​

Consensus and disagreements between China and the U.S. are the key factors influencing the direction that global climate negotiation is heading for. By taking into account the uncertainties of temperature increment and its impact on GDP growth, together with the positive, negative and spillover effects of climate change investment on utility, a strategic simulation model including China and the U.S. is developed. Based on utility and game theory, a sensitivity analysis is conducted. The results show that the first-mover disadvantage exists in the game, and the scale of each country’s climate change investment under non-cooperative scenario is too small to ensure the 2°C target. To guarantee the stability and win-win basis for global cooperation, the simulation results also indicate that it makes sense to assist and compensate technology transfer and funding to China.

​

==Keywords==

​

climate negotiation ; utility ; game theory ; strategy

​

==1. Introduction==

​

Owing to the Doha Round talks, the second stage of the Kyoto Protocol succeeds to continue. But in reality, the countries with binding commitments only contribute about 15.0% of global greenhouse gas (GHG) emissions, while the top two economic entities, i.e., the U.S. and China, release about 41.8% of global GHG annually. With a 32.8% proportion on the global GDP (purchasing power parity-adjusted, 2010) [ [[#bb0045|IEA, 2012]] ], the consensus and disagreements between China and the U.S. may greatly influence the trend in climate negotiation. The U.S. currently refuses to sign the Kyoto Protocol and emphasizes that its reduction responsibility should be based on mutual commitments including developing countries. China and other developing countries insist on the principal of common but differentiated responsibilities (CBDR), and persuade the countries with historical responsibilities to take action immediately. The potential restrictions and economic opportunities hidden behind the disagreements between the U.S. and China are essential. For the investigation of strategic actions between the U.S. and China, the application of game theory is not only necessary for a better understanding of the current global climate negotiation predicament, but also instructive for the design of a better international negotiation framework [ [[#bb0025|''Eyckmans and Tulkens,''  2003                                ]] ].      

​

[[#bb0130|''Wood''  [2011]                                ]]  has already pointed out the free-rider phenomenon behind the climate negotiation from a perspective of public goods. The externalities of GHG effects provide the basis for the game research on climate negotiation. Currently, most studies focus on the theoretical amendment to solve the problem of cooperation failure, including [[#bb0085|''Ostrom''  [2010]                                ]]  and [[#bb0145|''Zhang et al.''  [2011]                                ]] , but few researches discuss the specific individual’s decision procedure except for [[#bb0040|''Hovi et al.''  [2012]                                ]] , who explain why the U.S. quit the Kyoto Protocol. In order to better understand the hidden purposes and strategies of the main emitters in climate negotiation, and the predictable results of such strategies, it’s necessary to combine the theoretical model with simulations. This article will not only build a strategic simulation model including the top two global emitters, but also investigate the existence of first-mover disadvantage under climate negotiation and the possibility of staying below 2°C increment.      

​

==2. Utility functions of the U.S. and China==

​

Assume that the national utility functions of the U.S. and China meet the standards of constant relative risk aversion (CRRA), and the utility ''U''<sub>t</sub>  at time t can be connected with the homochromous consumption ''C<sub>t</sub>'' , i.e.,

​

{| class="formulaSCP" style="width: 100%; text-align: center;" 

|-

| 

{| style="text-align: center; margin:auto;" 

|-

| <math>U_t=\begin{array}{lll}

\frac{C_t^{1-n}}{1-\eta }, & if & \eta \not =1;\\

ln\left(C_t\right), & if & \eta =1\mbox{.}

\end{array}</math>

|}

| style="width: 5px;text-align: right;white-space: nowrap;" | ( 1)

|}

​

''η''  represents the level of risk aversion. The consumption ''C<sub>t</sub>''  at time ''t''  can be further expressed as ''C''<sub>''t''</sub>  = exp(∫<sub>0</sub><sup>''t''</sup>''g''<sub>''s''</sub> d''s'' ) by standardizing the initial cosumption (Here, ''go'' ,CHN and ''go'' ,US represent the stable GDP growth of China and the U.S., respectively.) The respective growth rates are 4.5% and 1.5% with a PPP-adjusted reference of global 2.5% annual GDP growth as given by [[#bb0090|''Pindyck''  [2012]                                ]] , which means that the gross GDP of China will overtake the U.S. by 2023. This assumption is in accordance with the years of 2039 given by [[#bb0125|''Wilson and Purushothaman''  [2003]                                ]] , 2031 by [[#bb0135|''Wu''  [2010]                                ]] , and 2016 by the OECD [ [[#bb0060|''Johansson et al.,''  2012                                ]] ].      

