7th May 2016
We hear a lot about climate change from many bad economists. It is time to hear the views of one of the greatest economists of all time and a winner of the 2002 Nobel Memorial Prize in Economic Sciences. These comments (below) are originally from 2012, but they make perfect sense even today. [I’ve slightly amended formatting for ease of reading and added a few annotations including colour, bolding and underlining. My commentary in blue.] I’ve taken the comments from Smith’s FB page:
==VERNON SMITH’S COMMENT==
Here is the long version of my specific comments on global climate change (2012). Its happening, and we should prepare for it, but stopping it is way beyond current technology, which is why CC [Copenhagen Consensus] plugged for energy saving and CO2 sequestration research with over a dozen WW [Sanjeev: worldwide?] problems easily besting climate given the budget we worked with. Since this is 2012 it would need updating, but you get the idea. You may have to click for the charts. [Sanjeev: I’ve pasted the charts here]
Comment : Vernon L. Smith
Carbon (dioxide) emission mitigation.
This is the solution that I and the panel rated lowest. I begin with this option because reducing carbon emissions is widely perceived by politicians, journalists, and many scientists (although skeptics abound) as (1) necessary to reduce global climate change; (2) worth the cost.
On (2) my view, given the state of current knowledge, is that the cost in sacrificed human betterment and poverty reduction would be prohibitive in achieving reduced near-term effective atmospheric carbon inventories (new emissions have an uncertain half-life estimate of 40 or many more years.
On (1) in my view there is so much uncertainty in the relevant sciences that we cannot yet make that judgment. That there is climate change in recent history (see temperature charts below) is not at issue, but rather its cause and reversibility through reduced carbon emissions.
• The leading scientific hypothesis is that global climate change is due to anthropogenic carbon emission forcing, but other hypotheses are not dead, e.g., solar forcing in combination with complex surface-air interactions and lags.
• But forcing is recognized as being much modified by physical principles operating through endogenous feedback loops—some recognized, others in discovery process—that are poorly understood in systems as complex as the global land-ocean-lower-upper atmospheric interactions. We cannot know, only estimate, what might be the climate state if carbon forcing were absent.
• Since 1850, emissions have grown exponentially, and therefore, if indeed anthropogenic emissions are the cause of recent warming, we have no observational experience with how emission reversals, might asymmetrically map into temperature change (+, 0, −) over short or long intervals.
• Climate change has both benefits and costs, with perhaps the largest cost being that of adaptation. Humans have thrived during Pleistocene cooling, and have thrived during Holocene warming. That adaptation is likely to continue as our knowledge base grows beyond anything that is conceivable across century-length episodes (compare Einstein’s year, 1905, with 2005)?
Uncertain causes of global climate change and the carbon “balance.”
Our knowledge of the dynamics of global temperature derives in part from well known experimentally verified spectral properties of carbon (a trace gas in the atmosphere, 0.036% by volume) that enables one to estimate the marginal isolated contribution to surface temperature of the net incoming radiative energy from the sun that is due to the carbon concentrations in the atmosphere.
The estimated equilibrium contribution of carbon to earth surface temperature symmetrically increases and decreases in the logarithm of its relative concentration in parts per million by volume.
How that marginal contribution interacts through dynamic lagged positive and negative feedbacks in the complex air-ocean-land-forest system is governed by hypotheses of great logical and quantitative uncertainty and subject to constant new scientific discover, subject to error. Error includes both model specification error and measurement error conditional on the specifications.
Comparable error properties pervade other complex systems like those in ecology, culture and economy. For the economy, in spite of well developed models of economic and financial interactions, sometimes even tested against samples of data not used to calibrate them, monetary policy utterly failed to anticipate that a bursting of the housing-mortgage market bubble beginning in 2006 would engulf the international economy by 2008. The chain of events was inherently unpredictable in a manner that policy actions could anticipate and prevent. Comparable uncertainty issues plague climate complexity.
I begin this brief summary with a myopic view—myopic relative to the totality of temperature observations begging for better understanding—from −400 to 2000 AD, with Roman and other historic events flagged in Figure 1.
Figure 1. Climate of the last 2400 years
[Source: Greenland ice core charts from here]
The deviations, inferred from ice core data, are plotted relative to the mean over this period. Recent temperatures have been trending up since 1900, and have exceeded the 2400 year mean only since about 1950. Anthropogenic carbon is not usually implicated in warming before about 1850.
Note also how common are “natural” (non-anthropogenic) “tipping points” when temperature signatures suddenly change direction especially at extreme points outside the range ±0.5 C. Although temperatures appear to have become more volatile since 1850 (this could be a data splicing problem), rapid 50-year upward trends are unremarkable over the whole 2400 years (e.g., 3rd, 5th ,9th, 11th,13th centuries). Always keep in mind that these indirect (ice core) measures of temperature are subject to error as are any “direct” observations of mean global temperature..
