Neutrality of money

The “neutrality of money” refers to the notion that the effect of changes in an economy’s nominal supply of money will have no effects on the real variables like the real GDP, employment and consumption and only the nominal variables such as the prices, wages and the exchange rate are affected. It was the standard feature of the classical[1] macroeconomic model of unemployment and inflation that was based upon the assumption of quickly clearing perfectly competitive markets and the money market was governed by the quantity theory (Ackley, 1978).
This resulted in what was known as the “classical dichotomy” – the real and monetary sectors of the economy could be analysed separately as real variables like output, employment and real interest rates would not be affected by whatever was going on in the nominal segment of the economy and vice-versa. The objective of the present endeavour is to explore this concept of neutrality by delving into its theoretical motivations and basis and thereby introspecting upon the extent to which distinguishing between short run and long run neutrality are important before briefly exploring the possible methods of empirically investigating the notion and concluding.
In the standard classical macroeconomic model, which was the basis of answering all macroeconomic questions before Keynes’s General theory brought forth its capturing  assault onto it, the connection between the money supply and the price level was made through the quantity theory thus implying that the price level would vary to ensure the real aggregate demand, which was assumed to be a function of the real money supply, was in alignment with the available supply of output determined in the market for labour.

The quantity theory simply posits that real money balances are demanded in proportion to real income. This can be expressed as
MD/ P = (1/v).Y; where MD represents the nominal demand for money balances, P the price level, v the velocity of circulation of money and finally Y the real GDP.  Now by assumption, v is constant; MD equals the supply of money which is exogenous (MD, = MS = M) in equilibrium and Y is fixed at its equilibrium value (Y= Y*) determined in the labour market. As a result the quantity theory equation essentially becomes an equation that determines the price level for different levels of money. We have,    P = v.(M/Y*) .
Evidently, changes in the money supply now shall only influence the prices. This is the basis of the notion of neutrality of money which therefore is a direct derivative of the assumption of the quantity theory itself (Carlin and Soskice, 1990).
An increase in the supply of money initially leads to a rise in the aggregate demand above the real output (Y*, which is exogenous to the money market) due to increased availability of cash balances. Due to the excess demand situation the prices are pushed up until the demand for real output reduces to equal the supply of it. Note that in the classical system, the rate of interest plays the role of equating savings and investment at full employment and does not enter the money market.
However, in the 1930s the great depression which was essentially a situation of cascading mass unemployment had no convincing explanation in terms of the classical framework which proposed that an economy would always operate at full employment. This situation of mass unemployment and the lack of forthcoming explanations of the phenomenon in terms of the classical full-employment framework provided the context for the introduction of the Keynesian model of unemployment.
Although he upheld the assumption of perfectly competitive markets, he assumed prices to be fixed and money wages to be rigid and inflexible especially in the downward direction in the short run thereby implying the inability of the prices and wages to adjust to excess supply situations in the labour market; employment and output were determined by the effective aggregate demand in the product market. Consumption was assumed to be a function of real income implying savings, essentially the remainder of real income after consumption to be a function of real income as well rather than a function of real rate of interest as in the classical framework, and aggregate demand was made up of the planned expenses for consumption, investment and government expenses (for a closed economy).
Contrary to the classical model, in the Keynesian framework the rate of interest serves in equating real demand and supply of money rather than equating investment and full employment savings. This set up not only brings forth the possibility of equilibrium with unemployment prevalent in the labour market, it also dispels the concept of neutrality of money. An exogenous increase in the money supply through its effect on the real rate of interest affects the amount of investment and through that causes a change in the aggregate demand and thus in the real output and employment. So, this framework proves the non-neutrality of money the short run (Mankiw, 2000).
But in the long run, money can be deemed to have neutral effects through the following reasoning. An increase in the money supply will reduce the interest rates and increase investment. However, as the money supply rises, the real stock of money balances exceeds the desired level thus necessitating the expenditure on goods to be raised in order to re-establish the optimum and in that creating an excess demand in the goods market. In the long run prices and wages are perfectly flexible and in the presence of excess demand, there is a rise in the price level until the excess demand is satisfied, at the new equilibrium.
Again this rise in prices leads to an increase in the demand for money and thus leads to a restoration of the real interest rates and investments to their initial levels (Patinkin, 1987). Therefore, in the long run money supply increases have no effects on real interest rates, investment, or output in the long run. So, we find that although money is actually non-neutral in the long run due to the wage-price inflexibility in the short run, in the long run money has neutral effects. Infact, Patinkin (1956) notes that not only is money neutral in the short run but this short run neutrality is absolutely necessary for the quantity theory to hold. If this non-neutrality is denied and the classical dichotomy is accepted, then there is no theory of money, quantity theory or otherwise.
Testing the neutrality of money would require one to measure the effects of altered money supply has on real variables like the real GDP, employment and real interest rate. One approach possible would be to use a time series data set with values for these variables.
A regression would be run to ascertain the extent of effects if any, the changes in money supply over time has had on the real variables. In fact, Fisher and Seater (1993) have used time series data in this manner to test the neutrality of money. Their methodology however requires the usage of advanced econometric tools. Many consequent studies[2] have adopted this methodology to test time series data for different regions and check for neutrality of money.
Another option would be to use cross section data with different regions specified by different money supply values. By gauging the differences in the values of the real variables of these regions and relating these with the differences in the money supply values through regression analysis can be another way of testing for neutrality of money.
So, to sum up, we have seen that although short run neutrality of money is not a valid proposition, money does not have real effects in the long run. In the final section we have suggested two possible approaches to testing the neutrality of money.
References:
Ackley, G., (1978). Macroeconomics: Theory and Policy, New York: Macmillan
Boschen, J.F. & Otrok, C.M., (1994) Long run neutrality and superneutrality in an
ARIMA framework: comment, American Economic Review 84, 1470-1473.
Carlin, W., & Soskice, D., (1990) Macroeconomics and the Wage Bargain: A Modern Approach to Employment, Inflation, and the Exchange Rate, U.K.: Oxford University Press
Fisher, M.E. & Seater, J.J., (1993) Long run neutrality and superneutrality in an ARIMA framework, American Economic Review 83, 402-415.
Mankiw,  N.G., (2000) “macroeconomics” 4th ed, Worth publishers, New York
Patinkin., D. (1987) “Neutrality of money,” The New Palgrave: A Dictionary of Economics, v. 3, pp. 639-4
Patinkin, D., (1956) Money, interest and prices: An integration of monetary and value theory, New York: Row Peterson
[1] One should be beware of the misleading potential of the term classical and note its distinct presence in macroeconomics and its modern adoptions in the forms of new classical economics and thereby avoid confusing it with the school of economic thought associated with Marx, Smith and Ricardo.
[2]  e.g., Boschen and Otrok (1994) for the US