About a year ago, my trade theory professor wondered about the concept of capital intensity and its application to production processes where the amount of physical capital employed is actually quite limited. The specific example he had in mind was the computer. It’s not obvious that a computer, used as a factor of production, is comparable in capital intensity to, for example, a production process that uses heavy machinery. However, if we measure capital intensity in units of value, we can make the case that the computer is actually much more capital intensive than a large, physically bulky machine.
What is capital intensity? Consider the production function Q(K,L), where K stands for capital and L stands for labor. We say that production process a is more capital intensive than production process b, if the ratio K:L is greater in process a than it is in process b. How do we define K, so that we attach some meaning to the concept of increasing K? We can think about K in terms of physical units, such that an increase in K means an increase in the physical quantity of capital. But, I’m not sure this is a good way to go about the problem, at least for a marginalist. As an Austrian, I prefer to think about K in terms of (subjective) value, so an increase in K implies an increase in the value of the capital stock being allocated towards the production of some output.
Example: take a firm in the last stage of the production process (the firm sells directly to the consumer), that employs certain quantities of K and L. We can conceive of a situation where the capital intensity of the firm’s production changes, strictly assuming that the physical quantities of K and L remain the same. Assume also that K and L are perfectly non-specific, meaning they could just as well be employed towards the production of various alternative outputs. Suppose that the marginal value of L falls for whatever reason (say, an increase in the aggregate supply of labor), and the physical K:L ratio remains unchanged. We can argue that the firm’s production is now more capital intensive, because the value of K remains the same while the value of L falls.
Going back to the computer, assume a firm must employ a certain value ratio, K:L, to produce an additional unit of output. The firm can accumulate the necessary value of K by employing one computer and a printer, or ten large, bulky movable type machines. Either of these choices imply the same degree of capital intensity. Yet, without considering value, I think many of us would at first think that the latter is more capital intensive than the former. It’s hard to think of a production process consisting of a person, her computer, and her printer as more, or equally, capital intensive than a process where five people are using ten large machines.
We can say that the capital intensity of U.S. production has increased over time. The U.S. is expected to specialize in relatively capital intensive production. But, this doesn’t seem to square with the fact that many manufacturing processes are being moved to countries with inferior capital accumulation. The manufacturing sector, as a share of total value produced, has been declining, as it is increasingly encroached on by the service sector. The interesting thing is that we could argue that many service sector processes are more capital intensive than their kin in the manufacturing sector (although, clearly, this isn’t universally true), even though at first this seems counter-intuitive. The type of capital employed in the service sector is typically of very high value — and to physical capital we also have to add human capital, which in the service sector is often many times greater than in manufacturing —, even if physically the opposite seems true.