In Figure 2 is displayed a longer view (the Holocene includes the post agricultural revolution since about 9000 BC), with the last 2400 years revealed to be one of relative temperature decline.
Figure 2 Climate of the last 12,000 years.
(Reference cited in Figure 1)
A global warming skeptic might easily argue that warming in the last century is merely part of a 500 year return to average temperatures prevailing since before the beginning of agriculture. But uncertainty cuts both ways, and we are well advised to be skeptical of the skeptics.
Also evident is the sharp 6 C increase in temperature in the two millennia just prior to agriculture. This warming was part of a 120 meter (four hundred feet) increase in sea level, beginning 18,000BC in which, for example, the Persian Gulf was filled to its present level, submerging the river of four heads from Mesopotamia.
Ice core data records a small dip in carbon from 9000BC to about 4500BC after which it increased, possibly due to anthropogenic forest clearing. This hypothesis has just been reported to be “confidently rejected” in favor of natural causes through the study of carbon isotope signatures. (See J. Elsig, et al, 2009,“Stable Isotope Constraints in Holocene Carbon Cycle…” Nature, 461, pp 446, 507-510.)
Looking only at the recent industrial era, as in Figure 3, warming is both prominent and coincides with accelerating growth in anthropogenic carbon emissions. [Sanjeev: It only seems to “coincide”. If one goes back in time in the geological record, there is absolutely no correlation between temperature and CO2 emissions. When I started studying this issue in 2008, this was the first question I asked: show me the very long term correlation.]
Figure 3 Climate since 1880
But is it a causal relationship? The greatest error source is concerned with how the climate system responds to the flow of new carbon into the far larger global inventory of carbon: Does the response decrease warming (negative feedback) or increase it (positive feedback), with what lags, and how does the response change over time? What are in short supply are observational tests of the predictive implications of the models for any relevant data not used in their calibration.
Because of the higher quality data base for calibrating the simulation models, modelers have focused on the period since 1975, leaving insufficient data for out-of-sample (subsets of data not used in model calibration) tests of models validity. As time goes on, however, and the data accumulate this problem is diminished. For example, concerning recent decadal relative cooling trends we have:
“Observations indicate that global temperature rise has slowed in the last decade…much less than the 0.18°C decade–1 recorded between 1979 and 2005… This is despite a steady increase in radiative forcing as a result of human activities and has led some to question climate predictions of substantial twenty-first century warming…Near-zero and even negative trends are common for intervals of a decade or less in the simulations, due to the model’s internal climate variability. The simulations rule out (at the 95% level) zero trends for intervals of 15 yr or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy with the expected present-day warming rate.” (Do Global Temperature Trends Over the Last Decade Falsify Climate predictions [in “State of the Climate in 2008”] . Bull. Amer. Meteor. Soc., 90 (8), S22-23.)
The quotation shows a refreshing commitment in advance to observations that would constitute falsification of a model. But this should not be comforting, as the complexity problem is emphatically not one of just measuring the effect of an equilibrium disturbance in the carbon heat balance, as the entire system already has a complex dynamic motion quite apart from anthropogenic sources, as is evident in Figures 1 and 2.
Calculating confidence intervals for deviations from unknown sources of trend is a deep mathematical challenge. (See Wu, Zhaohua et al, 2007, On the trend, detrending, and variability of nonlinear and nonstationary time series. Proc Natl Acad Sci, 104, pp 14889–14894).
My purpose is not to detract from the enormous recent advances in climate science, but to emphasize that our ignorance of global dynamics continues to be overwhelming.
Cut Black Carbon.
I rated this fairly high essentially because of the recent scientific claims that these particulate emissions may account for much of lower atmospheric temperature increases and particularly the regional warming associated with loss of Arctic and glacial ice. This may turn out to be a promising break-through, or just one more dead end, but it is worth aggressive investigation. Since the Asian stove sources are also a health hazard, black carbon merits cutting in any case; the principle problem has been to implement a change in remote stove use against powerful cultural norms.
On recent issues in black carbon (soot) and related brown cloud science:
“Here we use three lightweight unmanned aerial vehicles that were vertically stacked…over the polluted Indian Ocean…(that)… deployed miniaturized instruments measuring aerosol concentrations, soot amount and solar fluxes….(making)… it possible to measure the atmospheric solar heating rates directly. We found that atmospheric brown clouds enhanced lower atmospheric solar heating by about 50 per cent…. brown clouds contribute as much as the recent increase in anthropogenic greenhouse gases to regional lower atmospheric warming trends. We propose that the combined warming trend of 0.25 K per decade may be sufficient to account for the observed retreat of the Himalayan glaciers.” V. Ramanathan et al (2007) “Warming trends in Asia amplified by brown cloud solar absorption.”Nature 448, 575-578
Also see V. Ramanathan & G. Carmichael (2008), “Global and regional climate changes due to black carbon.” Nature Geoscience 1, 221 – 227; and J. R. McConnel, et al (2007) “20th-Century industrial Black Carbon Emissions Altered Artic Climate Fording.” Science. 317, pp 1381-1384.
