Compliance Technology and Self-Enforcing Agreements: Bård Harstad Lecture Notes - March 2016
Compliance Technology and Self-Enforcing Agreements: Bård Harstad Lecture Notes - March 2016
Self-Enforcing Agreements
Bård Harstad
Lecture Notes - March 2016
Abstract
These notes analyze a repeated game in which countries are polluting
as well as investing in technologies. While folk theorems point out that
the …rst best can be sustained as a subgame-perfect equilibrium when the
players are su¢ ciently patient, we derive the second best equilibrium when
they are not. This equilibrium is distorted in that countries over-invest
in technologies that are “green" (i.e., strategic substitutes for polluting)
but under-invest in adaptation and “brown" technologies (i.e., strategic
complements to polluting). It is in particular countries which are small or
bene…t little from cooperation that will be required to strategically invest
in this way. With imperfect monitoring or uncertainty, such strategic
investments reduce the need for a long, costly punishment phase and the
probability that punishment will be triggered.
1 Introduction
The purpose of this paper is to analyze how the society can succeed with climate
treaties of the type signed in Kyoto, 1997, and in Paris, 2015. To be successful, in
our view, any treaty must address two major challenges of climate change. First,
in the absence of international enforcement bodies, any international treaty
must be self-enforcing. In principle, sanctions could be imposed by threatening
free-riders with trade barriers, the seizure of infrastructure, or armed con‡icts,
but such options are never on the table when climate negotiators meet. In the
absence of such sanctions, one might hope that countries would follow the treaty
in order to motivate other nations to cooperate in the future. This motivation,
however, may not always be su¢ ciently strong. For example, for many years
These lecture notes are based on an article with the same name which is writtten jointly
by Bård Harstad, Francesco Lancia, and Alessia Russo.
1
it was clear that Canada would not meet its commitments under the Kyoto
Protocol. So, in 2011, it simply withdrew.
The second challenge to confront climate change is to develop new and envi-
ronmentally friendly technology. The importance of new and green technology
is recognized in the treaties, but there has been no attempt to negotiate or
quantify how much countries should be required to invest in these technolo-
gies.1 Instead, the negotiators focus on quantifying emissions or abatements
and leave the investment decision to individual countries. Nevertheless, some
countries do invest heavily in green technologies. The European Union aims for
20 percent of its energy to come from renewable sources by 2020, and to increase
that number to 27 percent by 2030. China is a still larger investor in renewable
energy and has invested heavily in wind energy and solar technology. Other
countries have instead invested in so-called “brown” technology: Canada, for
example, has developed its capacity to extract unconventional oil such as tar-
sands while, at the same time, “Canada risks being left behind as green energy
takes o¤" (The Globe and Mail, September 21st, 2009).
The interaction between the two challenges is poorly understood by econo-
mists as well as policymakers. To understand how treaties can address these
challenges and how these challenges interact, we need a theory that allows tech-
nology investments as well as emission decisions to be made repeatedly. Since
the treaty must be self-enforcing, strategies must constitute a subgame-perfect
equilibrium (SPE).
There is no such theory in the literature, however, and many important
questions have thus not been addressed. First, what is the best (i.e., Pareto
optimal) SPE? Second, folk theorems have emphasized that even the …rst best
can be sustained if the players are su¢ ciently patient, but what distortions occur
if they are not? Third, will non-cooperative, sel…sh investments result in the
optimal level of environmentally friendly technologies? Or are there reasons,
beyond the traditional argument about technological spillovers, for including
technology investments in the agreement? Which kinds of countries ought to
invest the most, and in what kinds of technology?
To address these questions, we analyze a repeated extensive form game
where, in every period, countries can invest in technology before deciding on
emission levels. In the basic model, all decisions are observable and investments
are sel…sh (i.e., there are no technological spillovers). Consequently, equilib-
rium investments would have been …rst best if the countries had committed to
the emission levels. The …rst best can also be achieved if the discount factor
is su¢ ciently high, in line with standard folk theorems. For smaller discount
factors, however, the best SPE must be distorted. We show that the distor-
tions take the form of over-investments in so-called “green” technologies, i.e.,
renewable energy or abatement technologies that can substitute for pollution.
