Knowledge that Transforms

The theory of the firm has been developed,by and large, in terms of static equilibrium. Even in this limited field, much of it rests on hypotheses that are rarely tested empirically. When one comes to dynamics, economists have been more preoccupied with problems of growth of an economy instead of the firm. Thus we have conjectures such as, the growth path of the economic system will pass through a turnpike, but we have hardly any conjectures as to how a small firm ends up as a big corporation. In this paper we offer a few conjectures dealing with the growth of a firm. Our inquiry is limited to only one facet of this growth process. The conjectures that we offer relate to the behavior of technological factors, namely: (1) elasticity of factor substitution, (2) technological impact on labor efficiency, and (3) biassedness of technological change. The method we have employed is based on the celebrated and censured production function. However, we have made an attempt to compromise with the non-neoclassicists. The conjectures we offer are not the result of hypothetical formulations of the assumptions and the derivation of conclusions therefrom, but are instead implications of our empirical exercises. These are based on inductive rather than deductive method. This paper suggests that all three technological factors describe an inverted U-shaped curve in the technological factor and output (or firm size) spacethe latter being the X-axis. In other words, these technological factors, contrary to the average costs, grow with the firm, reach a maximum and start falling off. These conjectures, particularly the one on the technological impact on labor efficiency, have relevance to the theory of growth recently enunciated by Nicholas Kaldor in 1967. They also suggest the reintroduction, in economic analysis, of concepts like optimum scale or Marshall's representative firm. The plan of the paper is as follows. A general production function is outlined in Section I. Like Kaldor's technical progress function, a labor efficiency progress function is suggested in Section II and its implications with respect to technological characteristics are examined in Section III. Section IV formalizes the theory into a testable hypothesis. Sample size is discussed in Section V and the empirical results presented in Section VI.

The growing demand for spectrum-communication frequencies-has once more come to outstrip its supply. Electrical interference, congestion, and resultant spectrum scarcities threaten to make air, ground, and sea travel slower and less safe; construction, distribution, mining, and manufacturing more costly; data transmission and information processing networks harder to develop for wide use; and some potential advances in education, law enforcement, and national security difficult to realize. This silent crisis seems due at least as much to deficiencies in our allocational practices as to any inherent characteristics of the radio spectrum resource. Today, rights are awarded gratis by the Federal Communications Commission and the President's Office of Telecommunications Management. With right-holders prohibited from selling any portion, they lack incentives to economize use today, to withhold current use if greater future value will result, or to transfer rights to others who may value them more highly. Large portions of spectrum are allocated to different services with no chance for interservice transfers despite the sometimes permissive policies that govern intraservice usage [24, p. 271. Some consequences of this centralized nonprice system have been aptly described by the President's Task Force on Communications Policy: