Arguments for (and against) introducing a Basic Income (i.e. a life-long, regular and unconditional payment to individuals that is sufficient for a modest but decent standard of life, i.e Pateman 2008) can be made from different perspectives:

Seen in the context of „values“ a quite convincing argument can be made, that the „Declaration of Human Rights already implies some guaranteed minimum income for each individual. From a political perspective it may either be seen as linked to democracy, to self determination or to radical social reform. Economists usually focus on the financing side and the effects of changing incentives for labor markets and growth.

Whatever perspective chosen, the evaluation of a Basic Income and its design depend crucially on its impact on welfare and happiness of individuals. Unfortunately it is quite unclear how the middle-run effects of such an institutional innovation will be:

Will the basic income succeed in permanently reducing poverty?

Will a basic income lead to a social divide, to a split society – or will it lead to more participation of the “poor”?

What are the consequences for the labor market, for labor supply, for education and consequently for innovation and growth?

How do these effects of a basic income scheme depend on its design?

The starting point of this paper is that these and similar questions cannot be understood without looking at the relation of Basic Income with relational goods. Already simple design questions underline this: If a Basic Income is received automatically and anonymously (maybe like in a negative income tax) or if one has to apply for it and to fulfill conditions will make a big difference for participation and social acceptance. How living solely of Basic Income – which may replace e.g. current unemployment schemes - will impact and social status will bear on how such a period is experienced by individuals and as a consequence how it is used for education or not and also how social relations are influenced. These features in turn will influence decisions of individuals and the consequences for economic growth and financial sustainability of the Basic Income scheme.

In this paper we make a first attempt of modeling these some decisions, focusing on the allocation of time spent for friends, education or work. As pointed out above it crucial to an understanding of a Basic Income, to take into account their relational dimensions, especially some sort of social “evaluation” of certain actions. This entails to enrich a dynamic macroeconomic model with the dynamics of “norms” governing the time allocation. These norms are in turn dependent on the individual decisions and their perceived success.

As this goal is quite ambitious we start with a simplified agent-based model. Agents are living on a cellular automaton. The relational goods are locally determined by the neighborhoods of agents. The “norms” are global “feedbacks” on the desirability of certain actions, which are derived by aggregations of the different actions of the agents and which in turn influence these decisions. Given this environment agents decide on how much of their time they allocate for friends, for knowledge and for work. These decisions are fed into a endogenous growth model, where the production function depends on technology, local public goods, determined by the relations of the agents, and labor input.

Simulations of this model with different designs of basic income schemes are then used to evaluate the difference in outcome between the different designs and to look for patterns that might be useful in evaluating basic income schemes. Obviously, the results are strongly dependent on the parameter values chosen. If the influence of the neighborhood and the norms and relational goods is set to zero the model nests a usual endogenous growth model where only the basic income basically only changes the incentives and design issues make little difference. The norms and the relational goods introduce additional feedbacks which lead to complex behavior including sensitive dependence on design for some parameter regions.

Obviously, at this point it is too early to use such exercises for design and policy issues regarding Basic Income. However, some improved version of such models may help to avoid costly policy mistakes in the future.