1 Chapter 16 of the Stern Review (2007) identi…ed technology-based schemes as an indis-
pensable strategy for tackling climate change. However, article 114 of the Cancun Agreement
2010 con…rmed in Durban in 2011 states that "technology needs must be nationally deter-
mined, based on national circumstance and priorities". The 2015 Paris Agreement follows this
tradition of letting countries decide on technology themselves.
2
The reason is that such over-investments reduce a country’s temptation to cheat
by emitting more rather than less, and they are thus necessary to satisfy the
compliance constraint at the emission stage. For so-called “brown” technolo-
gies, including drilling technologies and other infrastructure investments that
are strategic complements to fossil fuel consumption, investments must instead
be less than the …rst-best amount to satisfy the compliance constraint. Our
most controversial result may be that countries should also be required to in-
vest less than the …rst-best amount in adaptation, i.e., technologies that reduce
the environmental harm in a country (and thus also the country’s bene…t from
continuing cooperation and less emissions).
Our analysis is normative since it describes how to de…ne the best possible
self-enforcing climate treaty. The comparative statics have important policy im-
plications: Naturally, it is harder to motivate compliance if the discount factor
is small, the environmental harm is small, or the investment cost is large. In
these circumstances, the best SPE (i.e., the best self-enforcing treaty) requires
countries to invest more in green technologies and less in adaptation or brown
technologies. If countries are heterogeneous, the countries that are small and
face less environmental harm are the most tempted to free-ride. Thus, for com-
pliance to be credible, such countries must invest the most in green technologies
or the least in adaptation and brown technologies. This advice contrasts the
typical presumption that reluctant countries should be allowed to contribute
less in order to satisfy their participation constraint. While the participation
constraint requires that a country’s net gain of cooperating is positive, the com-
pliance constraint requires that the net gain outweighs the positive bene…t of
free-riding for one period, before the defection is observed. The compliance
constraint is therefore harder to satisfy than the participation constraint in this
model, and, to satisfy it, the reluctant countries must invest more in green
technologies and less in adaptation and brown technologies.
3
be better o¤ if everyone emitted less instead of more:
Variable ri 2 <+ is here capturing the fact that a country’s bene…t depends
on more than its emission levels. We will refer to ri as the country’s technology,
but ri can actually be any variable which in‡uences the bene…t of emitting.
In fact, we also allow ri to in‡uence a country’s environmental cost by letting
ci hi c (ri ), where hi measures country-speci…c environmental harm.
To simplify, we use subscripts for derivatives whenever this is not confusing,
and we abuse notation by writing b00i;gr @ bi (g; ri ) bi g; ri = g g =@ri .
To illustrate the relevance of technologies, we will occasionally refer to the fol-
lowing special types:
De…nition 1.
(A) Adaptation technology is characterized by b00i;gr = 0 and c0 (ri ) < 0.
(B) Brown technology is characterized by b00i;gr > 0 and c0 (ri ) = 0.
(C) Clean technology is characterized by b00i;gr < 0 and c0 (ri ) = 0.
4
capita utility can be written as:
X
ui = bi (gi ; ri ) hi c (ri ) sj gj ki ri : (3)
j2N
Note that it is without loss of generality to assume that the investment cost
is linear in ri , since ri can enter a country’s bene…t function in arbitrary ways.
If the investment cost were another function i (ri ), we could simply de…ne
ebi (gi ; i (ri )) bi (gi ; ri ) and e
ci ( i (ri )) hi c (ri ), treat i (ri ) as the decision
variable, and then proceed as we do below.
Since investments are “sel…sh", in that they are without spillovers, each
country is voluntarily investing the socially optimal amount, conditional on the
emission levels. To see this, note that if it happened that gi = g8i, the …rst best
would require:
Proposition 0.
(i) In the …rst-best, ri ri (g) and g = g.
(ii) In the unique SPE of the stage game, ri ri (g) and g = g.
(iii) If countries had committed to gi = g, the outcome, including the equilibrium
investments, would be …rst best.
5
technology stock, we can already account for the future cost-savings today and
write the net marginal investment cost as ki (1 qir ) e
ki .2
r
If the qi ’s are small, then the analysis below is unchanged since countries do
need to invest in every period (even o¤ the equilibrium path). The investments
are then, in e¤ect, reversible. These assumptions are reasonable in the very long-
run context of climate change, in our view.3 Furthermore, if the qir ’s were instead
large, it would actually be easier to motivate countries to emit less (see Section
3:2 for a further discussion). By ignoring stocks and instead considering the
one-period utilities given by (3), it is straightforward to interpret our dynamic
game as a simple repeated game.