​

The greenhouse effect can also affect the utilities of each individual. [[#bb0015|''Dell et al.''  [2009]                                ]]  gave a simplified relation between the GDP growth ''g<sub>t</sub>''  and a temperature increment ''T<sub>t</sub> ,                            ''  i.e., ''g<sub>t</sub> = g<sub>0</sub>  — γT<sub>t</sub>'' , where ''γ''  is the ratio of the marginal effect. Given the temperature increment ''T<sub>H</sub>''  at year ''H'' , the changing process of the temperature increment is calculated by ''T<sub>t</sub>''  = 2''T<sub>H</sub>''  × [1 — (1/2)''<sup>t/H</sup>'' ]. However, the temperature increment ''T''<sub>H</sub>  at year H is uncertain. [[#bb0090|''Pindyck''  [2012]                                ]]  used the three-parameter Gamma function (''r<sub>T</sub>'' = 3.8. λ''<sub>T</sub>'' = 0.92, ''θ<sub>T</sub>'' =–1.13) to simulate the distribution of ''T<sub>H</sub>''  based on the research results of the [[#bb0050|''IPCC''  [2007]                                ]] . This method is used in our research, where the probability density function can be expressed as

​

{| class="formulaSCP" style="width: 100%; text-align: center;" 

|-

| 

{| style="text-align: center; margin:auto;" 

|-

| <math>f_T\left(x;r_T{\lambda }_T,{\theta }_T\right)=\frac{{\lambda }_T^{r_T}}{\Gamma \left(r_T\right)}{\left(x-{\theta }_T\right)}^{r_T-1}e^{-{\lambda }_T\left(x-{\theta }_T\right)},</math>

|-

|<math>x\geq {\theta }_T\mbox{.}</math>

|}

| style="width: 5px;text-align: right;white-space: nowrap;" | 

|}

​

With the theoretical linear relationship between the marginal effect parameter ''γ''  and the loss function ''L(T)''  = exp(—''βT''<sup>2</sup> ), the three-parameter Gamma distribution (parameters: ''r<sub>γ</sub> , λ<sub>γ</sub> , θ<sub>γ</sub>'' ) can be applied to simplify the theoretical model based on the research results of integrated assessment models (IAMs) [ [[#bb0070|Nordhaus, 1994]]  and [[#bb0035|Hope, 2006]] ]. [[#t0005|Table 1]]  summarizes the GDP loss ratios of the U.S. and China under different temperature increments. Compared to the global reference, the variance of loss under the same temperature increment follows the order: China>world>U.S., which is in accordance with the economic structure and the resistance ability to natural disasters of each country. In this research the assumptions of the global y as given by [[#bb0090|''Pindyck''  [2012]                                ]]  and [[#bb0020|''Dietz and Stern''  [2008]                                ]]  are taken as the references. Here, we assume the loss of the mean, 17% and 83% proportion of the cumulative loss function under a 2.5°C increment for China (2.00%, 1.00% and 4.80%) and the U.S. (1.20%, 0.75% and 1.40%). [[#t0010|Table 2]]  further lists the respective parameters of the world, the U.S. and China. [[#f0005|Figure 1]]  illustrates the parameter ''y''  distributions. In order to enhance the robustness of our theoretical model, a sensitivity analysis of ''y''  is further carried out in this study.

​

<span id='t0005'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 1.

​

GDP loss ratios of the world, China and the U.S. for different temperature increments

​

|-

​

! Reference

! Temperature increment (°C)

! Effect to global GDP (%)

! Effect to GDP of China (%)

! Effect to GDP of U.S. (%)

|-

​

| [[#bb0030|''Fankhauser''  [1995]                                                            ]]

| 2.5

| –1.4

| –4.7

| –0.7

|-

​

| [[#bb0100|''Richard''  [1995]                                                            ]]

| 2.5

| –1.9

| –8.6[[#tf0005|<sup>a</sup>]] , –5.2[[#tf0010|<sup>b</sup>]]

| –1.5[[#tf0015|<sup>c</sup>]]

|-

​

| [[#bb0070|''Nordhaus''  [1994]                                                            ]]

| 3.0

| –1.33

| –1.0

| 

|-

​

| [[#bb0075|''Nordhaus and Yang''  [1996]                                                            ]]

| 2.5

| –1.7

| –2.4[[#tf0020|<sup>d</sup>]]

