticed this in a lot of what you write about. Your concerns are very near term. You have an extraordinarily high discount factor.
I’m going to guess you will find your equivalent locale somewhere south let’s say Atlanta, Tallahassee, or Tampa. Scary!!!!
If that’s what you think global warming is all about…turning up the AC a little higher and milder winters, then you really don’t have a clue.
CoRev,
Maybe you and Roy Spencer ought to take a second look at what his analysis actually shows. Setting aside some of the methodological problems with his analysis and for the moment just accepting his results, the results do not support the conclusion that there hasn’t been global warming. Pay careful attention to the words…he suggests the possibility that a substantial component of the warming trend may be “spurious” due to heat island effects. At most this means there is a quibble over the rate of temperature increase.
As to the methodology, there are some pretty obvious problems. First, he is only looking at the US, not global temps. A few very cold years recently can leverage things quite a bit. Second, his weighting technique tends to put too much weight on daytime temps in cities (note the times of his readings), which is not a particularly good way to do things. Night time temps give a much better reading, so it’s no wonder that he finds evidence of urban heating. Third, his assigning a single population density based on 2000 throughout the entire period is dicey and is a pretty obvious source of bias. Fourth, you cannot properly do a regression if the variance of the different groups is different. And unless he had exactly the same number of observations in each group the sample variance is guaranteed to be different. It leads to inefficient estimates. He might have corrected for this problem using some weighted least squares technique, but he makes no mention of it. Finally, he didn’t really try to tease out the true deterministic trend component from the random walk component. It looks like all he did was use Excel to find some trend line and then he called it a day’s work. He is right in not wanting to put this analysis up for peer review. There are a lot of other problems with his analysis, but these ought to be enough. And these are the obvious ones.
2slugs said: “An event with a 1 in 20 chance of happening is plausible. ”
and then this:
“The last time the earth was 5 degrees warmer was thirty to fifty million years ago, when, as I said, alligators were swimming in the around the north pole. ‘
Yup, “A ONE IN 20 EVENT” is plausible. So which event is that? So, how long ago was that again? 35M YA? I think I’ll warn the grandkids re: the alligators. As I said, you are reaching.
If you can conflate a geological and climatic event (alligators in the Arctic Ocean) with a possible calamitous event or condition within three to four generations really does indicate how far you must expand reality to reach a level of self proclaimed fear. You really do not understand a planet that is 2.5 to 5C higher in temps.
And all of this is based upon a bunch of short term (three generations) of very, very shoddy raw tremperature data, that is just now actually being analyzed. Sheesh!!!!!
2slugs, there you go again. Rephrasing an issue into your own issues, then arguing that issue. You said: “…do not support the conclusion that there hasn’t been global warming. ” So, who has said there is NO global warming? Not Spencer. NOPE not me! My views are well documented on this blog.
But, what is really funny is your now admitting: “At most this means there is a quibble over the rate of temperature increase. ” So, are you coming around to the skeptical view that it’s NOT ALL ABOUT THE CO2???? Then we agree.
Your discussion of Spencer’s process is just more 2slugs BS. You come across as knowing how to process this horrid, horrid, data, but after being offered the chance you agree that your approach(s) won’t add value (Jeff Id’s discusssion). Furthermore, you fault Spencer for not having perfect data or using the approriate tools, but he is working with imperfect, no horrid, data and a well established tool set. (BTW, Tamino has had two post on it not being a random walk. So, that comment makes no sense.)
I’m sure that Dr. spencer would listen to another statistician on how to improve his approach. Informal peer review is working. If you are serious about improving the processes, then do the hard work or at least make the offer to help those who are. Otherwise, it appears you are playing big frog in the little AB pond.
Recently you claimed I was all over the place, but comparing a proxy based temperature from a period of 35-50M years to a short, short, short 130 year snapshot of crappy real data is the epitome of hubris or ignorance. Making claims of some calamitous condition equal to that 35-50M YA period which was obviously not calamitous (at least for some animals still alive today). Leaves us with the obvious observation, if alligators were possible then what’s to say we are not? We do coexist today.
