Memorandum by Ross McKitrick, Associate
Professor of Economics, University of Guelph
INTRODUCTION
I am an Associate Professor of Economics at
the University of Guelph in Ontario, Canada. I specialize in environmental
economics and issues related to climate change. My research is
funded by the federally-funded Social Sciences and Humanities
Research Council of Canada, through peer-reviewed grant competitions.
I am pleased that your Inquiry is taking up
issues related to the IPCC process. I have observed this organization
very closely over the past few years and I believe a critical
outside assessment is overdue. The IPCC exerts tremendous global
influence over energy, environment and climate policies, yet is
effectively unaccountable. They have not won over any of their
prominent critics since the mid-1990s, meanwhile new, credible
experts continue to come forward with doubts about the IPCC's
credibility.
In this submission I would like to explain two
concerns I have regarding the IPCC:
— It appears to have little or no working
relationship with the mainstream academic economics community;
— It has exaggerated the rigor of its
scientific review process.
1. The lack of connection between the IPCC
and the academic economics community
One of the striking differences between the
Second Assessment Report of 1995 and the Third Assessment Report
of 2001 is the loss of participation of mainstream economists
in the latter. A comparison of the lists of chapter contributors
(especially in Working Group II) between the reports will confirm
that the IPCC could no longer claim to have the participation
of mainstream professional economists after 1995.
In recent years some economists have taken greater
notice of the IPCC's work because of the efforts of Ian Castles
and David Henderson to focus expert attention on the "Special
Report on Emission Scenarios" (SRES). This has led to a growing
body of criticism of the IPCC's handling of economic issues. The
SRES does not use conventional economic modeling to produce what
would normally be called "forecasts" or "projections".
They call their outputs "storylines" and "scenarios"
and emphasize that they are speculative, yet at the same time
they market the results as "predictions". For example,
the back cover of the SRES Report states (emphasis added):
The [IPCC] Special Report on Emission Scenarios
describes new scenarios of the future, and predicts
greenhouse gas emissions associated with such developments . . .
The scenarios provide the basis for future assessments of climate
change and possible response strategies.
The list of contributors to the SRES and to
the IPCC (WGII) Report[78]
includes a small and non-representative sample of economists,
amongst a long list of government bureaucrats and academics from
other disciplines. Moreover I know that some of the contributing
economists are quite critical of the final Reports. One of them
is John Reilly of the Massachusetts Institute of Technology. In
an article in Canada's National Post (27 November 2002)
he said that the SRES exercise was "in my view, a kind of
insult to science" and the method was "lunacy".
He said his lab refused a request from the IPCC to let their models
be "tweaked" to support the IPCC scenarios.
In Canada there is a large community of academic
economists, many with an international reputation, working in
the fields of natural resource, energy and environmental economics.
None of the participants in our annual research study group, numbering
close to one hundred members drawn from universities across Canada
and the US, is involved with the IPCC or had any hand in the SRES
Report.
I recently completed a study, [79]coauthored
with Mark Strazicich of Appalachian State University, that confirms
the SRES emission scenarios are unrealistically high. We used
time series econometric methods to analyze data on per capita
carbon dioxide emissions for 121 countries around the world. We
are able to show that the global per capita CO2 emissions level
is a stationary constant (neither drifting nor trending upwards)
with a long term mean of 1.14 tonnes per person and a standard
deviation of 0.02. The mean has not changed for several decades,
and indeed is trending slightly downwards since the early 1980s
(see figure below). If emissions average just over 1.1 tonnes
per person, and population peaks (as expected) at about nine billion
mid-century, we can expect peak emissions of about 10 billion
tonnes by 2050. Yet most IPCC scenarios are between 15 and 30
billion tonnes at 2050, a range that sits well above the plausible
upper bound.
We calculated the implied global per capita
emission levels associated with each of the 40 SRES scenarios
over the next 50 years and computed the probability of observing
each one. Only seven of the 40 SRES scenarios remain within five
standard deviations of the current mean through the year 2050.
