Professor Socolow describes the problem in a very clear way: “We
desperately want there to be low-cost solutions to climate change. So,
each time a "solution" arrives that looks like it is low cost, we
embrace it and are not adequately critical.”  Nevertheless, this
problem is not moral hazard; I would argue it is Time Inconsistency.
Time inconsistency is a term first introduced to talk about the
inability of the federal bank to be consistent across time with their
monetary policies, and it is now widely used by behavioral economists
to represent a situation in which a presentself cannot negotiate with
a futureself to ensure the actions decided today are still taken in
the future.  For example, you can always say you will stop smoking in
one week, or start running tomorrow, or dieting in the new year.
However when the next week, the next day or the new year arrive, your
promises are gone.  Several mechanisms give rise to this situation,
hyperbolic discounting being the most standard way of thinking about
the problem nowadays. Other mechanisms such as the overestimation of
the gains from the introduction of a technology, after the technology
has been introduced, will also give rise to this type of behavior.
The problem with situations exhibiting time inconsistent behavior, in
very general terms, is that it is impossible to generate a contract in
which the two parties agree on doing or not doing something.  The
reason this contract do not exist is because ...well, the futureself
or is not there to sign it.  The solution is then to generate path
dependent outcomes.  That is, create policies that will stick in and
that are very difficult to override. IN the case of the diet this
implies registering to a (expensive) program, in the case of running
the next day committing to run with another person (hopefully someone
you respect a lot) and in the case of smoking, I am not sure, I
haven’t figure that one out yet.
Assuming this framework is the right one, one could identify two
problems with arguing climate engineering induces moral hazard
behavior.  First, it obscures our understanding of the problem.
Second, it complicates the finding of a solution.  I should say here
that the fact that the problem is not correctly framed doesn’t mean it
doesn’t exists, as it was very subtly suggested before in this e-mail
exchange.   The fact that it obscures our understanding of the problem
is evident from this discussion and I shall not go there.
The real problem is when it interferes with finding a solution.  A way
to deal with moral hazard is by splitting the benefits of loses
between the principal and the agent (pardon my economics): In the
insurance industry this is due by means of deductibles; that is,
making the agent share on the damages.  In the labour market by
sharing on the benefits of the successful job.  It should be clear
these solutions are not available for the case of climate
engineering.  Precisely because the current generation cannot credibly
commit to split the gains or losses of a successful or unsuccessful
Climate Engineering adventure. Why? Well, the current generation will
not exist once the benefits or losses are weighted in.  IN this sense,
there isn’t a Principal-Agent framework to deal with.
On the contrary, the current generation can create policies today that
will stick in the future.  One such policy has been previously
suggested in the literature and consists on a ban on CE research.  If
there is no research, the technology will not be available and the
possibility of time inconsistent behavior is eliminated at its root.
This solution, however, seems to be inappropriate because the inertia
and uncertainty of the climate system my force the future generation
to use CE, even if it hasn’t been research on it.
A second solution, that is described on one of my papers jointly
written with Timo Goeschl and Daniel Heyen, is to increase our levels
of mitigation today.  By doing so, we decrease the risk for the future
generation; hence, decreasing the need for the use of climate
engineering technologies. This is solution is path dependent and will
seclude the use of CE only for extreme climate sensitivity cases. In
this case, a policy in which both, research on CE and higher levels of
mitigation seems to be adequate to deal with the problem.  Notice that
this solution can only be suggested and defended once we are aware of
the problem and we have frame it appropriately.
In general, I believe when David Keith make the parallel with Moral
Hazard he was thinking of suboptimal behavior induced by the
possibility of CE.  Time inconsistency is one of many other sources of
suboptimal behavior.  However, we need to quantify these suboptimal
behaviors and put them in context with the benefits from CE. But
again, we can only do that once the suboptimal behavior is properly
addressed.
Best,
Juan Moreno-Cruz

-- 
You received this message because you are subscribed to the Google Groups 
"geoengineering" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/geoengineering?hl=en.

Reply via email to