A scare story in the Oz today about the drastic impacts of an emissions trading scheme, with the impacts “so severe it cannot be reliably predicted by existing computer models”.
The problem is described a little more accurately, if less sensationally, a bit further on. Essentially, the scale of the changes required for the 90% cut in emissions by 2050 is large enough that the existing models can’t cope. Hence Professor Warwick McKibbin argues “That’s why you design policy with all sorts of safety valves and a capacity to adapt over time if it turns out the models were completely wrong.”
I’m no economist, let alone an econometrician, but I do know a teency bit about modelling. And this strikes me as exaggerated fearmongering by people who’ve been running similar arguments for a long time.
Every mathematical model ever constructed abstracts away irrelevant details, and many of them use approximations that are only valid over a certain range. To use a very basic example from physics, when modelling both very large and very small systems, Newtonian mechanics starts failing as an accurate predictive tool. But at a terrestrial scale, with familiar mass, velocities, and accelerations, its inaccuracies are typically orders of magnitude less than we can measure. Econometric models undoubtedly do the same thing. In a technical sense (and this is mildly informed guesswork here), they may well be using linear, or quadratic, approximations for certain things in their model (linear equations are straight lines, quadratic ones are the u-shaped curves from high school), that give sensible answers over a limited range, but start giving you garbage when extended further.
More fundamentally, the problem is that you simply can’t get to 90% cuts by linear extrapolations of our existing energy architectures. We can’t conserve our way to 90% cuts (not without what would indeed be fairly drastic changes to our lifestyles). We can’t rejigger our mixture of fossil fuel generators (the kind of thing that the model probably covers fairly accurately) to achieve the cuts. Existing renewables technologies can’t, on their own, replace a big enough fraction of the grid to get us there. So it’s not surprising that the models break down.
It’s important to remember here that Prof. McKibbin has been talking about the uncertainties of the cost of carbon reduction for many a year now; his big idea in the area is the McKibbin-Wilcoxen emissions trading proposal, where, in a nutshell, the price of emissions are heavily restricted, and if the costs of abatement turn out to be much higher than expected, more greenhouse gases get released into the environment. If your primary worry is the costs of reducing emissions (instead of the real issue, the costs of not reducing emissions), McKibbin’s proposed system appeals. So, maybe I’m a horribly cynical bugger, but it’s hardly likely that McKibbinis going to downplay the unpredictability of costs under a more straightforward cap-and-trade system like Garnaut seems to be leaning towards.
In any case, if the models break down, how can a responsible government possibly commit to such targets when they don’t know how much it’s going to cost us? Simple. For a start, the first thing to realize is that by 2050 it is highly likely - barring a disaster even greater in magnitude than the period 1930-1945 - we will be much, much richer then than we are now. Real GDP per capita has continued to improve by a percentage point or two every year since the Industrial Revolution, and shows no sign whatsoever of slowing down. As this paper suggests, a reasonable guess is that it’s going to be around $80,000 per person by then, more than double what it is today. So, even if the cost of emissions reductions is 20% of GDP by 2050, we’ll still be far better off, in material terms, than we are today. And, for the overwhelming majority of Australians, is our present standard of living really all that bad?
The second point to keep in mind is that when the highly precise modelling fails, there’s always the back of the envelope. And any way you crunch the numbers, you can set upper bounds on the cost of decarbonizing the economy - if worst comes to worst, we can artificially remove CO2 from the atmosphere and use the low-risk mineral sequestration to turn it back into a stable solid - essentially, carbonate rock. Even if you add considerable extra margins to the cost claims floating around, it’s still a cost which still represents a pretty small fraction of projected GDP.
Finally, we’ve currently undergoing a great unplanned experiment in what a radical increase in energy costs will do to our economy. As Quiggin noted here, even a carbon price of $100 per tonne of CO2 represents an increase in petrol costs of around 25 cents per litre. Petrol has gone up by at least 60 cents a litre since 2005. So, at least in the transport sector, we’ve seen roughly the equivalent of a $240 carbon price, imposed in three years. Yes, it’s clearly causing some angst, particularly in the airline industry. It’s also probably contributed a bit to a more general economic slowdown until some adaptation occurs. But where’s the disaster?
