ABSTRACT
What we can do and need to do is try to formulate theories that start with as full as possible a recognition of the actual facts that need to be explained, rather than ignoring the degree of unreality of the theories that we spin to give us a simple model we might prefer to reason about. The actual behavior of markets, of business decision-makers, of household consumers, of workers, and others needs to be examined first-hand, along with the perceptions of decision-makers and their possibly complex set of motives. A lot of contemporary economics is study of the logical implications of models that are based on simplifying assumptions. Starting instead from extensive research into real behavior and relationships that there is some reason to expect will continue to hold at least in the near future should give a better basis for predictions, though at best our ability to predict the future is not going to be very great in the field of economics. There are too many ways that people can do the unexpected. Indeed economic prediction cannot be made independent of the reactions to predictions. Some predictions, despite insufficient basis, can become self-fulfilling prophecies because of people's reactions to the predictions. In other instances, even well-based predictions can lead to actions that prevent their becoming true, because people did not like what they foresaw and acted to make the future different. We do need, however, to try to ascertain the most likely effects of different actions. For this purpose, the more realistic the model, the more useful its logical implications are likely to be.
