Conflicting Views of Inflation

What causes inflation?

One might think that this issue is fairly resolved in modern macroeconomics, but it is not. There are largely two competing visions of inflation circulating, the New Keynesian view and the Monetarist (Old and New) view.

The New Keynesian View

The basic New Keynesian model can be represented by the following three equations:

y_t = E_t y_{t+1} - a(R_t - E_t \pi_{t+1}) + e^{IS}_t

\pi_t = \beta E_t \pi_{t+1} + by_t + e^{PC}_t

R_t = R + \phi_{\pi} \pi_t + \phi_y y_t + e^{R}_t

where y is the output gap, R is the nominal interest rate, \pi is the inflation rate, E is the mathematical expectations operator, e^{IS}, e^{PC}, and e^{R} are stochastic shocks with mean zero and finite variance, and a, b, \phi_{\pi}, \phi_y are parameters. The first equation is the dynamic IS equation, the second is the New Keynesian Phillips curve, and the final equation is the Taylor Rule that governs monetary policy.

Within the New Keynesian framework, the central bank is an inflation fighter. What I mean by this is that the central bank adjusts the short term nominal interest rate, R, in response to realized inflation. The higher the value of \phi_{\pi}, the greater the responsiveness of the central bank. Inflation, within this framework, is pinned down by the Taylor principle (\phi_{\pi} > 1). To illustrate this suppose that there is some cost-push shock in the economy (e^{PI}_t is positive). The cost-push shock causes an increase in realized inflation. However, if the central bank adheres to the Taylor principle, the central bank will raise the short term interest rate by more than the increase in inflation. With sticky prices, this leads to an increase in the real rate of interest and therefore a reduction in “aggregate demand” and thereby the output gap. The reduction in the output gap reduces inflation (see the New Keynesian Phillips curve).

So what determines inflation in the New Keynesian model? In short, inflation is determined by the central bank target of inflation. Deviations from this target are caused by shocks to supply (e^{PC}) and demand (e^{IS}). The demand shock is (given the derivation of the model) a household preference shock (e.g. households wake up one morning and decide that they prefer current to future consumption). The supply shock is a shock to marginal costs, which causes firms to increase their prices. So long as the central bank adheres to the Taylor principle, the inflation rate can be uniquely determined.

A separate, but related question is what causes inflation in the New Keynesian model? Conceivably, it is the shocks to supply and demand. The central bank doesn’t create inflation, it prevents it through the Taylor principle.

The Monetarist View

The New Keynesian View is in stark contrast to the Monetarist View. In the Monetarist View, the central bank is not an inflation fighter, but rather an inflation creator. To understand why, let’s consider this from a New Monetarist perspective.

The New Monetarist framework is interested in examining money, monetary arrangements, intermediation, monetary policy, etc. from a perspective in which (*gasp*) money is actually important. You will note that in the New Keynesian framework above, money was never mentioned. This is because money is inconsequential for analysis within the NK model. The main reason why money is unimportant in NK models, however, is by assumption. Money, if it is introduced, is done through reduced form methods like putting money in the utility function or requiring that individuals must buy goods with money. These reduced form approaches, however, fail to capture the characteristics of money that make it important and therefore it is not surprising that they fail to identify that money isn’t important.

New Monetarist economics, in contrast, concerns itself with environments in which money is essential. It is therefore not surprising that one gets substantially different results. In NM models, the goods price of money is an endogenous determined. Put differently, the value of money is determined within the model rather assumed. If the goods price of money is zero, this means that money has no value and therefore a monetary equilibrium doesn’t exist (money is not simply assumed to be unimportant).

In equilibrium the goods price of money is determined as follows:

\phi_t = {{z(q_t)}\over{M}}

where \phi_t is the goods price of money, z(q) is the quantity of real money balances demanded given the quantity of consumption q, and M is the aggregate supply of money. If we think of the goods price of money as the inverse of the price level, this implies that the price level is determined by the following equilibrium condition:

P_t = {{M}\over{z(q_t)}}

In other words, the price level is pinned down by the ratio of the money supply and money demand. It is deviations between the money supply and money demand that cause fluctuations in the price level. It follows that the central bank in this type of framework is an inflation creator. Money growth causes inflation.

So why does this distinction matter?

It matters because it is an often overlooked aspect of inflation targeting. When forecasts are being conducted on inflation, care needs to be taken as to how inflation should be forecast. The NK approach would have inflation forecasts based on the NK Phillips curve. The NM approach would have inflation forecasts based on money growth.