| 

|-

​

| [[#bb0095|''Plambeck and Hope''  [1996]                                                            ]]

| 2.5

| –1.3 (–0.6 to –5.1)

| –4 (–1 to –4.8)[[#tf0025|<sup>e</sup>]]

| –1.2 (–1.4 to –0.8)

|-

​

| [[#bb0080|''Nordhaus and Boyer''  [2000]                                                            ]]

| 2.5

| –1.50, –1.88[[#tf0030|<sup>f</sup>]]

| –0.22

| –0.45

|-

​

| [[#bb0115|''Tol''  [2002]                                                            ]]

| 1.0

| –2.3 (1.0)

| –2.1 (5.0)[[#tf0035|<sup>g</sup>]]

| –3.4 (2.1)[[#tf0040|<sup>h</sup>]]

|-

​

| [[#bb0065|''Maddison''  [2003]                                                            ]]

| 2.5

| –0.1

| 

| –0.3

|-

​

| [[#bb0035|''Hope''  [2006]                                                            ]]

| 2.5

| –0.9 (–0.2 to 2.7)

| –2.6 (–0.4 to 10.0)[[#tf0045|<sup>i</sup>]]

| 

|}

​

Notes: Data in brackets represent the 95% confidence interval;

​

a. including south and southeast Asia;

​

b. referring to the centralized countries;

​

c. including the U.S. and Canada;

​

d. including developing countries;

​

e. including other Asian countries;

​

f. the former is weighted by GDP, and the latter is weighted by population;

​

g. referring to the centralized countries;

​

h. referring to the American OECD countries except for Mexico;

​

i. referring to Asian countries except China

​

<span id='t0010'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 2.

​

GDP loss ratios distributions and parameters for the world, China and the U.S. under a 2.5°C increment

​

|-

​

! rowspan="2" | District

! colspan="3" | GDP loss proportion under 2.5°C increment(%)

! colspan="3" | Three parameters of Gamma

|-

​

! Mean

! 17% distribution proportion

! 83% distribution proportion

! ''r<sub>γ</sub>''

! ''λ<sub>γ</sub>''

! ''θ<sub>γ</sub>''

|-

​

| World

| 1.92

| 0.63

| 3.15

| 3.96

| 19,645

| –6.26×10<sup>–5</sup>

|-

​

| U.S.

| 1.10

| 0.75

| 1.40

| 103.80

| 411,600

| –1.75×10<sup>–4</sup>

|-

​

| China

| 3.00

| 1.00

| 4.80

| 11.14

| 22,329

| –2.85×10<sup>–4</sup>

|}

​

Notes: The GDP loss distribution of the World is adjusted according to the results by [[#bb0090|''Pindyck''  [2012]                                                ]]  and [[#bb0020|''Dietz and Stern''  [2008]                                                ]]

​

<span id='f0005'></span>

​

{| style="text-align: center; border: 1px solid #BBB; margin: 1em auto; max-width: 100%;" 

|-

|

​

​

[[Image:draft_Content_513786166-1-s2.0-S1674927814500056-gr1.gif|center|307px|Cumulative probability distribution of parameterγfor the world, China and the ...]]

​

​

|-

| <span style="text-align: center; font-size: 75%;">

​

Figure 1.

​

Cumulative probability distribution of parameter ''γ''  for the world, China and the U.S.                  

​

</span>

|}

​

The externality of GHG effects will affect the utilities in three ways:

​

First, the increase of climate change investment will lead to the decrease of disposable consumption, which will have negative impacts on the utility.

​

Second, the increase of climate change investment will boost the GDP growth, as it is a capital investment into the economy. Such a positive effect has already been confirmed by the EU [ [[#bb0055|''Jaeger et al.,''  2011                                ]] ].      

​

Taking China as an example, if ICHN represents the proportion of climate change investment to its GDP, then the new disposable consumption and new GDP growth as the result of negative and positive effects should be modified as:

​

{| class="formulaSCP" style="width: 100%; text-align: center;" 

|-

| 

{| style="text-align: center; margin:auto;" 

|-

| <math>C_{CHN,t}^\ast =C_{CHN,t}\times \left(1-I_{CHN}\right),</math>

|-

|<math>g_{CHN,t}^\ast =g_{0,CHN}\times \left(1+{\xi }_{CHN}\times I_{CHN}\right)-</math><math>{\gamma }_{CHN}T_t\mbox{,}</math>

|}

| style="width: 5px;text-align: right;white-space: nowrap;" | 

|}

​

where ξ<sub>CHN</sub>  represents the multiplier of ''I''<sub>CHN</sub>  to the GDP growth.      

