From Monetarism to Inflation Targeting: What Actually Stuck?

Two Sentences That Changed Central Banking

In 1968, Milton Friedman told the American Economic Association that there was no long-run tradeoff between inflation and unemployment, that the Phillips curve was vertical at the natural rate, and that the central bank’s real job was to control inflation by controlling money. In 1990, New Zealand’s parliament passed the Reserve Bank Act, requiring the central bank to target a specific inflation rate, publicly announced. Between those two moments lies one of the most consequential intellectual migrations in modern economics: the journey from monetarism as a research program to inflation targeting as an institutional practice.

The story is not a straight line. It involves a dramatic experiment that partly failed, a quiet revolution in how economists thought about expectations, and a pragmatic compromise in which central banks took the spirit of monetarism while discarding its letter. Understanding what stuck and what did not is essential for anyone who wants to make sense of how monetary policy actually works today.

What Monetarism Meant in Practice

Monetarism, as a policy doctrine rather than a body of academic theory, boiled down to a handful of propositions. First, inflation is always and everywhere a monetary phenomenon, in the sense that sustained inflation requires sustained money growth beyond what is needed to accommodate real economic expansion. Second, the central bank can and should control the money supply, or at least the monetary base, and through it the broader aggregates. Third, discretionary fine-tuning of the economy through activist monetary policy is likely to do more harm than good, because the lags between policy action and economic effect are long and variable. Fourth, the best policy is a rule: Friedman’s preferred version was a constant growth rate for the money supply, the famous k-percent rule.

These propositions were not merely academic. By the late 1970s, inflation in the United States had reached double digits, and the Keynesian framework that had guided policy for a generation seemed unable to explain or fix it. The Phillips curve, which had promised a stable menu of inflation-unemployment tradeoffs, had broken down. Friedman and Edmund Phelps had predicted exactly this breakdown a decade earlier, and their prediction came true with devastating clarity. The intellectual credibility of monetarism was at its peak.

The Volcker Experiment

When Paul Volcker became chairman of the Federal Reserve in August 1979, inflation was running above 11 percent and accelerating. Volcker announced in October 1979 that the Fed would shift its operating procedure from targeting the federal funds rate to targeting nonborrowed reserves, a change designed to let interest rates move as needed to control monetary growth. The language was deliberately monetarist. The message to markets was that the Fed was serious about wringing out inflation, even if it meant sharply higher interest rates and economic pain.

What followed was dramatic. The federal funds rate spiked above 20 percent. The economy entered a severe recession in 1981-82, with unemployment reaching nearly 11 percent. But inflation came down, from over 13 percent in 1980 to under 4 percent by 1983. It was the most successful disinflation in recent American history, and it reshaped expectations about the Fed’s willingness to fight inflation.

But here is the part that complicates the simple monetarist narrative: the reserves-targeting procedure did not actually work as a monetary control mechanism. The relationship between nonborrowed reserves and broader monetary aggregates was erratic. The Fed struggled to hit its money supply targets, and when it did hit them, the relationship between money growth and nominal spending was unstable. Financial innovation was reshaping the banking system in real time. Money market mutual funds, NOW accounts, and sweep programs were blurring the boundaries between money and near-money, making the measured aggregates unreliable guides to the underlying monetary conditions that the Fed was trying to control.

By late 1982, the Fed had quietly abandoned the strict reserves-targeting procedure and returned to something much closer to interest rate management, though the rhetoric about monetary aggregates lingered for several more years. Volcker himself later acknowledged that the reserves-targeting framework was partly a political device: it allowed the Fed to let interest rates rise to whatever level was necessary to break inflation while deflecting blame from the FOMC onto the “impersonal” mechanics of reserve control. The substance was a commitment to reduce inflation regardless of the short-run cost. The monetarist operating procedure was the vehicle, but the vehicle turned out to be expendable once the destination was reached.

