Abstract
Solar geoengineering (SG) may have the potential to reduce extreme climate damages worldwide. Yet, international coordination will make the difference between success and failure in leveraging it. Using a simple game-theoretic framework, we investigate whether the stability of an efficient, self-enforcing international agreement on SG is attainable. We demonstrate that side payments from countries less vulnerable to climate change to those more vulnerable can guarantee the stability of an efficient agreement. The size of the side payments will vary within a zone of possible agreement, which will change depending on certain key assumptions. For example, assuming stronger mitigation reduces the necessary payments. Alternatively, asymmetry in national damages from SG over-provision vs. under-provision justifies larger payments; here, the welfare-optimal strategy may be deployment that makes no one worse off. We also show that an agreement may be stable without side payments if deployment costs are substantial and counter-SG is available, while a moratorium may be socially optimal if SG brings substantial global non-excludable fixed costs.
Similar content being viewed by others
Code Availability
The code used to obtain the results presented in this paper is available upon the request.
Notes
The dataset in Rickels et al. (2020) allows to calibrate ki but not \(\alpha\). In an attempt to avoid merging distinct and potentially conflicting climate-impacts datasets, we set \(\alpha\) at a constant value across countries. This simplifying assumption implies that national damage functions are calibrated so that damages in the absence of SG reflect relative vulnerability to climate change across countries but do not necessarily depict an accurate associated level of damages.
We chose the countries with benefits or costs above 1% of the change in global GVA.
Note that in this case the overall damages from SG are also increased. To demonstrate that the results are not crucially dependent on the impact of the change in total damages, we also consider an opposite formulation: reduction of the damages for SG deployment levels below optimal point. This case is presented in "Appendix B" and demonstrates a similar impact of the degree of asymmetry on the ZOPA.
Following the dynamic integrated assessment model of climate and economy (DICE), we state the abatement costs increase stronger than linearly with the abatement level.
We leave endogenous mitigation for possible future study.
References
Barrett S (1994) Self-enforcing international environmental agreements. Oxf Econ Pap 46:878–894
Barrett S (2001) International cooperation for sale. Eur Econ Rev 45(10):1835–1850. https://doi.org/10.1016/S0014-2921(01)00082-4
Belaia M, Moreno-Cruz JB, Keith DW (2021) Optimal climate policy in 3d: mitigation, carbon removal, and solar geoengineering. Clim Change Econ. https://doi.org/10.1142/S2010007821500081
Bunn M (2019) Governance of solar geoengineering: Learning from nuclear regimes. Harvard Project on Climate Change Agreement, pp 51–54
Caparrós A, Finus M (2020) Public good agreements under the weakest-link technology. J Public Econ Theory 22(3):555–582. https://doi.org/10.1111/jpet.12426
Carraro C, Eyckmans J, Finus M (2006) Optimal transfers and participation decisions in international environmental agreements. Rev Int Organ 1(4):379–396. https://doi.org/10.1007/s11558-006-0162-5
Carraro C, Siniscalco D (1993) Strategies for the international protection of the environment. J Public Econ 52(3):309–328. https://doi.org/10.1016/0047-2727(93)90037-T
Chander P, Tulkens H (1992) Theoretical foundations of negotiations and cost sharing in transfrontier pollution problems. Eur Econ Rev 36(2–3):388–399
Cherry TS, Kroll S, McEvoy D, Campoverde D, Moreno-Cruz J (2022) Climate cooperation in the shadow of solar geoengineering: an experimental investigation of the moral hazard conjecture. Environ Polit 1:1–9
Crutzen PJ (2006) Albedo enhancement by stratospheric sulfur injections: A contribution to resolve a policy dilemma? Clim Change 77(3–4):211. https://doi.org/10.1007/s10584-006-9101-y
d’Aspremont C, Jacquemin A, Gabszewicz JJ, Weymark JA (1983) On the stability of collusive price leadership. Can J Econ 16(1):17–25
Diamantoudi E, Sartzetakis ES (2006) Stable international environmental agreements: An analytical approach. J Public Econ Theory 17:1
Eigruber M, Wirl F (2018) Climate engineering in an interconnected world: the role of tariffs. Dyn Games Appl 8(3):573–587
Emmerling J, Tavoni M (2018) Exploration of the interactions between mitigation and solar radiation management in cooperative and non-cooperative international governance settings. Glob Environ Chang 53:244–251
Finus, M., & Furini, F. (2023). The Governance Architecture of Climate Agreements in the Light of Risky Geoengineering. preprint SSRN https://ssrn.com/abstract=4082787.