Remark on assumptions and extensions. In (3), we have assumed that tech-
nology investments are sel…sh in that such investments only a¤ect the investing
country’s technology. We have also abstracted away from uncertainty and policy
instruments, and we permit only two possible emission levels. These assump-
tions allow us to derive key insights in a simple setting. Starting from Sections
4, we relax all these assumptions and shows that our main results continue to
hold. The Appendix discusses time-varying parameters, rather than the station-
ary ones in our basic model. The key assumptions behind our results are that
investments are to some extent observable and decided on before the technology
can be used.
3 Self-enforcing Agreements
While the stage game is described above, we here assume that the stage game
is played repeatedly in every period t 2 f1; 2;P:::; 1g. We let 2 [0; 1) be
1
the common discount factor and vit = (1 ) =t
t
ui measures country
i’s continuation value at time t (normalized to per-period utility). The goal of
this section is to characterize the “best”(that is, the Pareto optimal) subgame-
perfect equilibrium (SPE). Since all parameters are invariant in time, the Pareto
optimal SPE is stationary and we skip t-superscripts for simplicity. The Appen-
dix allows for time-varying parameters and contains the proofs which are not in
the text.
costs were nonlinear functions. Allowing for payo¤-relevant stocks is reasonable but they open
up a host of other issues (some of them are discussed by Harstad 2012 and 2015) that are
tangential to the points we are making here.
3 In long-run problems such as climate change, countries must expect to invest repeatedly
partly to maintain the infrastructure and the capacity to produce renewable energy, but also
to invest in research and development e¤ort. See, for example, Dockner and Long (1993) or
Dutta and Radner (2004).
6
to ri = ri (g). Clearly, these strategies also survive as an SPE in the in…nitely
repeated game in which the stage game is played in every period. In fact, in
every SPE in which gi = g, we must have ri = ri (g). For any other equilibrium
candidate ri , country i could bene…t from deviating to ri (g) without any risk
of reducing vi . In other words, from country i’s point of view, emitting more
is the other players’ worst strategy (i.e., the minmax strategy), and an SPE
cannot be sustained with lower utilities. We refer to this equilibrium as the
business-as-usual (BAU) equilibrium and label it with superscript b.
Proposition 1. The worst SPE is BAU: rib ; gib = (ri (g) ; g). This equilibrium
always exists.
Since there are two decision-stages in each period, we must consider the
temptation to deviate at each of them. At the investment stage, a country must
compare the continuation value (vi ) it receives from complying with the SPE
by investing ri , to the maximal continuation value it could possibly obtain by
deviating. Since deviating at the investment stage implies that every country
will emit more beginning from this period, the compliance constraint at the
investment stage is the following:
vi vib
max bi (g; ri ) hi c (ri ) ng ki ri + . (CCri )
1 ri 1
The right-hand side of (CCri ) is maximized when ri = ri (g), implying that the
right-hand side is simply vib . Thus, (CCri ) simpli…es to vi vib . In other words,
as long as every country prefers the SPE to BAU, the compliance constraint for
the investment is trivially satis…ed.
7
At the emission stage, the investment cost for this period is sunk and the
compliance constraint becomes:
vi vib
bi g; ri hi c (ri ) ng + bi (g; ri ) hi c (ri ) si g + (n si ) g + ,
1 1 g
(CCi )
which implies that:
bi (ri ) vi vib
1 . (4)
bi (g; ri ) bi g; ri si hi c (ri ) g g + vi vib
Under condition (Gi ) for “green” technology, more investments relax the
compliance constraint by reducing the lower threshold bi (ri ). Above this thresh-
old, less emission can be sustained as an equilibrium outcome. Under condition
(NGi ) for “non-green” technologies, less investments relax the compliance con-
straint.
As the discount factor < i declines further, (CCgi ) becomes even harder to
satisfy and requires investment levels that increasingly di¤er from the …rst-best
level. Once the discount factor is smaller than a lower threshold referred to
as i < i , gi = g can no longer be sustained in an SPE. The thresholds are
explained in the Appendix, which includes the proofs of the following results.