I can only conclude that this is more scaremongering from the coal industry, not to mention the coal-fired generation industry. By 2050 we will all be vastly materially better off - unless, of course there’s a climate catastrophe. So let’s hope that the Rudd government grits its teeth, ignores the rent-seekers, and gets on with the job of Australia doing its bit to reduce the risk of that catastrophe.
Er, Peak Oil?
Yes, a carbon price much much higher than anyone is seriously considering has had very little impact on demand. Ok, there is some sign of total vehicle miles travelled levelling off in the U.S. but we’re not seeing massive reductions, and I dare say if oil prices take a breather we’ll see VMT creep up again.
Well, we all know who Matthew Warren really works for.
Robert Merkel wrote:
Or we run out of affordable oil, in which case the carbon emissions will drop off by themselves due to massively decreased economic activity
Economics seems to be sliding back into bunkum. If a scientist ran into problems like that with a model, they go and make a new model, not complain about the old one not working. The economists need to get out of the idiocy of “being an austrian” or whatever excuse they use to disguise their political leanings and start replicating the hard science of physics or biology like everybody else does - through experiments. Not models.
Look at the period 1900-1950. We saw two devastating world wars, the Great Depression, the Spanish flu epidemic (amongst others)…and living standards still improved greatly in the developed world over that period.
Furthermore, there world can and has reduced its oil consumption in response to past oil crises. I’ve no doubt that we can do so again, without descending back into the dark ages.
So I’m prepared to make an inflation-adjusted $500 bet with you - our per-capita real GDP, as measured by the ABS or its successor, will be bigger in 2030 than it is now.
Want to take it?
I would expect a lot more gloom and doom concerning all aspects relating to climate change policies. By which I mean policies that seek to answer the problem will be depicted as negatives in some way or another, cost too much for example, as the entrenched vested interests pull a full strings around the place in hopes of delaying or even averting the necessary changes.
I wouldn’t expect the polluters to be front and centre but to raise the issue via the back door, tame experts, tame think tanks, friendly journos and so on.
Is this such an example?
Of course! *smacks forehead* Why has nobody thought of that before? Economics should be a hard science. It’s so easy! Especially since humans are so like atoms in their predictability and consistency!
Furthermore, has anyone told the chemists, physicists and biologists tying up the world’s supercomputing resources that they’re wasting their time with this modeling nonsense? It’s a slide back to bunkum, dammit!
May I just interject…Austrian economics is proved by the fact that if you cut taxes then overall tax receipts do rise…slower of course but eventually they do ‘grow the pie’ overall. So the difference between Austrian economics and trad economics is…?
Then there’s Marxist economics. These are proved by actually existing Marxist states that manage to murder millions. They’re embodied in states like Zimbabwe today.
Liberal economics may be as dead as the Dodo but somehow we still manage to save enough on the micro-economic roundabouts to make up for the swings. I think we might federate EU style - and even switch to the last decent currency standing - even if there is every chance that we will muddle through somehow. Lucky countries’R’ us.
Lots of other ways a model an go haywire than just the breaking down of quadratic or linear approximations, robert.
My guess is it may be to do with actual numbers needed for elasticities, way outside the current range of data; and uncertainty about consumer and industry substitutions in reaction to sudden price changes - which goes to Jacques’ point about human behavior being more complex and multifaceted than that of molecules or photons
Why not take their word for it? They looked at the model outputs and discovered they were strange or inconsistent or ridiculous. At least they were able to check!
Robert, I think you misunderstood the point I was trying to make.
My point is: if a very large carbon price has such little impact on demand, how much effect will a very small carbon price (imposed by the ETS) have?
Lets face it, the ETS is going to be half-ar*ed and full of holes. Can you really see Courageous Kev whacking us with $100/tonne? He’d kill the ETS in a heartbeat if it actually hurt.
BTW, I don’t doubt that the rising oil price will eventually affect demand — it has to — and I appreciate that it takes time for the vehicle fleet to turn over and for other adjustments to be made.
Richer by 2030? Depends if there’s a peak in oil production between now and then, and whether viable alternatives are developed, and I think that’s almost unknowable.