Finus M, Furini F, Rohrer AV (2021) The efficacy of international environmental agreements when adaptation matters: Nash-Cournot vs Stackelberg leadership. J Environ Econ Manag 109:1
Finus M, McGinty M (2019) The anti-paradox of cooperation: Diversity may pay! J Econ Behav Organ 157:541–559
Heyen D (2016) Strategic Conflicts on the Horizon: R&D Incentives for Environmental Technologies. Clim Change Econ 07(04):1650013. https://doi.org/10.1142/S2010007816500135
Heyen D, Horton J, Moreno-Cruz J (2019) Strategic implications of counter-geoengineering: Clash or cooperation? J Environ Econ Manag 95:153–177. https://doi.org/10.1016/j.jeem.2019.03.005
Heyen D, Lehtomaa J (2021) Solar geoengineering governance: A dynamic framework of farsighted coalition formation. Oxford Open Clim Change 1(1):1. https://doi.org/10.1093/oxfclm/kgab010
Horton JB, Keith DW (2019) Multilateral parametric climate risk insurance: A tool to facilitate agreement about deployment of solar geoengineering? Climate Policy 19(7):820–826. https://doi.org/10.1080/14693062.2019.1607716
Keohane RO, Victor DG (2011) The regime complex for climate change. Perspect Polit 9(1):7–23. https://doi.org/10.1017/S1537592710004068
Lessmann K, Kornek U, Bosetti V, Dellink R, Emmerling J, Eyckmans J, Nagashima M, Weikard H-P, Yang Z (2015) The stability and effectiveness of climate coalitions. Environ Resource Econ 62(4):811–836
Lloyd ID, Oppenheimer M (2014) On the design of an international governance framework for geoengineering. Glob Environ Polit 14(2):45–63
Manoussi V, Xepapadeas A (2017) Cooperation and competition in climate change policies: mitigation and climate engineering when countries are asymmetric. Environ Resource Econ 66(4):605–627
Manoussi V, Xepapadeas A, Emmerling J (2018) Climate engineering under deep uncertainty. J Econ Dyn Control 1(94):207–224
McGinty M (2006) International environmental agreements among asymmetric nations. Oxf Econ Pap 59(1):45–62. https://doi.org/10.1093/oep/gpl028
Mendelsohn R, Dinar A, Williams L (2006) The distributional impact of climate change on rich and poor countries. Environ Dev Econ 11(2):159–178
Meya JN, Kornek U, Lessmann K (2018) How empirical uncertainties influence the stability of climate coalitions. Int Environ Agreements: Polit Law Econ 18:175–198
Millard-Ball A (2012) The tuvalu syndrome. Clim Change 110(3):1047–1066
National Academies of Sciences, Engineering, and Medicine (2021) Reflecting sunlight: recommendations for solar geoengineering research and research governance. The National Academies Press. https://doi.org/10.17226/25762
Nye JS (2019) Notes on Insights from Other Regimes: Cyber. Harvard Project on Climate Change Agreement, pp. 55–59
Parker A, Horton JB, Keith DW (2018) Stopping solar geoengineering through technical means: a preliminary assessment of counter-geoengineering. Earth’s Future 6(8):1058–1065. https://doi.org/10.1029/2018EF000864
Parson EA (2014) Climate engineering in global climate governance: implications for participation and linkage. Transnatl Environ Law 3(1):89–110. https://doi.org/10.1017/S2047102513000496
Ricke KL, Moreno-Cruz JB, Caldeira K (2013) Strategic incentives for climate geoengineering coalitions to exclude broad participation. Environ Res Lett 8(1):014021
Rickels W, Quaas MF, Ricke K, Quaas J, Moreno-Cruz J, Smulders S (2020) Who turns the global thermostat and by how much? Energy Econ 91:104852. https://doi.org/10.1016/j.eneco.2020.104852
