On the St. Louis Equation and an Alternative Definition of the Money Supply

Abstract

IN THEIR WELL KNOWN and controversial paper, Andersen and Jordan [2] presented evidence that, when responses of GNP to monetary and fiscal actions are analyzed, only money matters. The equation from which these results were derived has since been subjected to much criticism. This criticism has centered mostly on the statistical specification of the St. Louis equation (e.g., [14]) and on the "reversecausation" argument (e.g., [12]). In perhaps its most popularized form, the St. Louis equation measures monetary actions by employing changes in the narrowly defined money stock (?\M1). Fiscal actions are measured by changes in high employment federal expenditures (AE). In addition, statistical constraints are imposed upon the equation for estimation purposes. These constraints are: 1) each independent variable has the same lag length, 2) the distributed lag weights are estimated by a fourth degree polynomial by the Almon technique with both endpoints constrained to zero. Schmidt and Waud [14] show that if the statistical constraints are removed, estimation of the St. Louis equation indicates that money matters, but so do fiscal actions. Lombra and Torto [12] demonstrate that, if the variable employed to measure monetary actions is adjusted to account for "reverse-causation," the conclusion derived from the St. Louis equation is that money still matters, but again, so do fiscal actions. In spite of these criticisms, the St. Louis equation continues to be used in its original form (e.g., [4]). The purpose of this paper is to show that, even if 1) the Andersen-Jordan constraints are imposed upon the equation, and 2) the monetary variable is not adjusted for reverse-causation, estimation of the St. Louis equation with a weighted aggregate money stock variable leads to conclusions different from those discovered by Andersen and Jordan. The money stock variable considered in this paper, arbitrarily called Mw, is derived and more fully explained in a recent paper by the author [9]. MW is a weighted average of six financial assets and is an attempt to take into account Gurley's suggestion that assets be weighted according to their degree of moneyness in determining a money stock.' Moreover, in comparison with both the weighted average case and the case of combinations of unweighted sums of the component assets, it was found that MW satisfies the Friedman-Meiselman dual criteria for * Professor of Management Science, The Pennsylvania State University. 1. Other studies have also attempted to implement Gurley's suggestion. Among these are the studies by Chetty [5], Feige [6], Kane [8], Laumas [10], Lee [11], and Timberlake and Fortson [15]. For the calculation of the weighted monetary aggregate employed in this paper, the weights are determined by factor analyzing the assets into two factors defined as money and near-money [9, p. 1084]. The assets and their corresponding weights for calculating Mw are: currency (0.45733), demand deposits (0.53713), time deposits (0.09883), mutual savings bank shares (-0.16489), savings and loan ...

Money and Household Consumption Expenditure in Developing Asia: Some Panel ARDL Evidence

“…When we compare these results with broad money (M3) as a proxy for money supply, the results (Table 5) continue to hold true and steady, indicating the superior performance of monetary aggregates over its fiscal counterpart, government expenditure. The results obtained are in line with Modigliani and Ando (1976), Koot (1977), Barth and Bennett (1974), Hafer (1982) and Jordan (1986).…”

“…A review of studies supporting or opposing the F-M and A-J models and results are found in Batten and Thornton (1983), De Leeuw and Kalchbrenner (1969), Davis (1969), Corrigan (1970), Goldfeld and Blinder (1972), Blinder and Solow (1974), Modigliani and Ando (1976), Koot (1977), Barth and Bennett (1974), Hafer (1982) and Jordan (1986). Bias’s (2014) comprehensive survey of all the Friedman–Meiselman/Andersen–Jordan single equation debate succinctly presents the entire body of work.…”