Climate Policy under Sustainable Discounted Utilitarianism

Table 2. Change in expected stationary equivalent
of 2 CO2 and 1.5 CO2 policies compared with
business as usual, according to SDU and DU.

aversion is set to two. For these (and all subsequent) calculations, we use the full
modelling horizon from 2005 to 2395.

The table contains our core result, showing that willingness to pay for emissions
abatement is significantly larger under SDU than under DU. For the 2 CO2 policy,
the expected increase in the stationary equivalent is 4.7% under SDU, over twentyfold
higher than the corresponding estimate of 0.2% under DU. For the 1.5 CO2 policy,
the expected increase is 3.83% under SDU, but -0.67% under DU. Intriguingly, this
policy reduces social welfare according to DU, but according to SDU it increases it.
These results follow directly from the finding, detailed in Figure 1, that consumption
is more likely to fall under business as usual than under either of the two abatement
policies. SDU places greater value on these policies than DU as a consequence: they
are more likely to guarantee sustainability, defined as non-decreasing wellbeing.

What Table 2 also shows is the influence of uncertainty, specifically the small
number of random draws in which climate damage is severe. This is evident in
comparing the expected change in the stationary equivalent with the 95^th percentile
change under SDU. For both policies, the expected change is in fact greater than the
95^th percentile, indicating that a few random draws (less than 5%) have a change
in the stationary equivalent so large as to drive the expectation above the 95^th
percentile. This is one way of showing that concerns about intergenerational equity
and concerns about uncertainty are closely linked in the context of climate change.

Figure 2 explores the sensitivity of the expected change in the stationary equiva-
lent as estimated under both SDU and DU to ρ. We examine values for ρ ∈ (0, 0.05)
(corresponding to a range for the decadal discount factor of 1-0.62). It is evident
that, in line with the distribution of near-term abatement costs and longer-term ben-
efits, the expected change in the stationary equivalent of the two abatement policies
is a decreasing function of ρ, both under SDU and under DU. Indeed, it falls rapidly
as ρ is initially increased from 0.

19

<<   <   18   19   20   21   22   23   24   >   >>