Slugs,
It’s all moot. Obama is going to emerge from the healthcare debate regardless if he wins or loses with an empty tank. Global warming legislation is another casualty of Ahab Obama’s healthcare obsession. But that’s a good thing.
CoRev,
Yup, “A ONE IN 20 EVENT” is plausible. So which event is that? So, how long ago was that again? 35M YA? I think I’ll warn the grandkids re: the alligators.
Let me help you out. The last time the earth was 5 degrees C warmer was 30-50 million years ago. The IPCC estimated ~ a 1 in 20 chance of that happening in the next hundred years. So you better warn the grandkids. And MIT just found some errors in the IPCC study (shocking!) and now estimates the most likely temperature increase at 5 degrees in 100 years.
You really do not understand a planet that is 2.5 to 5C higher in temps.
We can understand a planet that is 2.5 degrees warmer. We have no concept of a planet that is 5 degrees warmer. That is completely outside of all human…in fact, virtually all mammalian experience.
And all of this is based upon a bunch of short term (three generations) of very, very shoddy raw tremperature data, that is just now actually being analyzed.
No. Most of it is based on estimates of CO2 and CH4 growth over the next 100 years and estimates of GHG effects on temperatures. Very little of it is based on extrapolating a time series of historical temperatures. The historical temperature data is important for calibrating models and parameters, but it is useless as a pure time series exercise. Climate scientists are not time series analysts.
CoRev,
after being offered the chance you agree that your approach(s) won’t add value (Jeff Id’s discusssion).
Huh? Geoff (the Aussie) begged off and said that his data set had too many problems and that he would send it when it was cleaned up and when he settled some personal issues that needed his attention.
Furthermore, you fault Spencer for not having perfect data or using the approriate tools, but he is working with imperfect, no horrid, data and a well established tool set.
No, I said that the econometrics was tricky and you really needed something like a Heckit or Tobit model. And the creators of both those models won Nobel prizes. And both of those models were designed to handle exactly the problem of censored data. I’m assuming you know what the term “censored” means in econometrics.
Tamino has had two post on it not being a random walk
I didn’t say it was. I said that Dr. Spencer didn’t give us any evidence that he had tested for a stochastic trend rather than a deterministic trend.
Cantab,
That may be true. And there’s no shortage of problems out there that need attention. Obama is also going to have to turn his attention to immigration reform. My sense is that Obama’s position on immigration is pretty close to George Bush’s and Bush didn’t have much success with the Buchanan wing of the GOP. So who knows. But global warming isn’t going to go away. Kicking the can down the road just makes it more expensive for future generations. My great-great-grandkids can blame CoRev.
I’m sorry 2slugs, you are just getting too desperate. appeal to authority, IPCC and MIT, just don’t cut it since Climategate.
Estimates based upon “…estimates of CO2 and CH4 growth…” and “Very little of it is based on extrapolating a time series of historical temperatures.” I’m sitting here just chortiling. I guess you are trying to say that the GHG you cite are not correlated against a”time series of temps.” Uh, huh. Yup, that’s why those GHGs are so dangerous. They will raise the, umh, ehh, erh, ah, yes, the alligator count.
I know you’re just joking around now. That was actually funny.
You’ve been doing the misdirection trick so much it’s getting tiresome. I said Jeff, and you say Geoff. Geoff ?Sherrington? is just a commenter. Jeff id offered you an opportunity to submit an article or apply your approach. QWhy not actually trying your vaunted approa(es) to add to the knowledge? Sitting on the sidlines sniping adds no value, and reflects poorly on the sniper.
Econometrics? I thought we were talking climate science.
As far as the models and censored data, you are just blowing smoke. Let me refer you to the Jeff Id issue and what appears to be your unwillingness to follow through.
It really does appear that you are not serious about anything you say here. Your comments are getting more and more desperate and bizarre.