​

And third, the spillover effect is the illustration of public goods. Assume that the global temperature increment ''T''<sub>up</sub>  can be controlled by the global investment ''I''<sub>world</sub>  in the relation ''T''<sub>up</sub>  = Φ(''I''<sub>world</sub> ), which can be further decomposed into each country’s investment, i.e., ''T''<sub>up</sub>  = Φ(''ω'' CHN × ''I'' CHN = ''ω'' US × ''I'' US + others). And ''ω'' CHN, ''ω'' US are the weighting factors of the country’s GDP. The PPP-adjusted GDP of China and the U.S. in 2010 are 10.16 trillion and 14.59 trillion international dollar each. It means the weighting factors are 13.26% and 19.02%, respectively.      

​

==3. Theoretical optimization procedure of game strategies between the U.S. and China==

​

For China, the net present gross utility can be simplified as ''U'' CHN = ''Ψ'' CHN(''U'' CHN, ''I'' US) by involving the three effects of climate change investments, i.e.,

​

{| class="formulaSCP" style="width: 100%; text-align: center;" 

|-

| 

{| style="text-align: center; margin:auto;" 

|-

| <math>U_{CHN}={\int }_0^{+\infty }{\int }_{{\theta }_{\gamma -CHN}}^{+\infty }{\int }_{{\theta }_T}^{T_{up}}U_{CHN,t}^\ast \times exp\left(-\right. </math><math>\left. {\delta }_{CHN}\times t\right)\times {\tilde{f}}_T\left(T;,,r_T;,,{\lambda }_T;,,{\theta }_T\right)\times f_{\gamma -CHN}\left(\gamma ;,,r_{\gamma -CHN};,,{\lambda }_{\gamma -CHN};,,{\theta }_{\gamma -CHN}\right)dTd\gamma dt</math>

|}

| style="width: 5px;text-align: right;white-space: nowrap;" | 

|}

​

The utility at time ''t''  for China (''U''<sub>CHN,''t''</sub><sup>⁎</sup> ) and the Bayesian-modified probability density function of the temperature increment (<math display="inline">{\tilde{f}}_T</math> ) can be expressed as:

​

{| class="formulaSCP" style="width: 100%; text-align: center;" 

|-

| 

{| style="text-align: center; margin:auto;" 

|-

| <math>U_{CHN,t}^\ast =\frac{{\left\{\left(1-I_{CHN}\right)\times exp\left[{\int }_0^tg_{0,CHN}\times \left(1+{\xi }_{CHN}\times I_{CHN}\right)ds-{\int }_0^t2T_t\times {\gamma }_{CHN}\times \left(1-2^{-s/H}\right)ds\right]\right\}}^{-1{\eta }_{CHN}}}{1-{\eta }_{CHN}}</math>

|-

|<math>{\tilde{f}}_T\left(x;,,r;,,\lambda ;,,\theta \right)=</math><math>\frac{f_T\left(x;,,r;,,\lambda ;,,\theta \right)}{F\left(T_{up}\right)},\theta \leq x\leq T_{up},\quad T_{up}=</math><math>\Phi \left(I_{CHN},I_{US}\right),F\left(T_{up}\right)=</math><math>{\int }_{{\theta }_r}^{T_{up}}f_T\left(x;,,r;,,\lambda ;,,\theta \right)dx\mbox{.}</math>

|}

| style="width: 5px;text-align: right;white-space: nowrap;" | 

|}

​

According to the hypothesis of rationality, and given the information that the proportion of the U.S. investment is ''I''<sub>US</sub> , the conditional optimal strategy for China is to select the optimal scale of climate change investment to maximize its utility, i.e., <math display="inline">\underset{I_{CHN}}{Max}{\psi }_{CHN}\left(I_{CHN},I_{US}\right)</math> . Then the conditional optimal strategy can be expressed as ''O'' CHN, which is closely connected with the exogenous information ''I'' US, i.e., ''O'' CHN = ''Θ'' CHN(''I'' US). Similarly, the conditional optimal strategy for the U.S. can also be expressed as ''O'' US = ''Θ'' US (''I'' chn).      