Why Velocity Broke

The failure of monetary aggregate targeting was not just a technical inconvenience. It struck at the empirical foundation of monetarism as a policy guide. Friedman’s framework depended on a stable money demand function, which implied that velocity would be predictable enough to make money supply growth a reliable indicator of future nominal spending and inflation. Through the 1970s, this assumption held up reasonably well for M1 and M2 in the United States. In the 1980s, it fell apart.

The proximate cause was financial deregulation and innovation. The Depository Institutions Deregulation and Monetary Control Act of 1980 and the Garn-St Germain Act of 1982 removed interest rate ceilings on deposits and expanded the powers of thrift institutions. The result was a proliferation of new financial instruments that were functionally similar to money but classified differently in the monetary statistics. When interest rate ceilings were binding, the boundary between money (which paid zero or regulated interest) and bonds (which paid market interest) was clear. When ceilings were removed, money became interest-sensitive, and the demand for measured monetary aggregates became unstable.

This was not a surprise to the most thoughtful monetarists. Friedman himself had always acknowledged that the appropriate monetary aggregate might change over time and that financial innovation could disrupt established relationships. But acknowledging the possibility in theory and dealing with it in real-time policy are very different things. The instability of velocity in the 1980s forced central banks to look for a different anchor.

The Pivot to Interest Rate Targeting

The shift from monetary aggregate targeting to interest rate targeting happened gradually and somewhat reluctantly. By the mid-1980s, most major central banks were de facto targeting short-term interest rates, even if they maintained a rhetorical commitment to monetary aggregates. The Bundesbank, often cited as the most monetarist of central banks, continued to announce monetary targets through the 1990s, but Bundesbank insiders acknowledged that the targets were subordinate to the interest rate decisions that actually steered the economy.

The theoretical justification for interest rate targeting came from several directions. One was simply the failure of monetary targeting: if you cannot reliably control the aggregates, you need a different instrument. Another was the development of New Keynesian models in which the interest rate was the natural policy instrument because it directly affected intertemporal consumption and investment decisions, while the money supply was a residual determined by money demand at the prevailing interest rate. In these models, specifying a monetary aggregate was unnecessary and potentially misleading.

A third justification came from the growing emphasis on expectations. If inflation expectations are the key variable that the central bank needs to anchor, then the question is not “what growth rate of M2 produces 2 percent inflation?” but “what instrument setting, combined with what communication strategy, makes people believe that inflation will be 2 percent?” The interest rate, which the central bank can control precisely on a daily basis, turned out to be much better suited to this task than monetary aggregates, which the central bank controls only indirectly and with considerable noise.

New Zealand’s Innovation

The idea of an explicit, public, numerical target for inflation crystallized in New Zealand in 1989-1990. The Reserve Bank of New Zealand Act of 1989, which took effect in early 1990, required the central bank governor to negotiate a Policy Targets Agreement with the Minister of Finance, specifying a target range for inflation. The original target was 0 to 2 percent. The governor’s job was explicitly on the line: failure to meet the target was grounds for dismissal.

This was a radical departure from previous practice. Most central banks had vague mandates: price stability, full employment, financial stability, some combination. The New Zealand model made the target specific, public, and enforceable. It was, in a sense, the implementation of the monetarist preference for rules over discretion, but with a different target variable. Instead of a rule for money growth, it was a rule for inflation outcomes.

The New Zealand experiment attracted enormous attention. Canada adopted inflation targeting in 1991. The United Kingdom followed in 1992, after its exit from the European Exchange Rate Mechanism. Sweden, Australia, and numerous other countries adopted some form of inflation targeting through the 1990s and 2000s. The European Central Bank, established in 1998, chose a definition of price stability as inflation “below but close to 2 percent,” which was functionally an inflation target even if the ECB resisted the label. The Federal Reserve was a latecomer, formally adopting a 2 percent inflation target only in 2012, though its behavior had been consistent with such a target for years before.

