Interest Rates and Inflation

Questions & answers

Q1. What tension between “consensus” monetary policy views and quantity-theoretic evidence does the paper open with?

The modern policy consensus holds “that the instrument of monetary policy ought to be the short term interest rate, that policy should be focused on the control of inflation, and that inflation can be reduced by increasing short term interest rates” – yet “the systematic evidence… consists almost entirely of evidence that increases in average rates of money growth are associated with equal increases in average inflation rates and in interest rates” (Section 1, p. 1). The paper notes that “the U.S. inflation of the 1970s and 80s can be fully accounted for by the corresponding increase in M2 (or M1) growth rates, and the return to relatively low inflation rates in the 1990s can be explained by the correspondingly low average rate of money supply growth in that decade” (p. 1-2), making an outright rejection of the quantity theory “a difficult step to take.”

Q2. Why do the authors reject standard new-Keynesian IS/Phillips-curve models as the right framework for this question?

Because those models “contribute nothing to our understanding of why the 60s and the 90s were low inflation decades, relative to 70s and the 80s, or why Germany has been a low inflation country, relative to Mexico” – the low-frequency, welfare-relevant questions that “one needs to rely on some explicit version of the quantity theory of money” to answer (Section 1, p. 3). They also note that most coherent monetary theories (Sidrauski, Brock, Lucas-Stokey) have no genuine liquidity effect at all – “with a complete set of financial markets, it is just not true that when the government buys bonds, the price of bonds increases” – so a different modeling ingredient is needed just to get interest rates to respond to open-market operations in the conventional direction (Section 1, p. 3-4).

Q3. How does market segmentation generate a liquidity effect in this model?

Only a fraction λ of agents (“traders”) attend both the goods market and the bond market where open-market operations take place; the rest (“non-traders”) never trade bonds, so an increase in the money supply is absorbed only by traders, temporarily depressing the return they require on bonds (Section 2, p. 4-5). Formally, from the log-linearized Euler equation, r_t = ρ̂ + φ(E_t[μ_{t+1}] − μ_t) + E_t[μ_{t+1}] + E_t[v_{t+1}] − v_t (eq. 6), “the immediate effect of an open market operation bond purchase, μ_t > 0, is to reduce interest rates by φμ_t. This is the liquidity effect that the segmented market models are designed to capture” (Section 2, p. 8). If markets are unsegmented (λ = 1), φ = 0 and “the liquidity effect vanishes” – interest rates can then only move via expected-inflation (“information”) effects (Section 2, p. 8).

Q4. What is the role of velocity shocks in the model, and how is inflation determined?

Velocity shocks (v_t) capture short-run instability in the money-price relationship; differencing the equation of exchange gives inflation as the sum of money growth and the change in velocity, π_t = μ_t + v_t − v_{t-1} (eq. 7), which anchors every subsequent example (Section 2-3, p. 9). Absent segmentation, prices would be “entirely determined” by the equation of exchange given the money supply path – “the quantity theory of money in its very simplest form” – with velocity shocks the only source of short-run money-price instability (Section 2, p. 7).

Q5. What does the “exact inflation targeting” example (Example 4) show, and why is it the paper’s benchmark?

That a money-growth rule conditioned on the contemporaneous velocity shock, μ_t = π̄ − v_t + v_{t-1}, can hit an announced inflation target exactly, for whatever the shock process is: “in our context, inflation targeting cannot be done any better than this” (Section 3, p. 10). This example becomes the benchmark against which the paper’s several Taylor-rule examples (5 through 8) are judged, since a Taylor rule uses the same current information but, by construction, ties the interest rate response to the deviation of inflation from target rather than directly targeting inflation via money growth.

Q6. What is the paper’s central comparison between money-growth rules and Taylor rules?

Because Taylor rules “use the same information as the rule in Example 4 that attains the inflation target perfectly,” but constrain the interest rate to move only in proportion to the inflation deviation from target (r_t = ρ + π̄ + θ(π_t − π̄), eq. 8), “committing to a Taylor rule amounts to tying the hands of the monetary authority in a way that can only limit its effectiveness” at hitting the inflation target, “as our examples illustrate” across both iid (Example 5) and serially correlated (Example 6) velocity-shock processes (Section 4, p. 15-16). A permanent change in the inflation target (Example 7) passes through one-for-one to money growth and interest rates regardless of the liquidity effect φ or the Taylor coefficient θ, showing those parameters affect only the transitional dynamics, not the long-run Fisherian link (Section 3, p. 13-14).

Q7. Does a policy of pegging the interest rate ever outperform a policy of fixing money growth?

Yes, conditionally: comparing an interest-rate peg (Example 4-type policy) to constant money growth with iid velocity shocks (Example 2), the paper finds “pegging the interest rate is inflation-stabilizing, relative to constant money growth, if and only if φ > 1/2” – that is, only when the liquidity effect is large enough (Section 3, p. 10-11). This result depends entirely on the size of the segmentation-generated liquidity effect, underscoring that the case for using interest rates as an instrument is a quantitative, model-dependent one rather than a general theoretical necessity.

Q8. If Taylor rules cannot beat a money-growth rule at inflation control, what does the paper suggest could rationalize using them?

Some additional objective beyond inflation-target attainment, most plausibly interest-rate smoothing: the authors note they “have in fact considered variations on the model presented here in which relative endowments of agents fluctuate, giving rise to gains from pooling endowment risk,” and that “in a model with segmented markets where such pooling cannot take place, there can be real gains from policies that smooth real interest rates” – what the founders of the Federal Reserve called an “elastic currency” – though the authors explicitly defer full analysis of this extension “to another paper” (Section 4, Conclusions, p. 16).