Kaldor and the relationship between ‘normal backwardation’ and the theory of storage

Explanations of the relationship between commodity prices at different delivery dates are based on the theory of storage. Holding stocks of commodities entails financial and storage costs. According to the theory of storage, these costs net of the stocks productive value (convenience yield) determine the difference between spot and futures prices through arbitrage operations. The theory of storage was developed, between the 1940s and the 1960s mainly by the US economist Holbrook Working, in alternative to the Keynes-Hicks theory of normal backwardation, which predates it by one decade. Having introduced the notion of convenience yield, Nicholas Kaldor is generally considered the initiator of the theory of storage and as such he is separated from Keynes and assimilated to Working. This separation is puzzling as Kaldor meant his theory to be a generalisation of Keynes’ own. The paper reconstructs how this separation occurred. As part of this reconstruction, we find that already in the 1940s at least two competing views on the formation of commodity prices existed, one giving prominence to fundamentals and homogeneous expectations, the other to speculation and heterogeneous expectations. This finding is interesting in view of its connections with the debate on the determinants of commodity prices today. Interestingly, we find that that Kaldor’s own theory was capable of encompassing both views on the formation of commodity prices as special cases and that Kaldor’s separation from Keynes on the theory of forward markets occurred when Working grafted the notion of convenience yield onto a non-Keynesian theoretical framework, anticipating the efficient market hypothesis