Schelling TC (1990) The strategy of conflict, 2nd edn. Harvard University Press, Harvard
Smith W (2020) The cost of stratospheric aerosol injection through 2100. Environ Res Lett 15(11):1104. https://doi.org/10.1088/1748-9326/aba7e7
Tol RSJ, Downing TE, Kuik OJ, Smith JB (2004) Distributional aspects of climate change impacts. Glob Environ Chang 14(3):259–272
Urpelainen J (2012) Geoengineering and global warming: a strategic perspective. Int Environ Agreements: Polit Law Econ 12(4):375–389
United Nations Environment Programme (2022). What you need to know about the COP27 Loss and Damage Fund, 29 November. https://www.unep.org/news-and-stories/story/what-you-need-know-about-cop27-loss-and-damage-fund
Weitzman ML (2015) A voting architecture for the governance of free-driver externalities, with application to geoengineering. Scand J Econ 117(4):1049–1068. https://doi.org/10.1111/sjoe.12120
Wilson B (2021) Keynes Goes Nuclear: Thomas Schelling and the Macroeconomic Origins of Strategic Stability. Mod Intellect Hist 18, 171–201.
Acknowledgements
The authors are deeply indebted to Robert Stavins for constructive feedback and inspiring conversations. The paper greatly benefitted from valuable comments by Scott Barrett, Johan Eyckmans, Michael Finus, Alton Frye and Dale S. Rothman, and helpful guidance at early stages of this work from Edward Parson. Also, thanks are due to Daniel Heyen and Christian Traeger for their help in improving the clarity of the model presentation. We are grateful to two anonymous referees for their constructive comments, which helped to improve the paper. The authors are responsible for any and all remaining errors.
Funding
Author Irina Bakalova was supported by Harvard’s Solar Geoengineering Research Program and the Basic Research Program at the HSE University.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors have no competing interests to declare that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
A Default Specification
In the default model specification, damage function for country \(i \in N\), \(i=1,2, \dots n\), reads:
where \(G_N\) is realized SG level and \(k_i\) - country \(i's\) optimal SG level. The damage function can be rewritten as:
with \(d_i \equiv \frac{1}{2} \alpha k_i^2\) - country \(i's\) climate damages in the absence of SG.
In the following analytical derivations we offer a solution in a more general form, where parameter \(d_i\) is explicit. It must, however, satisfy the following condition:
This condition ensures that damages cannot be negative, i.e., SG cannot be used to bring benefits beyond reduced climate damages. The value \(d_i\) such that \(d_i > \frac{1}{2} \alpha k_i^2\) describes the case when the damage compensation potential of SG differs across countries. For example, \(d_i = \frac{1}{2 \cdot 0.9} \alpha k_i^2\) means that SG can be used to compensate for \(90\%\) of climate damages in each country (in each - different level of SG). Here, residual damages would be largest (in terms of its levels) for the country most vulnerable as illustrated in Fig. 8.
1.1 A.1 Third-Stage Solution
Signatories
Consider an arbitrary coalition \(S \subseteq N\). Total SG deployment of coalition members amounts to \(G_S = \sum _{i \in S} g_i\); the average level of SG preferences of coalition members is \({\bar{k}}_S = \sum _{i \in S} k_i/s\), where s - number of coalition members.
