The Non-Existent Hand...

My letter on the Stiglitz review of Robert Skidelsky's "Keynes: The Return of The Master" is printed in the May 27, 2010 issue of the London Review of Books. Paul Davidson

It reads as follows:(Sent to me by Paul Davidson__The economist we need to solve the present world crisis...)


The Non-Existent Hand...

Joseph Stiglitz criticises Robert Skidelsky, Keynes’s biographer, for not understanding Keynes’s theory, but in doing so reveals his own imperfect understanding (LRB, 22 April). The basis of his complaint is Skidelsky’s distinction between risk and uncertainty. Risk, Skidelsky explains, exists when the future can be predicted on the basis of currently existing information (e.g. probability distributions calculated from existing market data); uncertainty exists when no reliable information exists today about the future outcomes of current decisions, because the economic future can be created by decisions taken today. According to Stiglitz, this is a distinction without a difference, and ‘little insight’ into the causes of the Great Recession is gained from Skidelsky’s emphasis on uncertainty as opposed to risk.

But this is not what Keynes believed. The classical economics of Keynes’s time presumed that today’s economic decision-makers have reliable information regarding all future outcomes. I have labelled this the ‘ergodic axiom’. By contrast, Keynes argued that ‘unfortunate collisions’ occurred because the economic future was very uncertain. ‘By very uncertain,’ he wrote, ‘I do not mean the same thing as “very improbable”.’ No reliable information existed today for providing a reliable forecast of future outcomes.

This is the very proposition that Stiglitz denies. All that is needed to provide better insight into the workings of the market, he thinks, is ‘small and obviously reasonable change in assumptions’; for example, that reliable information about the future does exist but that different individuals have access to different information. The only necessary policy is ‘transparency’: to make complete information about the future available to all. The classical ergodic axiom is correct, provided one accepts that not everyone has access to all the information that exists.

For Keynes the inability of firms and households to ‘know’ the economic future is essential to understanding why financial crashes occur in an economy that uses money and money contracts to organise transactions. Firms and households use money contracts to gain some control over their cash inflows and outflows as they venture into the uncertain future. Liquidity in such an economy implies the ability to meet all money contractual obligations when they fall due. The role of financial markets is to assure holders of financial assets that are traded on orderly markets that they can readily convert these liquid assets into cash whenever additional funds are needed to meet a contractual cash outflow commitment. In Keynes’s analysis, the sudden drying up of liquidity in financial markets, occasioned by sudden drops of confidence, explains why ‘unfortunate collisions’ occur – and have occurred more than a hundred times in the last 30 years, according to Stiglitz.

By contrast, Stiglitz implicitly accepts the orthodox view that all contracts are made in real terms, as if the economy were a barter economy. Consequently people’s need for liquidity is irrelevant. Stiglitz indicates that he and Bruce Greenwald have explained that financial markets fail ‘because contracts are not appropriately indexed’, i.e., contracts in our economy are denominated in money terms rather than ‘real’ terms. He suggests that if only such contracts were made in real, rather than monetary, terms we would not suffer the ‘unfortunate collisions’ of economic crisis. If only we lived in a classical world, where contracts would be denominated in real terms! But in a money-using economy, this is impossible.

Paul Davidson
Journal of Post Keynesian Economics, New York