8
(ii) If < i, then: 4
The result that the …rst best is achievable when the discount factor is suf-
…ciently large is standard in the literature on repeated games.5 Thus, the con-
tribution of Proposition 2 is to characterize the distortions that must occur
if the discount factor is small. To understand the importance of this charac-
terization, it is useful to once again refer to the special cases in De…nition 1.
Clearly, condition (Gi ) is satis…ed for clean technology, while (NGi ) is satis…ed
for adaptation and brown technology. In other words, if the …rst best cannot be
achieved, countries are only motivated to comply with an agreement and emit
less if they have, in advance, invested less in adaptation or brown technologies,
or more in clean technologies. Intuitively, the temptation to free-ride is larger
after investing in adaptation or brown technology, but smaller after investing in
clean technology.
Corollary 2. Compared to the …rst best, the Pareto optimal SPE requires the
countries to:
(i) under-invest in adaptation technology;
(ii) under-invest in brown technology;
(iii) over-invest in clean technology.
These strategic investment levels, which are clearly ine¢ cient conditional on
the emission levels, must be part of the self-enforcing agreement in the same way
as are the small emission levels: any deviation must be triggered by a reversion
to BAU.
Distorting the choice of technology in this manner reduces the temptation
to deviate from the equilibrium. Note that it is not necessary to require so little
or so much investment that emitting less becomes a dominant strategy: it is
su¢ cient to ensure that the bene…t of emitting more is smaller (although still
positive) than the present discounted value of continuing cooperation. Also,
note that if technology were long-lasting and not reversible, it would be easier
to satisfy the compliance constraint. The reason is simply that the deviation
payo¤ would be less than the BAU payo¤ if the investments cannot adjust easily.
and Wolinsky (1995), who show that the Fudenberg and Maskin (1986) folk theorem can be
generalized.
9
other parameters of the model. Compliance is particularly di¢ cult to motivate
if the cost of reverting to BAU is small. The cost of BAU is small if relatively
few countries are polluting (i.e., n is small), if the environmental harm is small
(i.e., hi is small), or if the countries heavily discount the value of cooperating
in the future (i.e., is small). In all these situations, a country i will not …nd it
optimal to comply unless it is requested to invests less in adaptation and brown
technologies, or more in clean technologies. The result that investments in clean
technologies should decline with the discount factor, for example, is certainly at
odds with traditional results in economics.
Furthermore, we show that all investments should increase with the invest-
ment cost ki . For adaptation and brown technologies, we have ri < rib . A larger
ki thus reduces the value of BAU (v b ) compared to cooperation, and makes the
compliance constraint easier to satisfy. Thus, when ki increases, ri can increase
towards ri without violating (CCgi ). For clean technologies on the other hand,
we have ri > rib , and a larger ki again reduces the value of cooperating relative
to BAU. The compliance constraint becomes harder to satisfy. As a response,
countries must invest even more in clean technologies to satisfy (CCgi ) when ki
increases.
Corollary 3. In the Pareto optimal SPE, the smallest and the most reluctant
countries invest the least in adaptation and brown technology, and they invest
the most in clean technology.
The result that countries which are small or have high investment costs ought
to invest more in clean technology is in stark contrast to the idea that countries
should contribute according to ability and responsibility.
6 If, instead, (c0 )2 =c00 > c, investing in adaptation technology is so productive that if n, g,
or hi increases, country i’s environmental harm nghi c (ri ) actually declines when the changes
induce the country to invest more in adaptation technology. This is unrealistic, in our view.
10
The result that countries which are reluctant to cooperate (in that the harm
hi is small) ought to invest more is similarly in contrast to the intuition that
such countries must be given a better deal to make them cooperate.
It is true, of course, that countries that are reluctant either because they are
small or have high investment costs, or because they are subject to less harm,
have participation constraints (i.e., the constraint vi vib ) that are more di¢ cult
to satisfy than for other countries. However, as we have shown above, the
compliance constraint (CCgi ) is more di¢ cult to satisfy than the participation
constraint. Although all countries must obviously bene…t from cooperation
compared to BAU, they must in addition bene…t from cooperation at the stage
when they face the possibility of free-riding one period before the others revert
to BAU.
