Whatever Happened to the Aggregation Problem..

One of the biggest challenges before neoclassical economics remains the so-called “aggregation problem” raised by Piero Sraffa and Joan Robinson decades ago which makes a production function for the whole economy highly untenable.

The problem is that most neoclassical conclusions are based on a general production function for the whole economy:

Q= A f(K,L)

where Q is output, A is technology and institutional factors (assumed constant in the short-run), K is capital, and L is labor.

The problem is that whereas output can be theoretically reduced in the aggregate into units of one generic “commodity”, and labor can be reduced into simple labor, e.g., one labor-hour, or workday, capital IN THE AGGREGATE cannot be reduced into one well-defined unit. That is, physical capital, such as plant and especially equipment, varying from manual (such as shovels) to mechanical (such as trains) to electronic (such as computers), CANNOT be reduced into shovels!

If you say we can use the money value of all capital as K instead, then a bigger problem emerges for neoclassical economics: the value of capital depends on its rate of return, which depends on its quantity (due to diminishing returns), which creates an infinite loop in programming language or an element of circularity where the value of something is determined by its value, which is not saying anything at all!

As long as there is no convincing resolution to this question, production functions for the whole economy become flimsy at best, and with them a lot of the conclusions of neoclassical models, including policy implications.

Ibrahim Alloush