Robert #3
I’m inclined, subject to some nuances and rules about enforceability, to take on your $500 bet. Please email me at paul textpert com and we can set the terms.
Cheers
McKibbin, McKibbin. Name rings a bell. Hang on, isn’t he one of the blokes John Howard appointed to the board of the Reserve bank?
Along with a bunch of fellas [and one woman] from big businesses like Coke, Nufarm, Telstra, Bluescope Steel, James Hardie, Brambles, and Woodside Petroleum.
None of which have anything to do with carbon emissions of course.
A perverse effect of the increased petrol price has this year been an increase in the number of V8 motor vehicles sold.
I can’t find the article now, but there was a news article last week about V8’s being sold at an increased rate as men decide they want to own one before it just gets too expensive.
It also says many V8 owners drive below average km per year, (and that Hybrid and Hatch drivers put out more carbon because they drive a lot more kms than V8 owners).
Strange days indeed.
in any case, if you read Andrew Leigh’s blog, for instance, empirical economics is precisely what he specializes in.
A further point: the “hard sciences” do spend a hell of a lot of time simulating stuff on supercomputers these days. Whether you can rely on these models (for instance, global climate models), and how much, is always an interesting question. In the case of global climate modelling, the answer is probably “given the amount of effort put into them, quite a lot”.
#3 Robert merkel
I hope someone takes your bet, Robert. That’s easy money unless the inmates stay in charge of the asylum for too long.
Looks like the price addition for standard petrol of pricing CO2 at about $25 per ton is an additional 20-21 cents per litre. Before GST, of course.
Lot of money for ‘working families’ to cough up to solve a problem that does not have the good grace to actually exist.
Lots of interesting stuff too from the meetings “The Minister for Making the Sun Obey Krudd” has been having with the evil baseload boys and demon-spawn hydrocarbon companies. Apparently she’s thick as two short planks and thinks that money grows on trees.
For example, poor Penny thinks that thermal power stations and offshore platforms did not actually cost anything to build and operate! Who knew?
Apparently they grow overnight after one sprinkles pixie dust on the ground (or sea floor).
The ALP’s loopy plans pandering to the greenie fruitcakes are going to devalue all thermal power stations by many billions. The Glorious Minister for Making the Sun Obey Krudd apparently believes that this does not require government compensation for destroying the value of their assets! I seem to recall a section of the Constitution which has something to say about that…
Let me see… how long will power companies stay in business when they are forced to operate at a loss, from a ruined asset base ? Probably not long. Wonder who they will be forced to pass the costs on to? Hope Victorians like the idea of having their power bills quadruple in order to keep their brown coal thermals going.
Or the companies could just go bankrupt and close them down, and Melbourne could return to horse transport, kerosene lighting and no refrigeration.
We live in highly entertaining times. Rather soon, there is going to be an amusing sound as brainless greenie dreams meet the onrushing train of reality. One cannot run baseload without coal, oil, gas or nuclear when one has exhausted all the hydro options. Simple. Anyone but a greenie or a leftist intellectual can understand this.
I look forward to the loud ’splat’.
MarkL
Canberra
MarkL: Methinks you might be having unit problems.
The permits are per tonne of CO2 (technically “CO2 equivalent”, an accounting measure that takes into account the relative power of greenhouse gases like methane), not carbon. When you use that figure, a $25 carbon price equates to a price impost on petrol of about 6.25 cents per litre.
As far as quadrupling electricity prices, you do realize that wholesale baseload electricity tends to go for about 4 cents per kwh, whereas retail goes for triple that? Quadrupling the wholesale price (highly unlikely, for various reasons) would only double the price for consumers.
I’ll stay out of the compensation issue, except to note that the “compensation” in the EU has turned out to be a massive rort that has made the coal generations billions in windfall profits.
Good post Robert.
The most noticeable error in Mathew Warren’s article is probably what is causing the issues with the models. I suspect it is not so much the amount of emissions as it is the fact that they are being reduced in 2050. Economic models are uncertain enough as it is, but the idea of quantitatively modeling the economy in 2050 has huge uncertainties.