Interestingly, the international community appears to recognize these prob-
lems. The 2015 Paris Agreement, for instance, emphasizes the importance of
transferring technology to the less-developed countries (Articles 66-71). Such
technology transfers seem like an ideal way of satisfying the compliance con-
straints of reluctant countries without violating their participation constraints.
7 This
utility function is also considered in Battaglini and Harstad (2014), who do not study
SPEs, but instead the Markov-perfect equilibria when countries can commit to the emission
levels. The …rst best and the BAU equilibrium are as in that paper, of course.
11
where B and K are positive constants. Here, y is a country’s bliss level for
consumption, and consumption is the sum of gi (energy from fossil fuels) and ri
(energy from renewable energy sources). Since @ 2 ui =@gi @ri < 0, we explicitly
consider only clean technology. We can easily reformulate the utility function
such that the investment cost becomes linear,8 although there is no need to do
so here.
Since the emission tax is the only cost of consuming fossil fuel, gi is chosen
by the consumers to satisfy the …rst-order condition:
B (y (gi + ri )) = i.
The left-hand side is also equal to the consumer’s willingness to pay for green
technology, so private investors invest according to the …rst-order condition:
12
factors, the emission tax must be allowed to fall to satisfy the compliance con-
straint. The associated increase in emissions can be mitigated by introducing
an investment subsidy.
Note that the investment-stage compliance constraints will never bind …rst.
As soon as one country deviates by setting a di¤erent investment subsidy, in-
vestors in all countries anticipate that cooperation will break down and de-
mand for their technology at the emission stage will be reduced. This lowers
investments everywhere, not only in the deviating country. Deviating at the
investment stage immediately gives the deviator the BAU payo¤, plus the ben-
e…t of the other countries’larger investments induced by their subsidies. These
subsidies are zero for and are small for discount factors close to . Conse-
quently, some < exists such that the compliance constraint at the investment
stage is not binding when 2 ( ; 1). (The proof in the Appendix derives both
thresholds.)
= cn ( ) and
&= ( ) , where
p
( ) c (n 1) 1 2 + B=K 0:
Corollary 4. The sum of the equilibrium emission tax and the investment sub-
sidy is, for every , equal to nc, the …rst-best Pigouvian tax level.
5 Conclusions
To confront global climate change, an environmental treaty must address two
primary challenges. The treaty must be self-enforcing, and it must lead to the
development of green technology. This paper analyzes these challenges in a joint
9 The proposition implies that the equilibrium investment level, r , given by (5), stays
i
unchanged as the discount factor falls. On the one hand, the fact that a larger g must be
tolerated implies that it becomes optimal to invest less in clean technology. On the other
hand, the countries can dampen the increase in g by requesting countries to invest more in
green technology upfront. These two e¤ects cancel each other out when g and r are perfect
substitutes. Relative to the ex post optimal level, however, it is clear that r r (g) is positive
and increases as falls, just as the equilibrium investment subsidy. The optimal investment
level, conditional on the emission level gi , is decreasing in gi and given by:
B (y (gi + ri )) B (y gi )
r (gi ) = = :
K B+K
13
framework and uncovers policy-relevant interactions between them. Speci…cally,
we demonstrate that when free-riding is tempting and cooperation di¢ cult to
sustain, the best self-enforcing treaty requires countries to over-invest in “green”
technology, reducing the temptation to pollute, or under-invest in adaptation or
“brown” technology that would have made free-riding more attractive. When
countries are heterogeneous, it is particularly countries that are small or re-
luctant to cooperate (because their environmental harm is relatively small, for
example) that are most tempted to pollute. To ensure that compliance by these
countries is credible, small or reluctant countries must invest the most in green
technology, or the least in adaptation and brown technology.
