Trade and environmental policy strategies in the north and south negotiation game

Abstract

In recent years, several non‐tariff trade provisions have been regarded as means of holding back transboundary environmental damages. Affected countries have then increasingly come up with trade policies to compensate for or to enforce the adoption of environmental policies elsewhere. These non‐tariff trade constraints are claimed to threaten the freedom of trading across nations, as well as the harmonization sought towards the distribution of income and policy measures. Therefore the ‘greening’ of world trade issues essentially ranges over whether there ought or ought not to be a trade‐off between trade and environmental policies. The impacts of free trade and environmental policies on major economic variables (such as trade flows, balances of trade, resource allocation, output, consumption and welfare) are thus studied here, and so is the EKC hypothesis, when such variables are played against the resulting emission levels. The policy response is seen as a political game, played here by two representative parties named North and South. Whether their policy choices, simulated by four scenarios, are right or wrong depends on their policy goals, split into economic and environmental ones.

Keywords:

• Environmental policy
• International trade
• North‐South relationship
• Economic growth and environment
• Environmental Kuznets Curve

Acknowledgments

I am thankful to Wageningen University, of The Netherlands, which provided me with the know‐how and means to carry this research out. Particular thanks to Prof. Dr Frank G. Müller, President of the Canadian Society for Ecological Economics, for his suggestions, criticisms and encouragement.

Notes

The idea here is that of a maximax strategy – the achievement of the greatest (or best) outcomes a player can avail of, irrespective to which strategy the other players choose. So, the optimality, in this case, is the economic‐wise rationality i.e. that of Pareto; the greatest (or the least, in the case of costs or damages) outcomes attained – rather than the mathematical‐wise rationality, more conservative, which is assumed to rule over game theory (maximin or minimax strategies) [1].

Or Nash‐solutions. The conservative player assumption in game theory has been criticized on the basis of its underlying pessimistic or defensive approach. Therefore, in practice, a more usual concept for game solving is that of the Nash equilibrium, through which each player chooses the best strategy (or the one that provides him with the best payoffs) given the other player’s choice [2].

XNETS^GEPM − XNETS^GEPVL; XNETN^GEPM − XNETN^GEPVL.

GEP’s scenarios closely resemble co‐operative ones, since both regions must comply with a jointly shared (global) given emission ceiling.

A strategy is called dominant in relation to another when the payoffs attained from following it are better than those arising from any others, whatever the other players’ play be [2]. Dominant strategies seek for Pareto optimality. Often though, following close dominant strategies may lead the players to a ‘prisoner’s dilemma’ game. In this case, as J. Nash showed, the solution found is not necessarily the best for all the players nor is it a Pareto optimum, since one player’s situation may be improved without worsening anyone else’s [2]. Likewise, a prisoner’s dilemma arises when there cannot be found any ‘saddle‐point solution’. As assumed here, the game is of a ‘zero‐sum’ nature: what one player wins the other loses. A ‘saddle‐point solution’ does exist when one player, behaving somewhat conservatively, seeks to get the maximum out of his minimum gains, whereas the other tries to give away only the minimum of his maximum losses. If a ‘saddle‐point equilibrium’ can be found, then the payoff arising from the maximum strategy of one player is the same arising from the minimum strategy of the other [1].

A player’s best choice (i.e. which gives him the best payoff), given the choice of the other.

QS^REPN − QS^REPS

QN^REPS − QN^REPN

XNETS^GEPM − XNETS^GEPVL; XNETN^GEPM − XNETN^GEPVL.

rs^REP − rn^REP = 1 − 0.558.