Economic conditions change with time and in space. We raise a few modern questions on economic policy: delocalization, globalization, cycles … We have developed a highly simplified agent based model providing a stylized geographical type of framework in order to tackle such questions. We investigate some of these and report quantitative results similar to qualitative features. The model contains several parameters: lattice size(s), lattice symmetry (or symmetries), initial concentration(s), economic field time sequence(s), selection pressure, diffusion process rule(s), enterprise-enterprise "interaction"(s), business plan(s) and enterprise evolution law(s), economy policy time delay implementation. It is found that the model even in its simple forms can lead to a large variety of situations, including: stationary solutions and cycles, but also chaotic behavior. Warnings on the model simplicity are in order, but also suggestions for control and development of economic strategies.