show the possibility of chaos. In contrast with the other papers in this collection which assume a fixed product wage in industry because of the mark-up pricing rule, they implicitly assume a fixed industrial price and a constant wage in terms of food. This implies that increased capital accumulation in the industrial sector increases the food price and the product wage, causing a profit squeeze. The current growth rate of the capital stock or accumulation rate is assumed to be an increasing function of last period's rate of profit on capital. These two relations give rise to a non-linear first order difference equation in the capital accumulation rate. This difference equation exhibits chaotic dynamics for certain parameter values. The authors argue that observed economic fluctuation in India could arise from such a deterministic system rather than from shocks. While this analysis is an interesting exercise, two aspects of the model seem questionable. The assumption that a rise in industrial demand squeezes profit margins sufficiently to reduce the rate of profit is not borne out by the evidence on mark-up pricing in Indian industry. Secondly, investment behaviour is backward looking, depending upon past profits rather than expected future returns. As a consequence firms can invest although they know (since there is no underlying uncertainty) that capacity will be unutilised. However, the general message is that chaos is an important issue, since it can arise in very simple dynamical systems. The question is whether economists can distinguish between chaotic systems and stochastic ones since the policy implications could be drastically different. Recent econometric research (Stengos, 1990) has put forward a test for discriminating between the two, so that the empirical relevance of chaos may be examined.
