Joan Robinson was “probably the best economist alive” in Paul Samuelson’s own words.

@zachdcarter rightfully celebrated her work on monopolistic competition (without which there wouldn’t be new trade theory nor endogenous growth theory) & #monpsony.

Dismissing her work...

...on #capital theory is wrong-headed, because it has fundamental implications for the theory of #distribution.

She realized that there is no “quantity of capital” independent of prices in a world with heterogeneous capital goods.

The existence of...

...such quantity of capital is *crucial* for the #neoclassical theory of distribution, which rests fundamentally on the possibility that price movements will ensure that f’(k) = r.

This “closes” the Solow growth model: the real wage will be determined residually: w = y-f’(k)k..

...which -with flexible wages- ensures growth at full employment at all times.

If a “well-defined marginal product of capital does not exist (as is the case per Robinson’s argument) the neoclassical theory of distribution falls apart...

The classical political economists had a different view: a) that wages would be determined institutionally, and b) that profits would be determined as a residual.

This can be expressed through the basic wage-profit relation r = R(1-ULC) where R is output/capital ratio & ULC...

...is the unit labor cost.

Despite some snarky dismissals of the problem by some “very serious people”, this is a very important difference.

For example, wages could be determined through #bargaining or #labor discipline. No guarantee of full employment.

What about #growth?

The simplest way would be to use the “Cambridge equation”: g = s r where s is the propensity to save (invest) out of profits. With a real wage determined as above, and a constant Y/K, you have “endogenous growth” out of capital accumulation alone, & no full employment.

This would be ~ 2 the Lewis model before the “turning point.”

There are many different ways to think about growth & distribution without an aggregate production function. This was just to showcase one possibility, clear things up a bit, & hopefully elicit some interest. /end