RDP 1978-01: Two Essays in Monetary Economics 5. “Money” or “Assets”?
September 1978
Firstly, money may be a proxy for financial assets in general for a wide range of problems involving the effects of inflation on asset values and the subsequent goods market effects. The detailed portfolio models favoured by Tobin and his followers are designed for tracing the chain of substitution effects following say, an open market operation, but have been singularly unproductive in throwing light on the dynamics of inflation and output adjustment. Even if one is committed to a multi-asset view of the world, monetary disequilibrium is a good proxy for asset disequilibrium in general if the composition of portfolios adjusts relatively rapidly. While this is a relatively unexamined subject, rapid adjustment of the composition of portfolios but not of the levels of asset holdings seems to be the case in the U.K. in the period ot 1939, although the composition takes time to adjust in the post-war period when direct controls of various sorts, especially on the capital account of the balance of payments, have be endemic.[15]
This subtle and probably somewhat controversial issue aside, the strongest argument for the importance of money comes from insights about its role as a buffer stock and signalling device. If, in an uncertain world, people use money as a buffer stock and make their decisions about consumption, the labour they will supply or demand, purchases of durable assets, holdings of bonds and other decisions, and, at the end of the day, look to their money balances to signal the aggregate impact of their economic decisions, money plays a crucial role.
This argument suggests that it is disequilibrium in the money market rather than in a whole set of associated markets which is the relevant variable for inclusion in the appropriate adjustment functions. It does not answer the question about which markets or equations monetary disequilibrium will influence. One possible answer is “all of them”, but quite a bit of work with post-war Australian data decisively refutes this hypothesis in suggesting that monetary disequilibrium effects mainly the non-bank groups take-up of government bonds, bank lending, consumption and pricing decisions. To the extent that monetary disequilibrium influences consumption and, therefore output, and prices, it will of course have powerful induced effects on other variables.
A problem which arises for some with the above analysis is that the money as buffer stock approach is in apparent contradiction to the simple open economy monetarist model in which money balances always equal the demand for money. This superficial contradiction arises from failure to consider the distinction between the short run and the long run. Even in models in which money is determined residually as a buffer stock in the short run, appropriate incorporation of monetary disequilibrium in consumption and price adjustment functions ensures that money balances equal money demand in the longer run. Recognition of this point resolves the (usually unstated) conflict between economists who estimate the demand for money function[16] and those who emphasise the money supply process.[17] The arguments presented above suggest that the quantity of money will be determined as the outcome of all other decisions in the short run, both because governments tend to use the printing press for residual financing and because the private sector allows its money balances to take the burden of adjustment in the short run. The arguments also suggest that the standard practice of estimating the demand function for money by single equation techniques with observed money balances as the dependent variable may be quite perilous and may suggest that money demand is unstable when it is not. In this connection it is interesting that demand for money functions estimated in the normal way have tended to break down when the control regime is changed from one in which interest rates and exchange rates are pegged (when normal estimation techniques may be more or less appropriate) or when considerable monetary instability is introduced.[18] The alternative approach may involve identifying the parameters of the desired or long run money demand function indirectly by way of the disequilibrium real balance effect in price and quantity equations, whatever the control regime adopted by the authorities. This has the advantage both of being theoretically more general and, as suggested by results of the empirical work noted above, empirically more robust.