In the third stage of the game, coalition members minimize their total damages \(D_S\):
Coalition S reaction function:
Non-signatories
The non-signatory \(i \in N\setminus S\) minimizes its individual damage function:
The non-signatory i’s reaction function reads:
Total SG Deployment Level and Associated Damage Costs
Consider country j to be the country with the largest preferred SG level among non-signatories. Thereby, an arbitrary coalition S may be“active”, i.e. define the global thermostat, if it’s optimal deployment level exceeds \(k_j\) and “passive” otherwise.
1.2 A.2 The Proposition Proof
Consider an arbitrary active coalition \(S \subseteq N\). Optimal level of SG for this coalition is \(\bar{k_S}\). Let us denote by \(DR^S\) the set of drivers: \(j \in DR^S\) if \(j \in S\) and \(k_j > \bar{k_S}\), and by \(ND^S\) the set of non-drivers: \(i \in ND^S\) if \(i \in S\) and \(k_i \le \bar{k_S}\).
The proposition states that the total gain of non-drivers from cooperation (relative to the no-cooperation case) exceeds the sum of the gains of all drivers from their unilateral deviation from the cooperation:
Substitute the considered functional form of damages, here \(k_h\) denotes the highest SG preference of all considered countries from set N:
Proof
By definition, \(k_h\) is the largest value of all nationally-optimal SG levels. It is larger than the average of optimal levels for all non-drivers: \(k_h \ge {\bar{k}}_{ND^S}\). Multiplying both sides by the number of non-drivers in a coalition S, we arrive at:
where \(|ND^S|\) is a number of non-drivers in a coalition S, \(k_{ND^S} = \sum _{i \in ND^S} k_i\) is the sum all SG amounts preferred by non-drivers. Then we add them up and subtract from the right hand side (RHS) of the sum of all preferences of coalition members (\(k_S\)):
After some reshuffling we arrive at:
We then rewrite this inequality using summation operator as:
.
Multiplying both sides by \(\left( k_h - {\bar{k}}_S \right) \ge 0\), the difference between the largest preferred SG level among all countries in N and the average of coalition member’s preferred SG level.
By definition \(k_h \ge k_i ~ \forall i \in N\), the RHS is greater than or equal to \(\sum _{j \in DR^S} \left( k_j - {\bar{k}}_S\right) ^2\). Therefore, the following holds:
Finally, by multiplying both sides by \(0.5 \alpha\) we arrive at the following expression, which was to be demonstrated:
. \(\square\)
1.3 A.3 Illustration
Figure 9 illustrates our Proposition, which itself is in the spirit of Coase theorem. What differs from the Coase theorem case, is that we are dealing with public “gob" and thus country may alter between being a loser or winner dependent on the amount of "gob" supplied.
We consider two groups of countries: (i) non-drivers and (ii) potential free drivers. Note that marginal damages to non-drivers are negative at low SG values, indicating that non-drivers benefit from small amounts of SG due to the associated reduced climate damages. Similarly, marginal benefits to potential free driversis negative when SG levels are large. Thereby, there is a difference between the sum of damages to free drivers in Nash equilibrium (where some free drivers actually lose) and the sum of damages that potential free drivers bear from the unilateral deviations.
In Fig. 9, the green area shows the benefits to non-drivers from global cooperation relative to the Nash equilibrium case. Red area shows the benefits to potential free drivers in Nash equilibrium relative to the global cooperation case. Here, red area (triangle) in the negative zone indicates losses to potential drivers from SG over-provision in Nash equilibrium.
B Asymmetric Damage Function
We define the degree of asymmetry B as a deviation (in degrees) of \(D_i^{asym}\) from the default function \(D_i\) at a point \(G_N=k_i+1\). The resulting degree of asymmetry is:
where \(\alpha =1.5\) and associated \(arctan(\alpha )=56.3\) [degrees].