​

For the non-cooperative scenario, the dynamic game strategies involve three types: China moves first, China moves second, and China and the U.S. move simultaneously. Taking the first type as an example, the dynamic optimal strategy of China is to maximize its utility. Under the dynamic situation, China should take the subjective initiative of the U.S. into consideration, which means that the U.S.’s investment information should be''Θ'' US(''I'' US) rather than ''I'' US. Hence, the dynamic optimal strategy for China and the maximized utility can be expressed as ''O''<sub>CHN</sub><sup>first</sup>  and ''U''<sub>CHN</sub><sup>first</sup>  using the following equation: Max ''Ψ'' CHN(''I'' CHN, ''Θ'' US(''I  '' CHN)). Then the dynamic optimal strategy for the U.S. and the corresponding utility can be calculated as <math display="inline">\underset{I_{CHN}}{Max}{\psi }_{CHN}\left(I_{CHN},{\Theta }_{US}\left(I_{CHN}\right)\right)</math> . The theoretical optimization procedures for the second and third types of game strategies are similar.      

​

For the cooperative scenario, the goal is to maximize the gross utility under the 2°C target and to define the optimal global allocation of investment assignments. [[#t0015|Table 3]]  summarizes the theoretical equilibrium results of different game strategies.

​

<span id='t0015'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 3.

​

Game equilibrium results under different strategies

​

|-

​

! rowspan="2" | Country

! colspan="3" | Non-cooperative game

! rowspan="2" | Cooperative game

|-

​

! China moves first

! China moves second

! Simultaneously

|-

​

| China

| ''O''<sub>CHN</sub><sup>first</sup> , ''U''<sub>CHN</sub><sup>first</sup>

| ''O''<sub>CHN</sub><sup>second</sup> , ''U''<sub>CHN</sub><sup>second</sup>

| ''O''<sub>CHN</sub><sup>simul</sup> , ''U''<sub>CHN</sub><sup>simul</sup>

| ''O''<sub>CHN</sub><sup>coop</sup> , ''U''<sub>CHN</sub><sup>coop</sup>

|-

​

| U.S.

| ''O''<sub>US</sub><sup>second</sup> , ''U''<sub>US</sub><sup>second</sup>

| ''O''<sub>US</sub><sup>first</sup> , ''U''<sub>US</sub><sup>first</sup>

| ''O''<sub>US</sub><sup>simul</sup> , ''U''<sub>US</sub><sup>simul</sup>

| ''O''<sub>US</sub><sup>coop</sup> , ''U''<sub>US</sub><sup>coop</sup>

|}

​

==4. Simulation of the game theory and sensitivity analysis between the U.S. and China==

​

[[#t0020|Table 4]]  gives the initial setting of parameters for the simulation of the game theory between the U.S. and China. Apart from the GDP growth ''g''<sub>0</sub> , weighting factor ''ω'' , the distribution of temperature increment ''T'' , and the probability density of parameter ''γ''  discussed above, the discounting factor ''δ''  and risk aversion ''η''  still need to be calibrated. While [[#bb0005|''Conceicao et al.''  [2007]                                ]]  and [[#bb0010|''Dasgupta''  [2007]                                ]]  have different views, considering the inter-generation unfairness [ [[#bb0010|Dasgupta, 2007]]  and [[#bb0110|Stern and Britain, 2006]] ] and proper social discounting ratio, the discounting factor and risk aversion are set as ''δ'' = 0, ''η'' = 2, which is in accordance with the results of  [[#bb0105|Seda and Tran, 2012]]  and [[#bb0150|Zhang, 2011]]  and [[#bb0140|''Yang and Tao'' [2011]                                ]] . In addition, ''ξ''  can be theoretically proven to be equal to 1 if the climate change investment is treated as the classic investment[[#fn0005|<sup>①</sup>]] .

​

<span id='t0020'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 4.

​

Initial parameters for the simulation of the China–U.S. strategic game model

​

|-

​

! Parameter

! China

! U.S.

|-

​

| ''g''<sub>o</sub>

| 4.5%

| 1.5%

|-

​

| ''ω''

| 13.26%

| 19.02%

|-

​

| (''r<sub>γ</sub> , λ<sub>γ</sub> , θ<sub>γ</sub>'' )                                                    

| (11.14, 2.23×10<sup>4</sup> , –2.85×10<sup>–4</sup> )                                                    

| (103.80, 4.12×10<sup>5</sup> , –1.75×10<sup>–4</sup> )                                                    

|-

​

| (''η, δ, ξ, H'' )                                                    

| (2, 0, 1, 100)

| (2, 0, 1, 100)

|-

​

| (''r<sub>T</sub> , λ<sub>T</sub> , θ<sub>T</sub>'' )                                                    

| (3.80, 0.92, –1.13)

| (3.80, 0.92, –1.13)

|}

​

With the initial parameters calibrated, [[#t0025|Table 5]]  further illustrates the results of the game simulations. First, both China and the U.S. tend to choose the second-move strategy (China: –21.77453 < –21.77451; U.S.: –66.47602 < –66.47596), which is similar to the actual situation in the climate negotiations. Both countries are trying to persuade the other to take the lead in reduction and to put off its own responsibilities.

