Economic Substitution Revisited: Unraveling the Firm

Traditionally, the field of Economics has used something called a “production function” to describe a relationship whereby inputs (in the simplest of models, these might simply be “labor” and “capital”) are turned into outputs (many economists refer to some abstract unit of products or services, such as “widgets”). This transformation is usually assumed to be brought about by an entrepreneur or perhaps some notion of a firm — but this aspect is not normally considered to be an input. Instead, the firm (or the entrepreneur) is rather simply assumed to be an agent of change — and the production function describes the results of this agency.

I believe this is perhaps a gross oversimplification — and I wonder whether this depiction of what is actually going on in an economy clouds our ability to recognize issues and problems that might arise.

One example — it’s the one that first started me thinking about this and got me to this idea — is the way one input might be substituted for another (this is a “technological” issue, but the basic idea is that labor and capital can be exchanged for one another [economics is actually quite abstract that way] ). This idea springs from the way in “advanced” economies, one person can use more capital to produce more output than in so-called “underdeveloped” economies (e.g. in agriculture: a large combine plus fuel produces more output per person than rather simple tools used by hand).

What got me thinking about this was an article I recently read that argued that computer technology may be eliminating skilled labor (rather than labor more characterized as “manual” — even if the less skilled work involved manual skills). The idea in the article was that skilled labor referred to overhead (such as managerial, organizational, analytical skills) might easily be performed by a computer algorithm (for example: computers can tally up data, perform statistical calculations, etc.).

I wish to entirely sidestep the question of whether computers can actually think (or manage, or organize, or analyze, etc.). Instead, it occurs to me that these are tasks that have — in my opinion — been carried out by the entrepreneur, or by a firm’s leading executives. And then I thought: “Hey, actually individuals can also use computers to replace entire firms”.

But there is no place in the production function for such an “adjustment” or substitution. This is what leads me to believe that the function of transforming inputs into output actually represents the contribution that the entrepreneur or the firm itself contributes… and that perhaps networked computers can replace entrepreneurs and firms just as much as they can replace “less skilled” workers.