We consider two types of asymmetry: with relatively larger damages from SG over-provision and smaller damages from SG under-provision. This helps us to demonstrate the potential for the qualitative change in the results. The illustration of the considered asymmetric damage functions are presented in the Fig. 10.
Changes in the total amount of SG, benefits to ND of cooperation and benefits to DR from unilateral deviation for the left-side asymmetry are depicted in the Fig. 11. This figure demonstrates that ZOPA is positive and increasing with the degree of asymmetry.
We also offer results for the alternative measure of asymmetry, which is as follows: \((\beta -\alpha )/\alpha\). The corresponding graphs are presented in the Fig. 12.
C Non-negligible Direct SG Deployment Costs
When assuming that the deployment costs are non-negligible, the damage function reads:
where z is the slope of marginal SG deployment costs and \(g_i \ge 0\).
1.1 C.1 Third-Stage Solution
Signatories
Minimization problem for members of a coalition \(S \subseteq N\):
FOC:
Adding it up for the coalition members \(i \in S\), we arrive at:
Reaction function for a coalition S then reads:
Non-signatories
Non-signatory \(i \in N\setminus S\) minimizes it’s individual damage function as follows:
FOC:
The associated reaction function reads:
1.2 C.2 Selected Illustrations
Figure 13 shows individual and total levels of SG deployment in global cooperation, Nash equilibrium and under the unilateral deviation case. The Figure is presented for a uniform distribution of optimal values \(k^U = \{1,2,3,4,5,6,7 \}\) for countries 2–7.
For countries with very low preferred SG levels (1 and 2), the deviation always results in zero individual SG deployment and an increase in total deployment, since otherl coalition members have larger average over preferred SG levels. The difference between the total SG deployment in cooperation and deviation decreases with marginal deployment costs, z.
Preferences of countries 3 and 4 are the closest to the socially optimal SG deployment level. Therefore, the difference between total SG level in the cooperative case and their unilateral deviation is tiny. This creates incentives to free ride: individual SG deployment level (deviation) is either zero or is very low.
Countries 5–7 have large preferred SG level and free drive when marginal deployment costs are relatively low: their individual SG deployment level in the case of deviation determines the total SG level. However, increasing costs of deployment curb their free driving ability and the SG deployment level decreases below it’s cooperative level. This is the case because the optimal SG level of other coalition members is lower. As a result, these countries do not have an incentive to free drive when SG deployment costs are substantial. However, as deployment costs are increasing, countries may become free riders. In the considered range of marginal deployment costs this happens only for the country 5 shown in Fig. 14.
Figure 14 depicts individual losses in cooperation, Nash equilibrium and unilateral deviation cases for countries 2–7. Unilateral deviation occurs if the associated losses are smaller than in the cooperative case. This difference is depicted in Fig. 14 by the area. The area in green indicates the case when a country is better off in the cooperative state. Figure 14 demonstrates that countries with large preferred SG levels (countries 5–7) have an incentive to deviate when SG deployment costs are low. They also may have an incentive to act as free riders if deployment costs are substantial. In the considered range of marginal deployment costs, it happens only for the country 5.
Countries whose preferred SG level is close to the socially optimal value (i.e., countries 3 and especially 4) have the strongest free-riding incentive. For these countries deviation represents costs with no benefits and thus is not rational. Countries with small preferred SG level, such as country 2, have small incentive to free ride at a relatively large deployment costs.
D Counter SG with Non-negligible Deployment Costs
Damage function:
Where z is the slope of marginal SG or counter-SG deployment costs. The main difference with the previous case is that \(g_i\) may be both positive and negative: \(g_i \in R\).
