Monetarism and the Quantity-Theory Tradition
From Hume to Keynes: “internal stability” as a goal of monetary policy
Most of the major propositions of monetarism follow from Friedman’s restatement of the quantity theory of money (Friedman 1956).
The quantity theory, which in one form or another dominated monetary orthodoxy until the rise of Keynesian economics, has been subject to many different interpretations throughout the history of economic thought (see Laidler 1991). Common to all variants of the quantity theory is the view that there exists a tight relationship between the stock of nominal money balances, the flow of nominal income and, in the long run, the purchasing power of money. It was further argued that the causal chain runs from stocks to flows and prices, and not otherwise. To appreciate Friedman’s variation on the theme, it is important to go beyond the similarities, and to highlight some differences between the classical, neoclassical, and monetarist tradition (see Patinkin 1969, 1974). All versions of the quantity theory state a “quantity equation”.A basic variant is given by:
Mv = PY, (1)
where M denotes the nominal supply of money (usually, the monetary base), Y the level of aggregate real income, and P the corresponding price level. The “velocity of money”, denoted by v, measures money’s average rate of turnover (the number of transactions enabled by a unit of money per period), reflecting the fact that the quantity equation relates the stock of money to the flow of nominal income (PY). The Cambridgereciprocal, k ? 1/v, measures the average period over which money balances are held (as a fraction of the elementary period). If k is determined by economic behaviour, it is a measure of the real transaction demand for money.
The use of the equality sign is pure convention. In fact, the “quantity equation” is a flow identity and, as such, tautological.
Its left-hand side simply specifies nominal spending as “money in circulation”. It states the trivial fact that nominal spending in the aggregate is always and by necessity equal to nominal income; that what is spent (decumulation of cash balances) is by definition received (accumulation of cash balances). So far, the analysis is empirically empty and, in some form or another, common to all variants of the quantity theory of money. To arrive at predictions, assumptions were made, and it is in the details of these assumptions that the variations differ:(A1) The “relevant” definition of nominal money supply is exogenous: M = M.
(A2) There exists no systematic relationship between the nominal stock of money and its velocity (or the real demand for money). Equivalently, v (or k) is independent of M. In studying the impact of money supply on income and prices, it is therefore feasible to fix the velocity of money: v = V.
(A3) Classical dichotomy: monetary factors exert no persistent influence on the level of real output. The output level in stationary equilibrium (the classical centre of gravitation) is fully determined by “real” factors. Thus, in the long-run: Y = Y.
Given the quantity equation, (A1)-(A2) already yield a simple theory of nominal income. Changes in the stock of money supply suggest equi-proportional changes in nominal spending and, thus, income. The theory of nominal income suggests no particular equilibrium specification. It may hold in static (neoclassical) and stationary (classical and neoclassical) equilibria; it also holds in disequilibrium. All assumptions together then yield the prediction of “quantity of money theory of the price level” (Tobin 1974: 86):
(Quantity Theorem) Given equation (1), if (A1)-(A3) hold for whatever reason, then the following statements are true and equivalent in a closed economy in long-run (steady-state) equilibrium: the purchasing power of the aggregate money supply (M/P) is independent of the level of nominal balances; money-supply variations are reflected in equi-proportional changes in the price level; monetary policy fully controls nominal variables; monetary policy has zero grip on quantities (real variables); money is neutral.
Or, in the words of Hume, whom Friedman (2008) celebrated as the best early advocate of the quantity theory: “If we consider any one kingdom by itself, it is evident, that the greater or less plenty of money is of no consequence; since the prices of commodities are always proportioned to the plenty of money” (Hume 1752 [1955]: 33; emphasis added).
Hume’s quantity theory must be seen as a vital contribution to a general attack on mer- cantilistic principles. It ridicules the goal of increasing a nation’s “riches” by maximizing the domestic stock of gold. Mercantilism is irrational, so the message, in the sense that it is founded on a money illusion, the confusion of nominal and real economic magnitudes. In the aggregate, it is impossible to accumulate wealth by hoarding gold, since money’s purchasing power adjusts until any nominal supply purchases the equilibrium level of output. Friedman, at least, always expressed greatest sympathy for this interpretation of Hume’s quantity theory (Friedman 1956: 10-11, 1966 [1969]: 145, 1974: 1-3, 2008).
In the quotation above, Hume explicitly refers to a closed economy (see italics). In general, however, he was concerned with open economies, considering international trade between pre-industrial commercial clusters that share gold and silver as common media of exchange. His quantity theorem is a statement of pure logic, directed against the profound irrationality of mercantilist thought and policy. According to his famous price-specie flow mechanism, money supply is endogenous (A1 does not apply): any national increase in the nominal money supply would disturb the international balance- of-payment equilibrium. According to the quantity theory, such an expansionary impulse leads to an increase in the domestic price level and, thus, deteriorates the terms of trade. In response, the trade balance turns into deficit, financed by an outflow of money (or monies). Falling prices, as again predicted by the quantity theory, then restore the initial terms of trade and rebalance the current account.
In new and old equilibrium, the global stock of gold and silver is distributed across nations in proportion to their real transaction demand for money (Dimand 2013: 293-7). Given free financial markets, a monetary authority committed to maintaining the convertibility of national currency into gold and silver cannot persistently influence nominal variables.By contrast, the monetarist orthodoxy champions flexible exchange rates (Friedman 1953a). It argues that nominal exchange rate adjustments serve as a buffer against external shocks. Most importantly, flexible exchange rates restore monetary sovereignty and, thus, the exogeneity of money. Even though the monetarists celebrated Hume and the classical quantity-theory tradition, they advocated a completely different monetary regime and, therewith, a different interpretation of the quantity theory. In case of inconvertible currency and flexible exchange rates, the quantity theory suggests the monetary authority’s long-run responsibility for price stability (and the responsibility for stable money income in the short run). In fact, monetarists are Keynesian in that they are willing to sacrifice “external stability” for the sake of such “internal stability
Thus, it comes as no surprise that Friedman admired Keynes’s early work as a Marshallian quantity theorist, especially his advocacy of monetary sovereignty and price stability as the primary long-run goal of monetary policy (see Keynes 1923 [1971]; Friedman 1974: 171, 1997: 2). Irrespective of vital differences (discussed below), Friedman’s restatement of the quantity theory is best understood as an attempt to improve upon Keynes’s early monetary analysis. This view is not in conflict with Friedman’s firm belief that his restatement was consistent with the General Theory (1936 [1973]), for he also believed that Keynes’s theory of output in a monetary economy was meant to be a refinement of his earlier work in the Cambridge tradition:
Keynes was a quantity theorist long before he was a Keynesian, and he continued to be one after he became a Keynesian.... I maintain that Keynes’s discussion of the demand curve for money in the General Theory is for the most part a continuation of earlier quantity theory approaches, improved and refined but not basically modified. As evidence, I shall cite Keynes’s own writings in the Tract on Monetary Reform. (Friedman 1974: 159, 168)