Many depart more than 10 standard deviations above the observed
mean; eight lie more than fifty standard deviations above
the observed mean.
Most scenarios are so improbable they should
never have been published in the first place. The seven scenarios
that we found remotely possible imply a range of total global
CO2 emissions from 9.1 to 11.2 Gigatonnes as of 2050, with a mean
of about 10.1 Gigatonnes as of 2050. Yet, as I mentioned, the
bulk of the SRES scenarios imply emissions of 15 Gigatonnes or
more as of 2050.
I should emphasize that what Mark Strazicich
and I did was merely to apply some standard statistical tests
for evaluating economic forecasts. It would have been obvious
to most economists to do so. In presenting it to our colleagues
a typical reaction is surprise that the IPCC didn't check these
things themselves. A recently received comment (from one of the
few academic economists in North America who has studied the SRES
scenarios closely) stated: "the key findings really are important.
Essentially, I think they demolish the SRES exercise—something
that I think was overdue."
The fact that the SRES document was used for
the Third Assessment Report without discovering these (and other)
problems illustrates my first concern about the lack of serious
economics capability in the IPCC milieu.
2. The exaggerated rigour of the IPCC review
process
Of even more concern to me is that, even after
serious flaws in the SRES have come to light, the IPCC has chosen
to use the same scenarios for the Fourth Assessment Report
(AR4), even though it is not due out until 2007. In making this
decision, the IPCC has effectively communicated to the scholarly
community that external criticism will have little impact on its
work. This further diminishes the incentive for outside experts
to bother checking over the IPCC Reports.
Another case I have been involved in recently
illustrates both that experts in the field put little effort into
critically assessing the IPCC's work, and that the IPCC overstates
the quality of its own internal peer review system. In 2003 I
got involved in a project with Stephen McIntyre, a mineral financing
consultant in Toronto, to replicate one of the central aspects
of the IPCC case for global warming, the so-called "hockey
stick" graph[80]
of Mann, Bradley and Hughes. This graph purports to show that
the climate of the late 20th century is unusually warm compared
to the past thousand years. Full details of our work are available
on my web site[81]
and on a new web page set up by Mr. McIntyre (www.climateaudit.org).
The IPCC deliberately highlighted the Mann et
al hockey stick. It appears five times in the TAR, each time
in bright colour and often occupying at least half a page. No
other graph is so prominent. The IPCC Summary for Policymakers
(p 3) used this figure as the basis of its claim that it is likely
"that the 1990s has been the warmest decade and 1998 the
warmest year of the millennium" for the Northern Hemisphere.
The graph was subsequently reprinted countless times and used
by governments around the world.
Had the IPCC actually done the kind of rigorous
review that they boast of, they would have discovered that there
was an error in a routine calculation step (principal component
analysis) that falsely identifies a hockey stick shape as the
dominant pattern in the data. The flawed computer program can
even pull out spurious hockey stick shapes from lists of trendless
random numbers, a point we showed in a recent article in Geophysical
Research Letters. [82]
Further scrutiny of this study has been stymied
by the fact that the authors refuse to divulge most of the computer
code used to produce their results, a situation that the IPCC
never took notice of nor has shown any apparent willingness to
remedy. However our replication of the method is sufficiently
accurate to prove that it fails all basic tests of statistical
significance. Corrected calculations show that the data in question
do not support the claim that the late 20th century climate is
outside the bounds of observed natural variability, opposite to
the assertion by the IPCC. [83]
It was the IPCC that highlighted the hockey
stick data as the canonical representation of the Earth's climate
history. Due to a combination of mathematical error and a dysfunctional
review process they ended up promoting the exact wrong conclusion.
How did they make such a blunder? They cannot claim that since
they surveyed thousands of articles they should not be expected
to scrutinize each one closely. The vast majority of articles
they cite play little role in their main conclusions. The Summary
for Policymakers highlighted three lines of evidence: the hockey
stick, the 20th century surface thermometer record and the climate
model forecasts. Having singled out the hockey stick there is
no excuse for failing to exercise basic due diligence on it.