I do find it curious that Warwick McKibbin believes we should have a safety valve price cap on emissions in case they are more expensive than expected, but does not seem interested in a price floor. The issue of climate change is rife with uncertainties - as well as the uncertainty in abatement costs there is also the uncertainty in climate sensitivity, carbon cycle feedbacks, climate change impacts including whether Greenland will melt and so on. Because of the existence of backstop technologies, such as solar power and CCS, there is already a natural cap on the price of climate change abatement that is something like $400/t CO2-e. Unfortunately there is no cap on the costs of climate change - it’s impacts have “potentially unlimited downside exposure”. Some recent work from Weitzman suggests this means that low probability high impact effects of climate change dominate its expect costs.
There is also the possibility of ‘unknown unknowns’ - impacts of climate change that we have barely thought about, let alone attempted to quantify in an integrated assessment model. When it comes to abatement the role of the ‘unknown unknowns’ is reversed - the possibility of abatement technologies that we have not thought about yet could lead to the cost of abatement being lower than expected. So I find it peculiar that McKibbin is so concerned with abatement costing more than expected when both common sense and more complicated forms of cost-benefit analysis that deal with risk and uncertainty suggest that the risk of climate change being worse than expected is much more serious than any uncertainties in the cost of climate change mitigation.
A very interesting set of slides by Michael Hanemann describes some of the uncertainties associated with modelling the impacts of climate change and why the actual impacts could be worse than these ‘Integrated Assessment Models’ predict.
Jacques Chester wrote:
There are plenty of hard sciences that include unpredictability - that’s what statistics are for. Besides which, if humans are that unpredictable, why are so many economists happy to be associated with political leanings first and mathematical and investigative skills second? Andrew Leigh might be (somewhat) of an exception but the world is filled with lightweights like Sinclair Davidson and or heavyweights like Paul Krugman, both sides making conclusions first and doing their studies second. Roberts post is a perfect example.
Robert Merkel wrote:
How are we going to organise this bet Robert? Do you have an escrow in mind?
I’d like to increase (your) odds - I’ll bet it’ll be lower in 2020, 2030 and possibly even 2040.
Yes, they do, but they also have to follow it up with repeatable experiments or observations that fit those models, and make testable predictions. Economists seem to forget that last step, by and large.
I, too, would be interested in taking your bet, Robert, except that it’s quite likely I’ll no longer be alive in 2030. I suppose, in that case, my by then middle-aged children could collect my winnings.
The reason I’m confident it’d be a winning bet, btw, is because I don’t expect that any government will actually do anything meaningful about global warming until it’s much too late (if then). This http://www.abc.net.au/rn/bigideas/stories/2008/2257930.htm is germane.
OK, several people have wanted to take the bet. I’m prepared to put up a stake of something around the equivalent of $500 in total.
If anybody else is interested in taking this bet, let me know (either replying on here or by email), and we’ll send some group emails round to set the exact terms, which we’ll then publicize on LP.
That’s a very long-timeframe bet. If punters were willing to stick their $500 in an escrowed account you could make a lot of interest on it, your assumptions holding true, Rob.
“but I do know a teency bit about modelling”
I admire modesty, Robert
David, wtf are you talking about ’start’ replicating the hard sciences? They’ve been doing experiments for ages eg. http://youtube.com/watch?v=ZRrPcSNeKzk Which isn’t to take anything away from the, at times fanatical, ideological force of Walrasian general equilibrium amongst the dismal scientists…
dk.au wrote:
Ages? They might have started in the 1970s but it’s not exactly a common mode of operation. “Dismal Science” is far too kind a description - the use of scientific tools doesn’t make it science any more than telescopes make astrology science. Economics in 2008 is about where physics and chemistry were in the age of alchemy.
Why not run the bet through the Long Now Foundation? (http://www.longbets.org/)
d
If you’ve got the capital (and don’t mind waiting a long time for the collect) then I’d advise you to go hard on this one, Robert - it’s candy from a baby. By 2030, Australia’s GDP per capita will be at least 25% higher than it is today (in real terms). And that’s a very pessimistic forecast. More likely that it will be at least 50% higher, and probably more.
If I wasn’t trying to save for a deposit, I’d stump up a few bucks to help you in the process of parting these doom-saying fools from their money. It’s not often that somebody will offer you evens for a sure thing!
Although Matthew Warren’s article in The Rentseekers Review is misleading in many ways it does have some interesting pieces of gossip.