1.1 D.1 Third-Stage Solution
Minimization problems are similar to the previous case with non-negligible costs. The difference is in the reaction functions for the case where \(g_i\) is not restricted to be non-negative:
Reaction function for a coalition S reads:
Non-signatory \(i \in N\setminus S\) reaction function reads:
1.2 D.2 Illustrations
Figure 15 shows individual SG deployment levels for the scenarios of global cooperation, Nash equilibrium and unilateral deviation, where marginal costs of both SG and counter-SG deployment aret set to \(z = 0.1\). The behavior of countries is intuitive: non-drivers counteract SG deployment activities of drivers (subject to a specified limit). In Nash equilibrium (upper graphs) some countries may switch from counter-SG to SG deployment as the CSG limit increases. In unilateral deviation (lower graphs) countries are clearly divided into two groups: (i) drivers that deploy SG beyond cooperative optimum, and (ii) non-drivers that counteract SG deployment. Note that a country whose optimal SG level almost coincides with the global optimal level (country 4 in the uniform distribution, middle column) tends to free ride in the case of a unilateral deviation, avoiding either activity.
Figure 16 depicts countries’ individual losses in Nash equilibrium and in the case of a unilateral deviation. For comparison, individual losses in cooperation are indicated by dashed lines. Color stays the same for each country.
E Exogenous Mitigation
Consider the global share of emissions abatement \(\sum _{i \in N}{\bar{a}} = {\bar{A}}_N \in [0, 1]\).
Damage function then reads:
Where \(c_i\) in the parameter of abatement costs of country i.
Signatories
Minimization problem for members of a coalition \(S \subseteq N\) reads:
Non-signatories
Non-signatory \(i \in N\setminus S\) minimize their individual damage function as following:
1.1 E.1 Amount of Transfers
Minimum amount of transfers to sustain the cooperation is the sum of benefits from deviation of all free drivers:
Note that the actual transfers size may exceed this level.
F Fixed Costs Associated with SG Deployment
Damage function:
Where
1.1 F.1 The Proposition Proof
The difference between the default specification and the modification with fixed costs of deployment is that a corner solution may come up. Here, corner solution means moratorium on SG. We consider an active coalition S with the optimal deployment level \(\bar{k_S}\). We distinguish three cases depending on the magnitude of fixed costs, FC:
-
\(FC < 0.5 \alpha (\bar{k_S})^2\) - Deployment equilibrium (interior solution) in both coalition and no cooperation. The case is similar to the default specification, without FC.
-
\(0.5 \alpha (\bar{k_S})^2 \le FC < 0.5 \alpha k_h^2\) - Moratorium on SG in coalition and deployment in non-cooperative scenario.
-
\(FC \ge 0.5 \alpha k_h^2\) - Moratorium on SG in both coalition and non-cooperative scenario: no collective action problem.
The interesting case (and thus our focus in the following) is when \(0.5 \alpha (\bar{k_S})^2 \le FC < 0.5 \alpha k_j^2\).
Individual damages | |
---|---|
\(ND: k_i \le (2FC/\alpha )^{0.5}\) | |
\(D_i(S)\) | \(d_i\) |
\(D_i(S \setminus j)\) | \(0.5 \alpha k_j^2 - \alpha k_i k_j + FC + d_i\) |
\(D_i(S \setminus j) - D_i(S)\) | \(0.5 \alpha k_j (k_j - 2 k_i) +FC\) |
\(DR: k_j > (2FC/\alpha )^{0.5}\) | |
\(D_j(S)\) | \(d_j\) |
\(D_j(S \setminus j)\) | \(- 0.5 \alpha k^2_j + FC + d_j\) |
\(D_j(S) - D_j(S \setminus j)\) | \(0.5 \alpha k_j^2 - FC\) |
Proposition reads: \(\sum _{j \in DR^S} (D_j(S) - D_j(S\setminus j)) \le \sum _{i \in ND^S} (D_i(\emptyset ) - D_i(S))\)
After we substitute the considered functional form of damages, where \(k_h\) denotes the largest preferred SG level of all considered countries from set N, it reads:
Proof
\(|DR^S|\) is the number of drivers in a coalition S and \(|ND^S|\) is the number of non-drivers in a coalition S.
We can rewrite as follows:
Assumming \(0.5 \alpha (\bar{k_S})^2 \le FC\), it follows:
Multiply by \(2/\alpha\) and rewrite LHS in an identical form:
Open the summation operator:
Can be rewritten as follows:
Combine RHS in the square of a difference:
Rewrite LHS in an identical form:
Cancel out three terms from LHS:
The following is true by the definition:
\(\square\)
G Countries with Distinct Decision-Making Weights
Consider an arbitrary active coalition \(S \subseteq N\). We introduce country-specific weights, \(w_i\), which are normalized to one in a considered coalition S: \(\sum _{i \in S} w_i =1\). As we are interested in decision making within one considered coalition, we can normalize weights to one in any coalition we analyze. Then, cooperative solution is the result of the minimization of weighted sum of damage functions:
The solution reads:
where \(DR^S\) is the set of drivers: \(j \in DR^S\) if \(j \in S\) and \(k_j > k_{wgt}\), and \(ND^S\) is the set of non-drivers: \(i \in ND^S\) if \(i \in S\) and \(k_i \le k_{wgt}\).
1.1 G.1 The Proposition Proof
The Proposition reads:
In the considered specification it takes the following form:
Proof
Since \(k_h\) denotes the largest preferred SG level, the following inequality holds:
Then we add the term \(\sum _{i \in S} w_i k_i\) to both RHS and LHS, arriving at:
We then rewrite the LHS and substitute \(k_{wgt} \equiv \sum _{i \in S} w_i k_i\) in the RHS:
As country-specific weights in a coalition are normalized to one, we may use the following equality \(\sum _{i \in DR^S}w_i +\sum _{i \in ND^S}w_i =1\):
Now we take the sum operator out of brackets:
We then multiply both RHS and LHS by \((k_h - k_{wgt} ) >0\):
As \(k_h \ge k_j ~ ~ \forall j\in N\), we have \((k_h - k_{wgt} ) \ge (k_j - k_{wgt})\) , and therefore:
Or just:
Finally, by multiplying both sides by \(0.5\alpha\) we arrive at the expression, which was to be demonstrated:
\(\square\)
H Empirical Calibration
To identify individual preferences of countries, we take the following steps:
-
1.
From the paper by Rickels et. al (2020), we adopt the following estimates:
\(V_i(0)\) - absolute impact on country GVA in the absence of SG deployment
\(V_i(G^*)\) - absolute impact on country GVA in the optimal level of SG deployment
We then use these data to find the normalized difference:
$$\begin{aligned} \frac{V_i(G^*) - V_i(0)}{V_i(0)} = V_i \end{aligned}$$ -
2.
We use the normalized difference in countries’ GVA to estimate the normalized difference in countries’ damages when \(SG = 0\) and \(SG = G^*\) as formulated in our model:
$$\begin{aligned} 2/\alpha \cdot (D_i(0) - D_i(G^*)) = 2 G^* k_i - G^{*2}= V_i \end{aligned}$$. Following the approach in the paper by Rickels et al.,(2020), we define the optimal level of SG (\(G^*\)) as GDP-weighted average of countries’ preferences \(k_i , i \in N\):
$$\begin{aligned} G^* = \sum _{i \in N} \frac{GDP_i}{\sum _{j \in N} GDP_j} k_i \end{aligned}$$ -
3.
We solve the system of equations for each country \(i \in N\) :
$$\begin{aligned} 2 G^* k_i - G^{*2} = V_i \end{aligned}$$ -
4.
We put a lower bound of 0 on \(k_i\).
-
5.
Normalize preferred SG levels for all countries to meet the global average of 4 W/m2.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Bakalova, I., Belaia, M. Stability of Efficient International Agreements on Solar Geoengineering. Environ Resource Econ 86, 673–712 (2023). https://doi.org/10.1007/s10640-023-00807-0
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10640-023-00807-0
Keywords
- Solar geoengineering
- International environmental agreements
- Side payments
- Free driver
- Global cooperation