The hockey stick flew in the face of the more
conventional representation of millennial temperature history,
which suggests conditions in the medieval era were relatively
warm compared to today. Last year Steve McIntyre and I received
the following email from an IPCC expert reviewer.
". . . I was one of a myriad of `reviewers'
of the IPCC 2000, prior to its publication. One of the major concerns
I expressed was the high level of credence given to the Mann et
al temperature history, without it having been seriously subjected
to testing. I strongly recommended that this had some dangerous
implication, should the reliance upon that research prove premature
. . ."
It is ironic that, despite having received such
warnings, it took two people outside the IPCC process to provide
a critical reappraisal of the hockey stick. I think this points
to the need for some external oversight to put proper checks and
balances into place.
3. Recommendations for governments that rely
on the IPCC
In the private sector, no one would invest a
million dollars in a mining project without putting the prospectus
through multiple layers of due diligence, including complete audits
of supporting calculations by independent professionals. Imagine
if your financial advisor proposed putting a large fraction of
your pension funds into stocks of mining companies which have
only released unaudited financial statements and promotional brochures.
No rational investor would consent to this, yet in setting large-scale
climate policy, worth many billions of dollars, this is effectively
what is happening.
The standard should be set very high for the
information used in the policymaking process. I would like to
see the following initiatives considered.
— The economic aspects of the
IPCC work, including its emission scenarios and model runs based
thereupon, should not be used for policymaking purposes until
a representative panel of expert economists has conducted a thorough
critique. Such a panel should consist of experts fully independent
of the IPCC and government environment ministries, and should
include, at minimum, experts in growth theory, measurement theory
and international economics. It should also include representatives
of government finance ministries and statistical agencies.
— A group of experts fully independent
of the IPCC should be assembled immediately after the release
of any future IPCC Reports to identify the key studies on which
the Report's conclusions have been based, and audit those studies,
with a view to verifying that, at a minimum:
— The data are publicly available,
— The statistical methods were fully
described, correctly implemented and the computer code is published,
— If the findings given maximum prominence
are at odds with other published evidence, good reason is provided
in the text as to why these findings have been given prominence.
Any competent scientist can assess these things.
My strong recommendation is that such a panel be drawn from the
ranks of competent mathematicians, statisticians, physicists and
computer scientists outside the climatology profession, to prevent
the conflict of interest that arises because climatologists face
career repercussions from publicly criticizing the IPCC. Also,
participation should exclude officials from environment ministries,
because of the conflict of interest entailed in the fact that
environment ministries are the main financial beneficiaries of
the promotion of global warming fears.
I believe both initiatives are necessary, overdue
and would carry only minimal costs. They ought to be taken by
governments proposing to use the IPCC Reports for policymaking
purposes. For those who feel there is no need for such "audits",
and that implicit trust can be placed in the IPCC not to make
any serious mistakes, my research over the past few years prevents
me from sharing that optimistic assumption. Considering the enormous
costs of climate policy and the large budget allocations at stake,
surely a modest investment in due diligence is warranted.
24 February 2005
78 For the Working Group II list see http://www.grida.no/climate/ipcc_tar/wg2/688.htm Back
79
McKitrick, Ross and Mark C Strazicich (2005). "Stationarity
of Global Per Capita Carbon Dioxide Emissions: Implications for
Global Warming Scenarios" University of Guelph Department
of Economics Discussion Paper 2005-03. (submitted for publication). Back
80
Based on Mann, ME, Bradley, RS and Hughes, MK (1998) Nature,
392, 779-787 and Mann, ME, Bradley, RS and Hughes, MK, (1999).
Geophysical Research Letters, 26, 759-762. Back
81
http://www.uoguelph.ca/~rmckitri/research/trc.html Back
82
McIntyre, S and McKitrick, R Geophysical Research Letters
Vol 32, No 3, L03710 10.1029/2004GL021750 12 February 2005. Back
83
McIntyre, S and McKitrick, R Energy and Environment 16(1)
pp 69-100. Back
| 