​

<span id='t0025'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 5.

​

Simulation results of the dynamic game model

​

|-

​

! rowspan="2" | Country

! colspan="3" | Non-cooperative game

! rowspan="2" | Cooperative game

|-

​

! China moves first

! China moves second

! Simultaneously

|-

​

| China

| 1.3419% [–21.774532]

| 1.3368% [–21.774509]

| 1.3405% [–21.774532]

| 3.7520% [–21.794039]

|-

​

| U.S.

| 0.4548% [–66.475957]

| 0.5601% [–66.476019]

| 0.4547% [–66.475956]

| 1.2747% [–66.444107]

|}

​

Notes: Data in brackets indicate the utility level. The EU is assumed to take the corresponding burden of climate change investment for the global 2°C target

​

Second, the non-cooperative scenario can hardly meet the necessary requirement of the 2°C target. In other words, the results also demonstrate the importance and necessity of cooperation in global climate negotiations.

​

Finally, the simulation also demonstrates that the U.S. rather than China can benefit from the cooperation scenario. China’s utility under the cooperation scenario is smaller than that under the noncooperative scenario, but the situation is different for the U.S. The reason for this is that China will have higher shares in emission reduction for maximizing the global utility, compared to the proportion under the non-cooperative scenario. China will then contribute much more to the spillover effect than receiving from it, which is quite different for the U.S. Compared to the non-cooperative scenario, China should increase the climate change investment proportion by 2.41% for the cooperative scenario, while the U.S. should only increase it by 0.82%. Without any compensation mechanisms in place, this difference makes the cooperative framework quite unstable. Therefore, necessary assistance for the developing countries, including funds and technology transfer, is of great importance. The utility differences between the cooperative and non-cooperative scenario for the U.S. and China can further confirm and explain the U.S.’s negotiation strategies. The U.S. emphasizes that its emission reduction should be bound on the condition that large developing countries participate in the allocation of the reduction targets. The U.S. prefers to establish a new platform to redistribute the reduction responsibilities beneficial for himself, and to essentially restrict the rights of emerging countries. So the developing countries need to push for their rights and have to appeal for reasonable compensation.

​

In order to reach more robust conclusions, [[#t0030|Table 6]]  gives the results of the sensitivity analysis based on the initial parameter settings ([[#t0020|Table 4]] ). All results support the three key conclusions listed in [[#t0025|Table 5]]  generally, which are: the first-mover disadvantage phenomenon does exist in the climate negotiation between the U.S. and China; it is impossible for the world to stay below the 2°C increment under the noncooperative scenario; the compensation mechanisms are quite necessary for the realization of a successful cooperation in climate change negotiations.

​

<span id='t0030'></span>

​

{| class="wikitable" style="min-width: 60%;margin-left: auto; margin-right: auto;"

|+

​

Table 6.

​

Sensitivity analysis and the influences on the main conclusions

​

|-

​

! Conclusion

! Original model

! Higher ''g''<sub>0</sub>

! Lower ''g''<sub>0</sub>

! Higher ''η''

! Lower ''η''

! Increase variance of ''γ''

! Decrease variance of ''γ''

! Shift of ''γ''

! Higher ''δ''

! Lower ''ξ''

|-

​

| Q1

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

|-

​

| Q2

| N

| N

| N

| N

| N

| N

| N

| N

| N

| N

|-

​

| Q3

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

| Y

|}

​

Notes: Q1 indicates the existence of first-mover disadvantage; Q2 indicates the possibility to stay below the 2°C target under a non-cooperative scenario; Q3 indicates the necessity of compensation mechanisms to enhance the cooperation between the U.S. and China

​

==5. Conclusions and policy implications==

​

The strategic simulation model based on game theory together with the sensitivity analysis gives us a new perspective to reexamine the current climate negotiation and the U.S. strategies involved. The results indicate the necessity of compensation mechanisms, without which the 2°C target can hardly be reached whatever the strategies under the non-cooperative scenario would be.

