Late 1981 Romer's thesis was completed. 143 pages, most of which included very difficult math. He gives an examples of exogenous growth versus endogenous growth. During a warming trend in the Middle Ages he noted that the northern limit of farmland was had moved 100 miles to the north. this was an exogenous change, no human could have influenced this result. Endogenous change happened during those same years as people purposely planted better strains of grain to increase crop output. This change was brought about
within the system.
Romer asserted that the math that he had devised might be used throughout economics. In the theory of the firm, in asset pricing, in macroeconomic fluctuations, and so on.
It took another eighteen months to completely finalize his dissertation and it was three years after that that it was published in a journal. For those that worked on the frontier of economics it was clear that the world had changed. Suddenly somehow everyone just knew there was a great deal of new economics to be explored.
B for Will for grammatical errors.
ReplyDeleteThere's an important word and concept here: embodied.
The creation of knowledge has increasing returns to scale: one new idea begets two more, and so on.
But firms don't have increasing returns to scale - they reach a point where they can't get more efficient at using their capital by getting bigger.
The way knowledge and capital interact is that a new idea comes along, and it is used to improve a piece of capital. The idea is then embodied in the capital.
To the firm, it's just one more piece of capital. It can help them be more efficient/profitable, but it can't change the fact that they have a single efficient size that can't be exceeded. For example, no robot that can efficiently prepare fancy food is ever going to change the fact that the optimal size of a restaurant is smaller and more intimate.
But, that one idea can be embodied in the machines of more than one firm. So, each of those firms is benefiting from the increasing returns to scale in knowledge without actually having increasing returns themselves.
It's a neat idea, once you get your mind around it.