“We conclude that decreasing concentrations of sulphate aerosols and increasing concentrations of black carbon have substantially contributed to rapid Arctic warming during the past three decades”. Drew Shindell and Greg Faluvegi (2009). Climate Response to Regional Radiative Forcing During the 20th Century. Nature Geoscience, 2, pp. 294–300.
I rated this solution very high. Regardless of the causes of climate change, the trend in global warming, sea level rise and loss of glacial and ocean ice for the last 20,000 years is most likely to continue. This is shown in Figure 4 for the virtually monotone increasing sea level rise in the Persian Gulf long before anthropogenic causes could be implicated.
Figure 4. Sea Level Rise in the Persian Gulf.
[Sanjeev: the reference below, and title of the chart don’t seem to coincide, but the point being considered is the same]
MWP refers to various meltwater pulses or “sudden” rises of 10 meters (33 feet) or more in a few hundred years. The most recent was MWP-1C, ~8,200-7,600 years ago. See at http://www.giss.nasa.gov/research/briefs/gornitz_09/
If carbon is a principal new cause, its accumulated effects are thought to be already built in and irreversible short of an unanticipated natural reverse “tipping.” The failure to respond efficaciously to hurricane Katrina shows clearly the need to ask whether and in what way adaptive planning can be implemented. We need also to ask if New Orleans or other cities located below sea level should be protected, rebuilt if lost, or simply moved with migration assistance.
Adaptation also makes sense because in intervals of tens of thousands of years the ice core temperature record going back 420,000 shows that warm episodes have been rare and short-lived—a few thousand years—with carbon concentration lagging temperatures.
I rated research on cloud whitening highest, aerosol insertion lower.
Cloud whitening is scalable, subject to relatively controlled experiments and reversible so far as we know. It appears therefore to chart an incremental low cost learning path in which unintended consequences can be identified on a small scale before applying it more aggressively to counteract anticipated damages from warming.
Aerosol insertion is less attractive on these measures since it is less incrementally controlled, but research seems justified because of the prospect that it could act more quickly than carbon mitigation. Even cloud whitening, however, is fraught with incredible uncertainties that are just elementary reflections of our broader scientific ignorance: “Despite decades of research, it has proved frustratingly difficult to establish climatically meaningful relationships among the aerosol, clouds and precipitation.” (B. Stevens and G. Feingold, 2009, “Untangling aerosol effects on clouds and precipitation in a buffered system.” Nature, 461: p 607.)
These options have merit only because they offer promising new increases in our practical knowledge, not because they can be assured of rescuing us if that be necessary. Nevertheless, both of these technologies may have risks, whose origins are precisely the same as those governing the causes of global warming: we know precious little about systems as complex as that of the global climate, and we should proceed with caution to avoid unintended harm.
In line with the previous CC meeting conclusions, I am persuaded that if target anthropogenic GHG reductions are necessary to reduce global climate change—a distinctly speculative proposition—then the brute force approach with existing technology is not feasible.
If there is any effective means of reducing GHG emissions, it rests with R&D discoveries that will enormously increase energy savings (or, alternatively, finesse the whole issue through climate engineering as above). But we cannot assure discovery; we can only commit to trying, and the technical paper cautiously recognizes this potential outcome.
Even in the absence of a carbon tax and public R&D it is easy to underestimate the extent to which rising relative energy prices for long periods will induce innovations that will increase energy efficiency, as is evident by simply looking back to 1830 when kerosene-from-coal—or “coal oil”—was the response to the high price of whale oil.
My response to Fred Folavary who suggested we have a green tax.
Vernon L. Smith had also noted this article on his page a few days ago: http://www.telegraph.co.uk/news/2016/05/05/no-one-ever-says-it-but-in-many-ways-global-warming-will-be-a-go/ [Smith’s comment was: “Green on the increase? Why do climate Chicken Littles not put more emphasis on adaptation, and preparation for further warming?”]
– an analysis of the benefits of CO2. What if the benefits exceed costs? (which, in my view – since 2008 – they do. There are plenty of studies which show that there is a net benefit of CO2 to life on earth for at least the next few decades.)
As plants have increased, so has animal life on earth. There is clear evidence that current levels of CO2 are at least 10 times less than peak levels on earth – which caused the huge book in plant life that kickstarted evolution.
We should carefully consider why we want to tax something which benefits life on earth.
Further, almost all green technology is extravagantly expensive, and requires taxpayer subsidy.
A green tax therefore (a) reduces the benefits of CO2 and (b) destroys precious resources which would could have gone into useful production.
Finally, it is only a matter of a couple of decades (at most) when carbon will become one of the most outdated resources. Solar and fusion technologies are on the verge of becoming cost-effective. Coal and oil will die out on its own, anyway.
Our best bet is to let markets do what they do best, without taxing them or subsidising them unnecessarily.