Monetary trends in the UK and the USA from 1874 to 2020: a nonlinear approach to money demand

Since the influential works of Friedman and Schwartz (1963, 1982) on the monetary history of the United States of America and the United Kingdom from 1876 to 1975, there has been a great concern in the literature about the instability of money demand functions when monetary trends are explored historically. Several authors, at the end of the 1980s and during the 1990s, reconsidered their statistical approach based on the distinction between money demand in the long-run (cointegration) and money demand in the short-run (error-correction models). Recently, using M1 as the measure of money, Benati, Lucas, Nicolini and Weber (2021) have shown, for a shorter and more recent period of time, that there are stable long-run money demands for a long list of countries. However, to date there are no studies on whether a stable long-run and short-run money demand equations exists for the entire monetary history of the US and the UK. By means of a nonlinear cointegration and a nonlinear error-correction approach, this paper goes beyond the work of its predecessors and presents evidence of a well-specified, long-run and short-run money demands of real broad monetary balances in these two countries from 1874 to 2020. The estimated cubic-polynomial cointegrations (US and UK) for the long-run money demand specification, and the linear (US) and cubic equilibrium correction (UK) for the short-run money demands, are shown to be stable based on an appropriate measure of the opportunity cost of holding money.