Endogenous and Exogenous Money

Abstract

The issue of endogeneity or exogeneity of money is one that runs through the history of monetary theory, with prominent authors appearing to hold views on either side. Narrowly put, those who plug for the exogeneity view take one or all among the cluster of variables – price level, interest rate or real output – as being determined by movements in the stock of money. Those who hold the endogeneity view consider that the stock of money in circulation is determined by one or all of the variables mentioned above. This narrow definition begs several questions. The variables price level (P), interest rate (R), real output (Y) and money stock (M) are all at the macroeconomic level, i.e. in the context of a one-good economy. Some part of the continuing debate can be traced to the view held by various participants in the controversy about whether such a high level of aggregation is appropriate, e.g. is there a rate of interest? Another part of the debate refers to the choice of money stock variable. Is it commodity money (gold), fiat (paper) money, bank deposits or a larger measure of liquidity that is to stand for the money stock? The problem can be dealt with even at a one-good level either in the context of a closed economy or an open economy and either in an equilibrium or a disequilibrium context, static or dynamic, short run or long run. The basic issue is about the direction of causality-money to other variables or other variables to money. But as our understanding of the underlying statistical theory concerning causality and exogeneity has advanced in recent years, it must also be added that participants in the controversy conflate the exogeneity of a variable (especially of money) with its controllability by policy. Strictly speaking one can have exogeneity without any presumption that the variable can be manipulated by policy, for example rainfall. Also once posed in a dynamic context, we should distinguish between weak exogeneity, which allows for feedback from the endogenous to the exogenous variables over time, and strong exogeneity, which does not allow such a feedback (Hendry et al. 1983). Endogeneity or exogeneity are notions that only make sense in the context of a model. Frequently in the past, there has been a failure to specify such a model, which has then allowed the controversy to continue.