By the end of 1927, Sraffa, not yet in his thirties, showed Keynes a system of three linear and homogeneous equations to determine prices in an economy without surplus. We now know that these “first equations” were to be the starting point for his future (1960) book. Considered as such, this elementary system can hardly be considered
as particularly impressive. Indeed, it could be viewed as a rather trivial set of accounting identities. Keynes’s reaction is therefore surprising: “It is very interesting and original—he wrote to his wife Lydia—but I wonder if his class will understand it when he lectures”. In the present paper, it is suggested that Keynes’s enthusiastic approval of the equations relied on their being the outcome of a totally new theoretical approach to the problem of value. With the aid of the Sraffa Papers, preserved in the Wren Library, this approach is shown to be a most unusual one, so unusual as to be hardly intelligible and acceptable even for the “advanced” students for which
Sraffa’s lectures were intended. Moreover, if we adopt this new approach, a remarkable consequence becomes evident. This is that the vexed question of variable or constant returns becomes irrelevant—as Sraffa always maintained.