In a time when the world struggles to develop and reach an agreement on a
global climate change treaty, it is natural that the motivation for our analysis
has mainly been normative. We believe that the international community has
not yet implemented the best possible self-enforcing treaty, and thus do not
expect that our predictions are directly observable or consistent with the facts
of today. That said, our theory is testable: we do provide a number of predic-
tions that may eventually be compared to the data. Our assumptions are also
in line with the facts: Policymakers do have few sanctions available (meaning
a treaty must be self-enforcing) and they certainly consider the development
of technology to be of great importance. Furthermore, some of the countries
that have invested the most in green technology (notably the European Union)
are also the ones that have complied with the Kyoto Protocol to the largest ex-
tent.10 Other countries that have instead invested in brown technology (notably
Canada and Australia) ended up not complying or increased emissions. Finally,
although the 2015 Paris Agreement does not specify how much countries ought
to invest in green technology, it does recognize the importance of technology
and of technology transfers to less developed countries (Articles 66-71). In our
model, such technology transfers appear to be an ideal way of satisfying com-
pliance constraints without violating participation constraints. Future research
should empirically clarify the interaction between technology, emissions, and
compliance to test the model’s predictions.
Theoretical research should also continue. Our simple workhorse model has
proven to be su¢ ciently tractable to be extended in many ways, but our ap-
proach is still only a …rst cut. We have simpli…ed tremendously by abstracting
away from payo¤-relevant stocks of pollution or technologies. We have focused
exclusively on the Pareto optimal subgame-perfect equilibrium, although the
actual transition toward such a treaty appears to be characterized by high trans-
action costs and multiple wars of attritions. By focusing on the best subgame-
perfect equilibrium, we have also abstracted from the possibility of opting out
of the negotiations at the beginning of the game. When countries are hetero-
geneous, it may actually be optimal, even for the countries that cooperate, to
exclude certain reluctant countries, since these countries may, with some prob-
ability, cheat and thus trigger a costly and long-lasting punishment phase. One
1 0 “EU over-achieved …rst Kyoto emissions target” (October 9th, 2013, European Commis-
sion: ec.europa.eu/clima/news/articles/news_2013100901_en.htm).
14
of the goals with this project has been to provide a tractable workhorse model
that can be developed along these lines in future research.
15
6 Appendix
The proofs of the basic results, Propositions 2 3, allow for time-varying para-
meters.
Proof of Proposition 2.
Since investments are sel…sh, the Pareto optimal SPE satisfying gi = g is for
each i 2 N specifying the rit closest to ri ;t satisfying the compliance constraints.
That is, the Pareto optimal SPE solves:
max
t
uti = bti rit ; g nt cti rit g kit rit s.t.
ri
r;t
vit vib;t 0, (CCi )
vit+1 vib;t+1
t g;t
i bti g; rit + bti g; rit + sti hti c rit g g + 0. (CCi )
1
(i) Since vit > vib;t at ri ;t , both conditions hold if is close to 1. A binding
t
(CCg;t bt t bt r ;t , then r ;t satis…es both
i ) de…nes i (ri ) implicitly, and if i i i i
compliance constraints.
t
(ii) As soon as declines below the level i , (CCg;t r;t
i ) binds (before (CCi )
does). The problem above is then a Kuhn-Tucker maximization problem which
can be written as:
0 1
vit+1 vib;t+1
max uti + ti @ bti g; rit + bti g; rit + sti hti c rit g g + A,
rit 1
while the second-order condition is satis…ed for uti su¢ ciently concave. Since ri ;t
is de…ned by @uti =@rit = 0, rit must increase above ri ;t to satisfy this …rst-order
condition if the second term in (6) is positive, i.e., if:
16
As declines further, (CCg;t
i ) can only be satis…ed if ri
t
ri ;t increases more.
;t
Eventually, becomes so small than either (i) rit ri becomes so large that
0 0
(CCr;t t t g;t
i ) is violated, (ii) ui and thus vi is reduced so much that (CCi ) is
0 t
violated at some earlier date t < t, or (iii) reaches hzero. We let i measure
t
the maximum of these three thresholds. Clearly, ti 2 0; i . QED
Proof of Proposition 3.
t
If < i , the Pareto optimal SPE satisfying git = g ensures that rit solves
bt (rt ) = , so that (CCg;t ) binds. As long as (CCg;t ) binds, we can simply
i i i i
di¤erentiate the left-hand side of (CCg;t t
i ) to learn how ri must change with the
other parameters at time t:
t
@rit i;h sti c (rit ) g g
= = ;
@hti t
i;r
t
i;r
t
@rit i;s hti c (rit ) g g
= = ; where (7)
@sti t
i;r
t
i;r
t
i;r =@ bti g; rit + bti g; rit =@rit + sti hti g g c0 rit :
Thus, @rit =@hti and @rit =@sti are both negative if (Gti ) holds (then, ti;r > 0), and
otherwise they are positive.