“They simply don’t believe us,” - its about bloody time that people stopped believing the scaremongering from greenhouse polluters. It reminds me of the story of the boy who cried wolf…
Considering it took physics and chemistry millenia to get as far as alchemy, I’d say that’s a pretty good rate of progress.
Whiggish much?
In the sense that experiments help to achieve closure amongst a community of experts, Economics is every bit as scientific as every other discipline.
” Economics is every bit as scientific as every other discipline.”
…as long as all other things are equal…
Jacques Chester wrote:
Wake me up when they quit searching for the Philosophers stone equivalent (i.e. the ultimate statistical proof of correctness of political leaning). Then it might start being a useful tool rather than a blunt object to bludgeon your opponent with.
MarkL: Methinks you might be having unit problems.
The permits are per tonne of CO2 (technically “CO2 equivalent”, an accounting measure that takes into account the relative power of greenhouse gases like methane), not carbon. When you use that figure, a $25 carbon price equates to a price impost on petrol of about 6.25 cents per litre.
Comment: I thought so too at the time when doing back-of-the-fag-packet calculations. So I asked the interlocutor, who looked at me, smiled, and said ‘apply it as a VAT, not as a point of retail sale tax’. I have no knowledge of whatever policy is planned in regards to the new tax, but I hope his hint is wrong.
As far as quadrupling electricity prices, you do realize that wholesale baseload electricity tends to go for about 4 cents per kwh, whereas retail goes for triple that? Quadrupling the wholesale price (highly unlikely, for various reasons) would only double the price for consumers.
Comment: My understanding is that if no means of compensation for lost asset value is forthcoming, the asset value losses will have to be recovered from consumers, hence the information. Now, I have no idea as to the scale of asset loss (that would be commercially sensitive), but the very silence on that issue is telling, as is the ‘difficult’ nature of meetings with the Minister for Making the Sun Obey Krudd. At this stage, my assumption is that those meetings are in the ’cause and effect’ stage, with the Minister for Making the Sun Obey Krudd being given a fairly brutal education in real-world impacts of greenie dreams. That would apply (based on my stated assumption) that this issue is being used as a stick to beat some sense into her head.
The horsetrading would be fascinating to see, not that any of us will, because as Gordon Brown is finding out in the UK, ‘working families’ are averse to paying for the dreams of climate change fantasists.
I’ll stay out of the compensation issue, except to note that the “compensation” in the EU has turned out to be a massive rort that has made the coal generations billions in windfall profits.
Comment: Most interesting, and an issue I have no visibility of. Do you have any source data or links?
MarkL
canberra
MarkL - Re #14, #32:
Hepburn et al. Auctioning of EU ETS phase II allowances: how and why? is a good place to start. Auctioning 100% of emission permits to any greenhouse gas intensive firm will almost certainly result in a net profit for the firm.
I’m not an expert in constitutional law but I don’t think that every time a government makes a decision that adversely affects a firm they are eligible for compensation. Should governments compensate cigarette companies for regulations that affect their profits? When asbestos was made illegal because it was carcinogenic, should governments have compensated the firms that made the asbestos for lost profits? When I invest in shares, I am likely to make more of a profit than if I was to invest in cash, because I am making a risk. This is what investors in greenhouse gas intensive industries are doing. If greenhouse gas intensive industries were to be compensated for a carbon price then it would set a very strange precedent. A “section of the constitution” may mean that a firm is eligible for compensation if its assets were acquired by the state, but that is not what is happening here. As long as a carbon price in not in existence, greenhouse gas intensive firms are being subsidised by future generations for the climate change costs that will occur, I think torts law may have something to say about this.
Power companies will pass the cost of emissions on to consumers regardless of whether they are compensated for the polluter finally having to pay. It is the marginal cost that makes a difference. I find it amusing that they are still trying to use this argument. Garnaut is certainly not buying it.
I could make more comments but I prefer not to provide free advice to rent seekers. If I was to offer any advice to greenhouse gas intensive firms, I would suggest that they should reduce their emissions, rather than try to postpone action on climate change or shift the burden of a carbon price elsewhere - which is what they are trying to do when lobbying for free emission permits.