Unfortunately, simply taking these models to the data doesn’t resolve this issue. For example, NK advocates would argue that money growth doesn’t do a good job predicting inflation and therefore the NK approach is better. However, the empirical evidence upon which these claims are made rely on simple sum measures of the money supply, which are severely flawed (more on this later). In addition, from the NM side, one could easily point to the failure of the NK Phillips curve in fitting the data. While New Keynesians like to point out that the fit improves by incorporating lagged terms of inflation, it also undercuts the notion of microfounded macroeconomic models and suggests a circularity between fitted the data to the model and the model to the data.

In short, the differences between the New Keynesian view on inflation and Monetarist view on inflation are stark. In the NK view, the central bank acts as an inflation fighter using the nominal interest rate to mitigate and control inflationary pressures. In the NM view, the central bank is an inflation creator. The price level is determined by the relationship between the money supply and money demand. This distinction is not minor.

(Note: The inflation fighter vs. inflation creator terminology comes from the work of Robert Hetzel.)

8 responses to “Conflicting Views of Inflation

  1. why not explain Keynes’s theory of inflation? As I point out in my 2009 book THE KEYNES SOLUTION:THE PATH TO GLOBAL ECONOMIC PROSPERITY — Keynes identified two types of Inflation (1) Commodity Inflation and (2) Incomes Inflation. The firstrequired a buffer stock policy amnd the second an incomes policy

  2. Interesting post.

    Is the “new monetarist” theory “wrong” if the demand for money changes in unpredictable ways while the natural interest rate is very stable?

    Is the “new Keynesian” theory “wrong” if the natural interest rate fluctuates alot and the demand to hold money doesn’t change much?

    It seems to me that the “empirical” approach assumes that we are looking to help the Fed direct the economy. The argument that a better measure of the money supply makes a difference sounds to me like it is pointing to using that measure of the money supply as a central bank target.

    • Bill,

      Let me answer your first two questions with more questions. 🙂

      What is the source of money demand instability? This seems to be an empirical conclusion, not a theoretical one. In addition, I don’t think the empirical evidence is supportive of the instability hypothesis on close examination.

      What is the source of instability in the natural rate of interest?

      Regarding the empirical approach, let’s assume that they target the forecast. I am arguing that how they make that forecast matters. I am not arguing that they should partake in fine-tuning nor that they target the money supply.

      Ultimately, I think that the big divergence is whether one thinks of the central bank as an inflation fighter or an inflation creator.

  3. As a matter of political economy, I think inflation fighter vs. inflation creator explains why the new Keynesian perspective is dominant. Somehow, those monetary economists who are out to help the Fed seem to do better than those who are critics of the Fed.

    My “priors” is that there is no reason for either the demand for money or the natural interest rate to be expecially stable.

    I don’t mean to be critical of the Divisia approach. I have always rejected M2 for “theoretical reasons,” M1 is worthless because of sweep accounts, and Divisia is looking better than MZM.

    The alernative is really the monetary base. Of course, that makes no attempt to capture all assets that provide monetary services.

    My view is that Ms-Md is correct (the new monetarist approach.) If it turned out that the natural interest rate never changed and monetary disequilibirum always caused market interest rates to change, so (I think) a monetary authority could always avoid monetary disequilibrium with an interest rate target–Ms and Md is still right.

    Similarly, if there is no way to adjust Ms to Md because of, say, changing preferences, that doesn’t make it wrong.

    • Fair enough. I think that we agree.

      Regarding aggregates, I have been thinking a lot about which monetary aggregate is preferable and how we choose the composition of the money aggregate. The more I think about it, the more I think that the broadest money aggregates are preferable — mainly for reasons pertaining to excludability.

  4. Pingback: Varieties of Phillips Curves | The Everyday Economist

  5. The Keynesian and Monetarism perspectives have become insufficient for the determination of exo-multi economic variables (micro, macro and meso). The greatest fallacies of both those schools of economic management is the amount of uncertainty that is left to the operation and exception of the very assumptions of the schools for which realistically can be catastrophic (e.g. Zimbabwe, etc) and even advantageous with disastrous consequences (America as the fundamental for global recession). Concerning what causes inflation in both models can best be addressed by what is inflation and causes by virtue are endogenous to the model of inflation. Inflation is an economic measure of imbalances between aggregate demand and aggregate supply factors as predetermined by government regulation and the effects of business firms and households economic decisions as they respond to anticipated and unanticipated government actions and as influenced by global or meso economic factors that are complex and intertwined due to the myriad of economic structures of the world. Both schools have become complacent in an ever growing world. A concentric model of inflation, growth, stability and prosperity requires formulation and implementation. The new Keynesian and Monetarist approaches have too many patches and loop holes for any serious economic management of twenty first century economies.

  6. nelson kamugisha

    thanks for the good work

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s