摘要：This paper presents a two-agent economy, in which each agent has a
consumption-dependent time preference. The optimal dynamic paths of
accumulation will tend to one of many possible steady states, depending on
the location of the initial capital level. The qualitative properties of this
economic system have been analyzed elsewhere (Tohmé and Dabús, 2000). It has
been shown that the interaction between the two agents may drag the poorest
agent towards a higher steady state, getting her out of the potential poverty
trap in which she could fall in isolation. We are interested now in studying
specific functional forms of the joint production function, the
utility functions and the psychological discount rates. The goal is to
characterize both the individual and joint steady states in order to assess
the advantages of interaction. Following the lead of (Dockner and Nishimura,
2004) we will obtain the subgame perfect equilibria of the economy seen as a
two-person non-zero sum game. We will show that the non-linear
convergence path towards the steady state examined by Tohm´e and Dab´us also
obtains in a closed-loop solution..
