Academic journal article Journal of Money, Credit & Banking

The Behavior of Money Velocity in High and Low Inflation Countries

Academic journal article Journal of Money, Credit & Banking

The Behavior of Money Velocity in High and Low Inflation Countries

Article excerpt

THIS PAPER ANALYZES the long run behavior of money velocity and its relation with the inflation rate. On theoretical grounds, the relation between these two variables is governed by the growth rate of money. At the steady state, countries with higher growth rates of money should present higher levels of inflation and larger nominal interest rates. As the opportunity cost of holding money increases, real money demand is reduced, and velocity rises. Thus, there should be a strong positive relation between average money velocity and both average inflation and the average rate of money creation. The same reasoning should apply to the volatility of these variables. Countries with more volatile money growth rates should have more volatile inflations and velocities. However, the correlation between the average growth rate of money and the average velocity between the years 1960 and 2000 for a cross-section of 79 countries is just 0.22 while the correlation between the standard deviation of money growth and the standard deviation of velocity over the same sample is 0.40 for those same countries.

I argue that such small correlations arise in the data because the sample pools together countries with dissimilar transaction technologies for which the relation between velocity and inflation (or the growth rate of money) is different. Once countries are sorted by an indicator of transaction costs, the correlation between these variables rises dramatically. To understand why that should be the case and to compute a measure of costs in the financial sector, I develop a general equilibrium version of the models in Baumol (1952) or Tobin (1956) where the velocity of money is determined endogenously and changes in response to fluctuations in the interest rate.

General equilibrium models of the transactions demand for money have been generated in different ways in the literature. Jovanovic (1982) presents a deterministic economy where agents have access to a productive storage technology for capital. Agents want to consume continuously but capital can only be consumed if transferred to the market at a fixed cost. Since capital perishes once removed from storage, money is used to finance consumption between the dates when capital is liquidated. Another route has been taken by the so-called "shopping-time" models of money demand, such as the ones in Den Haan (1990) or Guidotti (1989). In these models, agents value consumption of goods as well as leisure, but transacting goods is a time-consuming activity. Money is demanded because real balances reduce the time needed for transactions. Feenstra (1986) and Marshall (1992) use the same idea but the cost involves goods instead of time. Finally, in liquidity models like Rotemberg (1984), Grossman and Weiss (1983), Alvarez and Atkenson (1997), or Alvarez, Atkenson, and Edmond (2002) agents decide on the composition of their portfolios between cash, needed for transactions, and bonds, bearing an interest rate. However, the number of trips to the bank per period is fixed exogenously in these models.

Romer (1986) presents a general version of the Baumol--Tobin model of money demand where the frequency of financial transactions is determined endogenously. He develops a non-stochastic, continuous time overlapping generations (OLG) economy where agents choose the pattern of money holdings throughout their lives. The model is then used to analyze the Mundell--Tobin effect, the optimum quantity of money, the effect of inflation on consumption, and the welfare costs of inflationary finance. However, because of the agent's heterogeneity and the non-recursivity imposed by the OLG structure he is only able to solve the model for the specific case of logarithmic instantaneous utility and no time discount. In fact, as he recognizes, with a different utility function, "... the equations describing equilibrium would therefore be extremely complicated, and the analysis of those equations extremely impossible. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.