​

Specifically for China, it will be reasonable and necessary to acquire technology and funding assistance, besides insisting on the principal of CBDR. On the one hand, the strategies under the non-cooperative scenario are also meaningful if the negotiations are deadlocked. On the other hand, assisting and funding the developing countries for reduction is also a favorable option for the U.S. if only an enforceable cooperation framework can be reached among the major emitters.

​

==Acknowledgements==

​

This study is supported by the Natural Science Foundation (No. 71273153) and the National Key Technology R&D Program (No. 2012BAC20B04). The authors appreciate the helpful suggestions and comments of the anonymous reviewers and the editors.

​

==References==

​

<ol style='list-style-type: none;margin-left: 0px;'><li><span id='bb0005'></span>

[[#bb0005|Conceicão et al., 2007]] P. Conceicão, Y. Zhang, R. Bandura; Brief on discounting in the context of climate change economics; Human Development Report Office occasional paper (2007), pp. 1–19</li>

<li><span id='bb0010'></span>

[[#bb0010|Dasgupta, 2007]] P. Dasgupta; The Stern Review’s economics of climate change; National Institute Economic Review, 199 (2007), p. 4</li>

<li><span id='bb0015'></span>

[[#bb0015|Dell et al., 2009]] M. Dell, B.F. Jones, B.A. Olken; Temperature and income: Reconciling new crosssectional and panel estimates; The American Economic Review, 99 (2) (2009), pp. 198–204</li>

<li><span id='bb0020'></span>

[[#bb0020|Dietz and Stern, 2008]] S. Dietz, N. Stern; Why economic analysis supports strong action on climate change: A response to the Stern Review’s critics; Review of Environmental Economics and Policy, 2 (1) (2008), pp. 94–113</li>

<li><span id='bb0025'></span>

[[#bb0025|Eyckmans and Tulkens, 2003]] J. Eyckmans, H. Tulkens; Simulating coalitionally stable burden sharing agreements for the climate change problem; Resource and Energy Economics, 25 (4) (2003), pp. 299–327</li>

<li><span id='bb0030'></span>

[[#bb0030|Fankhauser, 1995]] S. Fankhauser; Valuing Climate Change: The Economics of the Greenhouse; EarthScan (1995), pp. 66–90</li>

<li><span id='bb0035'></span>

[[#bb0035|Hope, 2006]] C. Hope; The marginal impact of CO<sub>2</sub>  from PAGE2002: An integrated assessment model incorporating the IPCC’s five reasons for concern                                        ; Integrated Assessment Journal, 6 (1) (2006), pp. 19–56</li>

<li><span id='bb0040'></span>

[[#bb0040|Hovi et al., 2012]] J. Hovi, D.F. Sprinz, G. Bang; Why the United States did not become a party to the Kyoto Protocol: German, Norwegian, and US perspectives; European Journal of International Relations, 18 (1) (2012), pp. 129–150</li>

<li><span id='bb0045'></span>

[[#bb0045|IEA, 2012]] IEA, 2012[2013-02-02]: Key world energy statistics. Accessed http://www.iea.org/publications/ .                                    </li>

<li><span id='bb0050'></span>

[[#bb0050|IPCC, 2007]] IPCC; Climate Change 2007: The Physical Science Basis; S. Solomon (Ed.),  ''et al.'' , Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press (2007), pp. 98–679                                        </li>

<li><span id='bb0055'></span>

[[#bb0055|Jaeger et al., 2011]] C.C. Jaeger, L. Paroussos, D. Mangalagiu,  ''et al.''; A New Growth Path for Europe, University of Oxford, ICCS, Université Paris (2011), pp. 1–27</li>

<li><span id='bb0060'></span>

[[#bb0060|Johansson et al., 2012]] Johansson, A, Y. Guillemette, F. Murtin, et al., 2012[2013-02-02]: Looking to 2060: Long-term global growth prospects. Accessed http://www.astrid.eu/La-produtt/Studi-ric/OECD_2060-policy-paper-FINAL_11_2012.pdf</li>

<li><span id='bb0065'></span>

[[#bb0065|Maddison, 2003]] D. Maddison; The amenity value of the climate: The household production function approach; Resource and Energy Economics, 25 (2) (2003), pp. 155–175</li>

<li><span id='bb0070'></span>

[[#bb0070|Nordhaus, 1994]] W.D. Nordhaus; Expert opinion on climatic change; American Scientist, 82 (1) (1994), pp. 45–51</li>

