 |
| 'Mitigation tomorrow and mitigation yesterday - but never mitigation today' Alice and the White Queen, John Tenniel, 1865. |
Utility maximization, perfect competition,
consumer rationality – when reading economic modelling literature, it doesn’t
take long to come across some pretty absurd statements. Because of the
complexity of real-world economic processes, simplification is recognised to be
a fundamental tenet of economic modelling that is necessary in order to
establish and examine specific economic relationships. Such simplification is
achieved by making various assumptions when constructing the model and it is
here that the absurdities start cropping up. It’s important to recognise that
economic models shouldn’t be automatically dismissed because of these
outlandish assumptions. In particular, I believe that abstract models are
useful for producing benchmark results that can be compared with the real
world. For example, examining model predictions that don’t hold in reality and
asking why different outcomes are observed can
highlight particularly important aspects of economic systems. However, economic
models aren’t used solely for academic purposes and often inform policy
decisions, which is when extreme assumptions start to be problematic.
Integrated assessment models (IAMs) are a
particular type of environment-economy model that are often used when assessing
the economic costs of climate change and mitigation policies[1].
Although specific models vary, the general framework combines an economic model
of production with a simple general circulation model. Production is assumed to
have a positive effect on ‘utility’ or wellbeing, as it enables consumption,
but also incurs economic costs or damages by generating greenhouse gas
emissions that cause atmospheric temperature increases. Mitigation is
incorporated by assuming that decision makers can invest in either
production or abatement technology, which has the effect of reducing the level
of emissions per unit of output. Policy makers, therefore, face an
inter-temporal problem: investing in abatement today is beneficial, as it
causes future climate change costs to be avoided, but also undesirable, as
it reduces present-day consumption and hence wellbeing.
From this, it’s pretty clear that IAMs are
gross simplifications of reality, but are these simplifications problematic?
This excellent paper by Ackerman et al (2009)
argues they are, outlining three issues with the IAM framework related to
the practice of discounting, speculative valuation methods and the modelling of
mitigation. Although the first two of these points are significant, I’m going
to focus here on the third, as discounting and valuation are already highly
debated topics and I believe Ackerman et al’s critique of mitigation modelling
relates to topics I have discussed previously.
A popular use of IAMs is to determine an
‘optimal’ path of mitigation that achieves particular emission reduction
targets while minimising the adverse wellbeing impacts of abatement investment and Ackerman et al observe that such modelling exercises often conclude the
‘optimal’ policy involves backloading investment. This observation is supported by a review of IAMs that found ‘perhaps the most surprising result
is the consensus that… modest controls are generally optimal’ (Kelly and Kolstad 1999, pg 19), which directly contradicts statements made by the
scientific community that argue immediate action against climate change is
imperative.
Ackerman et al argue that this result
occurs because the way technological change is modeled in IAMs is inadequate
and causes the costs of mitigation to be overstated. While IAMs acknowledge
that abatement technology will become increasingly efficient over time (meaning
that a given amount of investment in abatement will achieve a greater reduction
in emissions), this technological change is assumed to occur at a constant rate
that can be determined from past data. As the authors state, this awesome
simplification makes the optimality of backloading unsurprising:
‘Because climate change is a long term crisis, and predictable, inexorable technological change will make it easier and cheaper to reduce emissions in the future, it seems better to wait before addressing the problem of climate change’, (Ackerman et al 2009, pg 308).
Such a blithe treatment of efficiency
overlooks a significant and rich literature on innovation that emphasises
technological change is often path dependent and, as I have written about before, can be positively influenced by public policy. If politicians are
presented with the result that the ‘optimal’ policy is to do nothing – or very
little – now, the potential benefits of targeted public investment are unlikely
to be realised.
Although there have been recent attempts to
‘endogenise’ technological progress, which means efficiency is no longer taken
as a constant but is determined within the model, these progressive
developments haven’t been applied to all IAMs. Furthermore, it is often very
hard to establish what assumptions are being made in any particular model as
these are often buried in technical annexes rather than explicitly stated
in the main body of reports and research papers. As long as economists continue
to use IAMs that treat technological change as given and discuss underlying
assumptions evasively, I worry that climate policy will remain an unfulfilled promise
reminiscent of the White Queen’s policy of ‘jam tomorrow’ that Alice found so
frustrating in her adventures through the looking glass…
"The rule is, jam to-morrow and jam yesterday – but never jam to-day."
"It must come sometimes to 'jam to-day'," Alice objected."No, it can't," said the Queen. "It's jam every other day: to-day isn't any other day, you know." (Carroll 1897).
[1] Just to give you an idea of their significance, the whole Stern
Review was based on the PAGE2002 IAM (Hope 2006, Dietz et al 2007)
No comments:
Post a Comment