Hi Chris,
Thanks for responding, and also for addressing that particular issue,
which is the one I have the most questions about.  Let me see if I
understand the issues.

>Thanks, Jim, for taking the time to post some of this discussion.
>
>You'll be relieved to know that I'm not going to reproduce and comment on
>all of it, just this one point from Lomborg that I think is clearly wrong:
>
>"The $5T is the cost of global warming under BAU. You can only 'do nothing'
>in exactly one way. Therefore there is just one number."
>
>A moment's thought will show that:
>
>If we 'do nothing' about climate change, we do not know exactly how fast
>our economies will grow, we do not know exactly how quickly we will find it
>economic to introduce lower-carbon sources of energy, we do not know
>exactly how much the climate will respond to the extra carbon we put into
>the atmosphere, we do not know exactly how that changed climate will affect
>economic and non-economic activity, and we do not know exactly how the
>valuation of that changed activity will develop.

I had to go back to Lomborg to see exactly where he gets the $5
trillion figure under "business as usual."  In a section titled "Then
what should we do?" (pp. 305-317) he presents the figure as generated
from Nordhaus's and Boyer's (cite below) RICE-99 model.  Lomborg
writes:

        "It is perhaps worth noting that the Kyoto reductions in the
Nordhaus and Boyer model will cause a surprisingly small reduction in
the temperature (0.03 deg. C) in 2100, partly because the developing
countries will increase their CO2 emissions compared to the
business-as-usual model. . . .
        "But what will be the overall cost of these interventions?
Figure 163 [JT: I'll summarize below] shows the cost of the various
types of intervention, with the baseline being a situation with no
global warming.  It shows that the business-as-usual will present
society with a total, one-time cost of $4,820 billion. [footnote
#2610] This cost can also be seen as the cost of the anthropogenic
greenhouse effect as such--if it should prove that our CO2 emissions
have no effect on the climate it would mean a gain of of a little
less than 5 trillion dollars.
        "Of course we would prefer to be without the anthropogenic
greenhouse effect, but the phenomenon is not something we can simply
wish away.  If global warming is coming we must pay the bill--and
then the central question is just how small we can keep this bill.
It turns out, as we have already argued [earlier in the chapter],
that the optimal policy costs a little less.  The total cost in this
case is $4,575 billion, or $245 billion cheaper than taking no
action.  These total savings of $245 billion pertain to a slightly
higher cost in the short term of controlling CO2, offset by the
rather greater advantage of slightly less warming in the long run"
(p. 310).

Figure 163 reads:
Total costs (trillion Y2000$):

Business as usual (no action): $4.820T
Optimal policy: $4.575T
Global stabilization of CO2 at 1990 level: $8.553T
Limit temp. increase to 2.5 deg. C:  $7.803T
Limit temp. increase to 1.5 deg. C: $37.632

Additionally, Lomborg's footnote #2610 reads:
"One should not put too much faith into all these digits, given the
numerous assumptions and approximations--it would be more correct to
say about $5 trillion.  However, the important point is here to
compare the outcome with other scenarios, in which case the relative
costs are much more robust" (p. 426).

Maybe I'm being dense, but I'm still having a bit of trouble seeing
the point to Schneider's criticism here: Lomborg's (actually
Nordhaus's and Boyer's) $5 trillion figure is simply a baseline cost
estimate given present Y2000 conditions of taking no action for
purposes of comparing relative costs of alternative policies.

In my confusion, I've tracked down Nordhaus and Boyer 2000 to see
what it is that they say.  They actually use a round figure of $4
trillion, but I'll assume that Lomborg has legitimate reasons for
coming up with his $4.820 trillion figure (I think Nordhaus may be
using 1990 US dollars).

In a section titled "Costs and Damages," Nordhaus and Boyer write:

        "The focus has been primarily on the abatement costs, but it
is always important to keep in mind that the point of reducing
emissions is to reduce future damages.  Our estimates indicate that
there are likely to be substantial costs of global warming in any of
the cases examined here; the discounted value of damages in the base
case are approximately $4 trillion in present value.
        "The impact of different policies on both costs and damages
is shown in . . . table 8.8.  .  .  .  [These figures] show that the
policies reduce (discounted) damages by only a modest amount--between
$100 and $300 billion out of total damages of $4 trillion.  The
maximum damage reduction from the Kyoto Protocol is $160 billion.
        "Many readers may express surprise about the small impact of
the Kyoto Protocol on the damages from global warming.  The reasons
are that, because there is so much inertia in the climate system and
because the protocol does not limit the emissions of developing
countries, the Kyoto Protocol reduces the global temperature increase
by only a fraction of a degree over the next century.  The other
point .  .  .  is that where- and when-inefficiency raise the costs
of abatement substantially with little or no improvement in benefits.
For example, moving from no controls to the Kyoto Protocol plan with
Annex I [developed countries] trading incurs discounted abatement
cost of $217 billion; however, the discounted value of damages
decreases by only $96 billion.  Moving from the Annex I version to
the no-trade version increases benefits by $65 billion while
increasing costs by $667 billion.  .  .  .
        "The main conclusions that come from an examination of
damages are that there are likely to be substantial damages from
climate change, but that the Kyoto Protocol does virtually nothing to
mitigate the damages" (Nordhaus and Boyer 2000, pp. 163-166).

Jim again: sorry for the length of this, but I guess I'm still a bit
thick on why the $5 trillion base figure is not reputable.  I could
understand if the complaint were that the $5T figure ought to be
proportional to the range of predicted temperature rise (e.g. 1 deg.
C increase=$1T cost; 5 deg. C increase=$5T cost, or whatever), but
that doesn't seem to be Schneider's complaint.

I thought Lomborg's point in using these figures is to demonstrate
relative costs *generally*, as he states in the footnote: "However,
the important point is here to compare the outcome with other
scenarios, in which case the relative costs are much more robust."

I read Lomborg (and Nordhaus) as saying, if we were to do nothing and
policies stay the same, then we could predict up to $5 trillion in
costs associated with global warming.  That figure is to be compared
*generally* with the relative costs of the following alternative
policy scenarios:
Lomborg's Optimal policy: $4.575T
Global stabilization of CO2 at 1990 level: $8.553T
Limit temp. increase to 2.5 deg. C:  $7.803T
Limit temp. increase to 1.5 deg. C: $37.632

Again, I might not be thinking clearly enough here, but it seems to
me that Chris's comment above applies equally to each of the
alternative policy scenarios, i.e.:

>If we 'do nothing' about climate change, we do not know exactly how fast
>our economies will grow, we do not know exactly how quickly we will find it
>economic to introduce lower-carbon sources of energy, we do not know
>exactly how much the climate will respond to the extra carbon we put into
>the atmosphere, we do not know exactly how that changed climate will affect
>economic and non-economic activity, and we do not know exactly how the
>valuation of that changed activity will develop.

Sorry if I'm just being obtuse, but I am grateful at any rate for
your helping me work through these issues and figures.

Jim T.

Nordhaus, William D. and Joseph Boyer.  Warming the World: Economic
Models of Global Warming (Cambridge, MIT Press: 2000).

>
>So any reputable analysis needs to consider the uncertainties, both if we
>'do nothing' and if we 'do something', rather than producing 'just one
>number'. And that is what the reputable analyses do. It seems to me that
>Schneider is right on this fairly central point, and Lomborg, on this
>showing, is wrong.
>
>Chris
>
>
>Chris Hope, Judge Institute of Management,
>University of Cambridge, Cambridge CB2 1AG, UK.
>Voice: +44 1223 338194.   Fax: +44 1223 339701
>e-mail: [log in to unmask]