Econometric Analysis of Money Supply, Inflation, and Unemployment

Introduction and Research Justification

The title of the study proposed herein is: "An Econometric Analysis of the Relationship Between Money Supply, Inflation, and Unemployment."

The ongoing global economic downturn makes an understanding of the relationship between the supply of money and unemployment rates a timely and valuable enterprise. Despite a growing body of econometric research over the past several decades, much remains unknown or unclear. As Keuzenkamp emphasizes, "Basic economic phenomena, such as the consumption and saving patterns of agents, remain enigmatic. After many years of econometric investigation, there is no agreement on whether money causes output or not" (2000, p. 1).

The unemployment level in the United States — one of the major economic drivers of the global economy — stood at over 9 percent at the time of this writing, with indications that global economic stagnation would keep that rate elevated for the foreseeable future (Employment situation summary, 2010). According to Ghysels, Swanson, Watson, and Granger, "The model of the unemployment rate consists of two equations: the money growth equation and the unemployment rate equation" (p. 323).

An early study by Chow and Megdal (1978) notes that "numerous studies have appeared to refine, respecify, and estimate structural equations explaining the rates of change in wage rates, the price level, unemployment and related variables [and] the relationship between unemployment and inflation is implicit in the econometric model" (p. 446). As Klein (1991) points out, however, "such a relationship must be conditional on a number of factors, including past patterns of inflation and unemployment, as well as the supply conditions of some raw materials whose prices are determined in worldwide commodities markets" (p. 202).

Preliminary Literature Review

These observations clearly indicate that economic forecasts used to guide budgetary decisions must account for a wide range of dynamic variables, making modeling all the more challenging. Hicks (2009) emphasizes that "forecast accuracy is obviously important, but it is widely recognized among economists that no forecast is truly right — though some might be helpful in informing the decisions for which they are formulated" (p. 1).

According to Duo (1997), "Econometrics is a frontier discipline in the introduction of scientific means and methods into economics. It was born out of the desire to bridge the gap between economic theory and economic data" (p. 1). The "gap" to which Duo refers is the need to develop robust analytical tools capable of measuring the effects of a wide range of variables on the economy. Modeling these variables can take a number of different approaches. For example, Srivastava and Rao (1990) note that "variables that can be chosen instead of the unemployment rate or in addition to the unemployment rate are the vacancy rate, the layoff rate, the quit rate, real or nominal wage inflation rates, etc." (p. 131).

A comparison of available econometric models used for macroeconomic analyses, provided by Klein (1991), identifies one approach — the Indiana University model of the U.S. economy — as particularly well suited for this type of analysis. The Indiana University econometric model is comprised of 100 equations, 151 identities, and 56 exogenous variables (Klein, 1991). The model regards output as being influenced by overall demand, with prices treated as a variable markup over the costs of labor per unit (Klein, 1991). The determination of wage rates uses an inflation-supplemented Phillips curve, and corresponding expectations are adaptive (Klein, 1991).

The Indiana University model also incorporates 20 endogenous components of real GNP, two inventory components, 11 consumption expenditure components, three fixed nonresidential investment components, residential investments, imports, exports, and purchases made by state and local governmental entities (Klein, 1991). According to Klein, "Federal government defense and nondefense purchases are exogenous. Government and rest-of-world output are modeled separately and subtracted from real GNP to yield real private domestic output" (p. 63).

Real private domestic employment and man-hours worked are factor demand equations that rely on output as well as capital stock levels determined using a production function (Klein, 1991). The Indiana University model also derives the wage bill from the wage rate and hours equations, along with other separately modeled components of income (Klein, 1991). As Klein states, "The [Indiana] model maintains consistency among wage rates, hours, employment, productivity, unit labor costs, the wage bill, and prices for the private domestic economy" (p. 63).

Nonborrowed reserves M1 and M2 are endogenously identified in the Indiana University model and serve as the basic monetary policy aggregate. An inverted M2 money demand equation is derived from the Treasury bill rate, and the relationship between the bond rate and the Treasury bill rate is developed using a term structure equation (Klein, 1991). The results of the Indiana econometric model can be used to provide periodic forecasts and to evaluate empirical macroeconomic hypotheses (Klein, 1991). Taken together, the Indiana University econometric model provides a comprehensive analytical approach to understanding the relationship between money supply, inflation, and unemployment levels.

Methodology: The Indiana University Econometric Model

1. To apply the Indiana University econometric model to historic economic data in the United States in order to develop informed views concerning the relationship, to the extent that it exists, between unemployment, money supply, and inflation; and,

2. To provide a series of recommendations for economic policymakers concerning this relationship and what steps can be taken to increase employment while restraining inflation levels.

The data sources for the proposed study will include peer-reviewed and scholarly literature from university and public libraries, as well as reliable online resources such as EBSCO and Questia. The statistical data required will be obtained from the U.S. Bureau of Labor Statistics and the U.S. Bureau of Economic Analysis, as well as other federal government resources that maintain archival economic data as required.

Based on its perceived advantages, the proposed study will employ the Indiana University econometric model to determine the relationship, to the extent that it exists, between historic unemployment levels in the United States and corresponding levels of inflation and money supply. According to Klein (1991), "In the Indiana model, prices are determined as a variable markup over unit labor costs. Wages are determined by the degree of slack in labor markets and by inflationary expectations" (p. 105).

The Indiana University econometric model assumes adaptive expectations; therefore, the level of change in prevailing wages relies upon the lagged values of the inflation rate at the time (Klein, 1991). In the model, the extent to which output increases is also the extent to which employment increases, thereby driving the unemployment rate down — which in turn places upward pressure on wages (Klein, 1991). The model also accounts for the upward pressure on wages placing upward pressure on prices, again with a lag. Accordingly, any increases in output tend to result in increases in prices (Klein, 1991).

Klein also notes that "in the Indiana model, employment and hours adjust with a lag to changes in output and thus produce a procyclical movement in productivity" (p. 105). As a result, increases in output produce short-term increases in productivity that cause a temporary decline in the costs of labor per unit, placing downward pressure on prices — an outcome that dissipates as employment levels adjust to the new situation (Klein, 1991). Recent applications of the Indiana University economic model have produced reliable forecasts of unemployment levels (Hicks, 2009).