<li><span id='bb0075'></span>

[[#bb0075|Nordhaus and Yang, 1996]] W.D. Nordhaus, Z. Yang; A regional dynamic general-equilibrium model of alternative climate-change strategies; The American Economic Review (1996), pp. 741–765</li>

<li><span id='bb0080'></span>

[[#bb0080|Nordhaus and Boyer, 2000]] W.D. Nordhaus, J.G. Boyer; Warming the World: The Economics of the Greenhouse Effect; MIT Press (2000), pp. 76–128</li>

<li><span id='bb0085'></span>

[[#bb0085|Ostrom, 2010]] E. Ostrom; Polycentric systems for coping with collective action and global environmental change; Global Environmental Change, 20 (4) (2010), pp. 550–557</li>

<li><span id='bb0090'></span>

[[#bb0090|Pindyck, 2012]] R.S. Pindyck; Uncertain outcomes and climate change policy; Journal of Environmental Economics and Management, 63 (3) (2012), pp. 289–303</li>

<li><span id='bb0095'></span>

[[#bb0095|Plambeck and Hope, 1996]] E.L. Plambeck, C. Hope; PAGE95: An updated valuation of the impacts of global warming; Energy Policy, 24 (9) (1996), pp. 783–793</li>

<li><span id='bb0100'></span>

[[#bb0100|Richard, 1995]] S.J.T. Richard; The damage costs of climate change toward more comprehensive calculations; Environmental and Resource Economics, 5 (4) (1995), pp. 353–374</li>

<li><span id='bb0105'></span>

[[#bb0105|Seda and Tran, 2012]] K.C. Seda, C. Tran; Temptation and social security in a dynastic framework; European Economic Review, 56 (7) (2012), pp. 1422–1445</li>

<li><span id='bb0110'></span>

[[#bb0110|Stern and Britain, 2006]] N.H. Stern, G. Britain; Stern Review: The Economics of Climate Change, HM Treasury (2006), pp. 179–298</li>

<li><span id='bb0115'></span>

[[#bb0115|Tol, 2002]] R. Tol; Estimates of the damage costs of climate change. Part 1: Benchmark estimates; Environmental and Resource Economics, 21 (1) (2002), pp. 47–73</li>

<li><span id='bb0120'></span>

[[#bb0120|Weitzman, 2009]] M. Weitzman; Additive damages, fat-tailed climate dynamics, and uncertain discounting; Economics: The Open-Access, Open-Assessment E-Journal, 3 (39) (2009), pp. 1–29</li>

<li><span id='bb0125'></span>

[[#bb0125|Wilson and Purushothaman, 2003]] Wilson, D., and R. Purushothaman, 2003[2013-02-02]: Dreaming with BRICs: The path to 2050. Accessed http://www.atlanticfinancial.com/global-investing/emerging-markets-investing/goldman-sachs-bric-research-paper-2050.pdf .                                    </li>

<li><span id='bb0130'></span>

[[#bb0130|Wood, 2011]] P.J. Wood; Climate change and game theory; Annals of the New York Academy of Sciences, 1219 (1) (2011), pp. 153–170</li>

<li><span id='bb0135'></span>

[[#bb0135|Wu, 2010]] Q.-J. Wu; Forcasting the change in economic strength of China and America; East China Economic Management (in Chinese), 24 (12) (2010), pp. 56–59</li>

<li><span id='bb0140'></span>

[[#bb0140|Yang and Tao, 2011]] Y. Yang, Y.-C. Tao; Estimation of proper Chinese international reserves 1994–2009; Studies of International Finance (in Chinese), 6 (2011), pp. 4–13</li>

<li><span id='bb0145'></span>

[[#bb0145|Zhang et al., 2011]] F.-S. Zhang, Z. Xu, J.-K. He,  ''et al.''; Gametheoretical study on international transfer of low-carbon technologies under incomplete information; Advances in Climate Change Research (in Chinese), 7 (1) (2011), pp. 41–47</li>

<li><span id='bb0150'></span>

[[#bb0150|Zhang, 2011]] Zhang, T., 2011[2013-02-02]: Endogenous discounting, precautionary savings and the current account: The case of China. Accessed http://halshs.archivesouvertes.fr/docs/00/55/69/79/PDF/2007.31.pdf .                                    </li>

</ol>

​

== Notes ==

​

<span id='fn0005'></span>

​

[[#bfn0005|①]]. ''g''  = ''ΔY/Y, Y''  indicates the GDP. If the climate change investment is treated as the classic investment, then ''g*''  = (''ΔY + I × ΔY)/Y''  = (1 + I) × ''g''

​