If we could write bti ( ) = it b ( ), where the importance of consumption,
t
i , could be particularly high at times of recessions, then we could show that
countries ought to invest more at such times in clean technologies, and less in
adaptation or brown technologies, for compliance to be credible:
17
Under (??), i;r > 0 at ri and i;r remains positive as long as a larger ri
weakens (CCgi ) (that is, for > i ). Similarly, under (??), i;r < 0 at ri and
g
i;r remains negative as long as a smaller ri weakens (CCi ) (that is, for > i ).
(i) E¤ ect of ki : The compliance constraint depends on ki because:
To see the sign of ri rib , assume, for a moment, that ri remains at ri . If so,
all investment levels are given by the …rst-order condition:
@bi (g; ri )
hi c0 (ri ) ng = ki ,
@ri
which we can di¤erentiate to get
@ 2 bi (g;ri )
dri @ri @g + hi c0 (ri ) n
= @ 2 bi (g;ri )
; (9)
dg hi c00 (ri ) gn
(@ri )2
Thus, for adaptation and brown technologies, ri < rib . For such technologies we
also have ri ri < rib and i;r < 0, so from (8) we have that dri =dki > 0. For
clean technologies, Eq. (10) gives ri > rib . For clean technologies, we also have
ri ri > rib and ir > 0, so from (8) we again have that dri =dki > 0.
(ii) E¤ ect of si and : The e¤ect of si on ri is exactly as in the case with
time-dependent variables (7). The e¤ect of is trivial. Note that ri does not
depend on si or .
(iii) E¤ ect of n: Consider …rst the case where c0 (ri ) = 0 (brown or clean
technologies). In this case, @ vi vib =@n = g g hi c > 0, so
dri i;n g g hi c
= = ;
dn i;r 1 i;r
which has the opposite sign of i;r . Note that ri does not depend on n.
Consider next adaptation technologies, where c0 ( ) < 0 but @bi ( ) =@ri = 0.
If ri were invariant in n, we would have @ vi vib =@n = hi c rib g c (ri ) g ,
18
but ri and rib may di¤er since ri (g) depends on g. If c (ri (g)) g increased in
g, then we would have c rib g > c (ri ) g. If we use Eq. (9), we can show that
c (ri (g)) g increases in g if:
2
[c0 (ri (g))]
c (ri (g)) > 0: (11)
c00 (ri (g))
So, under this condition, @ vi vib =@n and i;n would be positive for ri close
to ri and, then, dri =dn > 0.
E¤ ect of hi : This e¤ect is derived in a similar way. For adaptation technolo-
gies, if ri were invariant in hi , we would have @ vi vib =@hi = n c rib g c (ri ) g ,
where the bracket is, as before, positive under condition (11). Then, i;h > 0
and, therefore, dri =dhi > 0 for adaptation technologies. For brown technologies
we have i;h > 0 and dri =dhi > 0 while for clean technologies we have i;h < 0
and dri =dhi < 0, even though ri is independent of hi . QED
Proof of Proposition 4.
If we de…ne di y gi ri to be the decrease in consumption relative to
the bliss level y, the …rst-best emission (and consumption) level is simply given
by d = cn=B, while, in BAU, db = c=B. We can write the continuation value
as:
v = d (nc Bd=2) + r (nc Kr=2) cny:
(i) The compliance constraint at the emission stage can be written as:
v K 2 2
nc r rb r rb (CCgc )
1 2
vb
+ c (n 1) d db + ;
1
which implies that:
dB = nc , where
h p i
( ) c (n 1) 1 2 + B=K
" r #
2 g g
= c (n 1) 1 (1 ) = ;
19
g
where ( ) decreases from c (n 1) to 0 as increases from 0 to . This d is
implemented by the emission tax nc . To ensure ri = r , the subsidy must
be .
Note that the investment-stage compliance constraint can be written as
v c vb
& (n 1) + , (CCrc )
1 K 1
which always holds when & = ( ) ! 0. When falls, v declines and & = ( )
increases. The threshold is de…ned implicitly by requiring (CCrc ) to hold with
equality at . QED
20