Friday, September 17, 2010

not a triple crisis IMO

update 20th Sept: For the full dialogue b/w me and steve see comments 13, 18, 21, 22, 25, 26, 28, 30, 32, 55, 56 at Can capitalism save the planet?

I left a comment on Steve Keen's blog in response to his recent talk at the Can Capitalism Save the Planet? forum. I only watched the first 9 minute video on his blog (link). I didn't like his triple crisis scenario that economic crisis, peak oil and global warming will combine to create a disaster by mid Century. This is a departure from Steve who up to now has focused on debt deflation and the instability of capitalism developing the ideas first advanced by Hyman Minsky. In response to another complaint in the thread Steve linked to a pdf (A comparison of Limits of Growth with 30 years of reality) by Graham Turner which argues that the Club of Rome Limits to Growth scenarios are being validated.

My comment is #13:
I had a quick look at the Graham Turner paper you linked to. I don’t believe that a computer model at this stage of their development could accurately predict a catastrophe by mid century.

You promoted a triple crisis view in your talk – economic crisis, peak oil and global warming. Each of those issues has its own complexities and specifics. But the computer modelling aspects of the global warming thesis is not its strong point. I read James Hansen’s book and he does not base his dire predictions on computer modelling -he specifically says they are not reliable enough yet – but relies much more on paleo-climate evidence. In light of this how anyone could say that a computer model will predict trends 100 years into the future is beyond me.

You create a bit of dilemma for those who want to discuss this further. Your blog is about debt deflation but in your talk the triple crisis theme was strong. I think the evidence for capitalisms instability is overwhelming but the other issues require extensive discussion in their own right and in how they connect to the economic crisis.

In your response to johnyh I think the issue you are missing is that there is not a linear relationship b/w the science of peak oil and global warming to the policy actions that might be taken in response to that science. The issue is not so much that the science is wrong (although I don’t think there is a consensus on these issues) but that alarmism at the policy level may not be warranted in response. In that respect I would argue that those issues are quite different from the economic crisis
update 18th September: Steve has replied to me as follows, comment #18:
I agree my talk does create a bit of a dilemma for this blog, so I’ll relax my resistance to discussing global warming here for a short while; but I’ll start by putting my position in perspective, in particular about the role of computer modelling here.

We need a bit of a perspective on what that particular computer model–World 3 and its developments that generated the results in Limits to Growth–were actually doing, and what I have come to agree is a fundamental blindspot in the human psyche (I think Sirius here first put this to me), our inability to grasp the impact of exponential trends.

If there is a fixed resource–say land area–and our use of it is growing exponentially and doubles every ten years, and after 100,000 years we have grown to the point where we are consuming 50% of the land, then in 10 years time it will run out.

If we somehow manage to increase the amount of this resource, or say improve our efficiency of use of it by a factor of four, that will buy us another 20 years. A thousand-fold increase in efficiency will buy us an extra century.

So the model was not as such trying to “predict trends 100 years into the future” as to say that IF exponential trends of usage continue THEN given the feedback between exponential trends and fixed inputs, a crisis will occur sooner rather than later. The models also acknowledged that we could perhaps improve our efficiency of usage of fixed resources (though not as a “magic bullet” but as another exponential trend over time), and that if we did and we reduced other factors as well exponentially (pollution) while reduced some pressures to sub-exponential growth (population), we could probably sustain an indefinitely improving standard of living.

That was written in 1972, and almost 40 years later it is manifestly obvious that we haven’t done any of that (save a continuing tapering in population growth which is still nonetheless growing); if anything we’ve increased the intensity of our exponential loads on the planet.

So the models were not so much a prediction as a warning that we had better come to understand the dilemma of exponential growth on a finite planet sooner rather than later. I don’t think there is much doubt that we have failed to do that, and the “climate sceptic” position, though it’s not consciously trying to refute that proposition, is in effect delaying us coming to terms with it.
My reply back to Steve #21:
Thanks for relaxing the guidelines.

I agree that there are always limits to exponential growth and that not everyone understands exponential growth. However, peak oil is not a problem given that we have long term energy alternatives, such as nuclear. The real problem here is the lack of R&D being devoted to energy alternatives. Nuclear would also solve the problem of excessive CO2 entering the atmosphere.

The real problems here are economic (nuclear is still more expensive than fossil fuels) not environmental.

I tend to agree with John McCarthy (progress and its sustainability) that in the case of energy supply the limit is roughly a billion years since nuclear can supply our energy needs for at least that time and in the case of population it “will eventually be limited by a sense of crowdedness rather than by material considerations”


Mark Miller said...

When I was in college in the early 1990s there was a lot more clarity about what stochastic computer models for non-linear systems were actually good for. What's disturbed me in recent years, with the climate issue, and now with economic issues, is the revelation (to me) that somehow people have gotten it into their heads that these models, which run on computers that are only really good at running linear models, can actually be relied upon to make accurate predictions. It makes no sense at all, except from the POV of the naive believing whatever a computer tells them, a problem with the human perception of computing that goes back decades.

As I think I mentioned on your blog before, one of the key revelations of this last economic crisis was when Alan Greenspan said that a computer model assured him and his people that there was no problem with the housing market, because it assumed that housing prices would *always* rise. Recently it was revealed that the Obama Administration relied on a computer model to make their prediction that the $800 billion stimulus bill that was passed in early 2009 would cap unemployment at 8%, and that unemployment would rise to 10% without it. We're now close to 10% with the stimulus spending, and now all the Obama Admin. can say is, "It would've been worse without it." Well, based on what model? Is it any more reliable than the last model they used? I heard commentary on this a couple nights ago, saying that what the model didn't anticipate was the level of fear that would cause organizations to withhold spending, despite the influx of cash. Well, of course. We're talking about a psychological phenomenon here. Do we have any reliable computer models for mass psychology?

I get the impression that otherwise intelligent people are fooling themselves with computers, because they're satisfied with the illusion of certainty that this medium provides in lieu of a real understanding of what's going on, which may provide the uncomfortable answer of, "We don't know yet, and we don't know when we will know."

Bill Kerr said...

hi mark,

Steve has clarified his position further and I have replied at his blog. I've also updated this blog to include those comments.

Bill Kerr said...

Further discussion at Steve's blog so best to go there for updates: Can Capitalism save the Planet?