The Taylor Rule

The intellectual bridge between monetarist rule-following and inflation-targeting practice was built in part by John Taylor’s 1993 paper proposing a simple formula for the federal funds rate. The Taylor Rule says that the central bank should set the nominal interest rate as a function of the current inflation rate relative to target and the current output gap. When inflation is above target, raise rates. When output is below potential, lower rates. The weights on each term, and the assumed equilibrium real interest rate, are parameters that can be debated, but the structure is straightforward.

The Taylor Rule was not derived from first principles; it was a description of what the Fed appeared to be doing during the successful Greenspan era. But it quickly became prescriptive as well as descriptive. It offered a benchmark against which actual policy could be judged. When the Fed deviated from the Taylor Rule, economists asked why, and that question imposed a kind of accountability that was very much in the monetarist spirit.

The Taylor Rule also had an important intellectual property: it was a rule for the interest rate, not for the money supply, but it embodied the same commitment to systematic, predictable behavior that Friedman had advocated for monetary growth. The instrument changed; the philosophy of rule-following persisted.

The New Consensus in Monetary Policy

By the early 2000s, a “new consensus” had emerged in monetary economics, at least among mainstream practitioners. Its key elements were: (1) the central bank should have a clear mandate focused on price stability; (2) the primary instrument is the short-term interest rate; (3) policy should be systematic and transparent, guided by something like an inflation target and something like a Taylor Rule; (4) central bank independence from short-run political pressure is essential for credibility; (5) expectations management, through forward guidance and clear communication, is as important as the interest rate setting itself.

This consensus drew from multiple intellectual traditions, but the monetarist contribution was central. The idea that inflation is fundamentally a monetary phenomenon, meaning it is ultimately the central bank’s responsibility, came directly from Friedman. The preference for rules over discretion, for transparency over opacity, for systematic behavior over ad hoc intervention, was a monetarist theme from the 1960s onward. The emphasis on central bank independence, while not uniquely monetarist, was powerfully reinforced by the monetarist critique of politicized monetary policy in the 1960s and 1970s.

What the new consensus did not take from monetarism was the specific instrument. Money supply targets were gone. The k-percent rule was gone. The detailed modeling of money demand functions, which had been the bread and butter of monetarist empirical work, was marginalized. In the canonical New Keynesian models that underpinned the new consensus, money did not even appear as a variable. The interest rate was the instrument, the inflation forecast was the intermediate target, and the money supply was whatever it needed to be.

What Central Banks Actually Took from Chicago

The ledger, then, is mixed. Central banks took from the Chicago tradition: the conviction that inflation is the central bank’s primary responsibility; the preference for rules-based or at least systematic policy; the emphasis on credibility, expectations, and the long-run neutrality of money; the insistence on central bank independence as an institutional safeguard against inflationary bias; and the deep skepticism of fine-tuning and activist stabilization policy.

They did not take: money supply targeting as an operating procedure; the k-percent rule; the specific claim that controlling monetary aggregates is the right way to control inflation; or the broader monetarist skepticism of Keynesian demand management, which was repackaged in New Keynesian form and incorporated into the inflation-targeting framework.

The result is a monetary policy regime that Friedman would have found partially satisfying and partially frustrating. He would have approved of the inflation focus, the rules-based philosophy, and the institutional independence. He would have been skeptical of the interest rate instrument, the reliance on models that ignore money, and the increasing resort to unconventional policies like quantitative easing that expand the central bank’s balance sheet without clear rules for how and when to reverse course.

Forward Guidance and the 2 Percent Target

Two features of modern monetary policy deserve special attention in the context of the monetarist legacy. The first is forward guidance: the practice of central banks communicating their likely future policy path to influence expectations and long-term interest rates today. Forward guidance is, in one sense, the ultimate expression of the expectations-focused approach that Friedman and Lucas pioneered. If expectations matter, then managing expectations is policy. The Fed’s “dot plot,” the ECB’s “forward guidance language,” and the Bank of Japan’s “yield curve control” are all ways of trying to shape expectations about the future path of interest rates.

But forward guidance also creates tensions that Friedman might have predicted. If the central bank commits to a future policy path, it constrains its future discretion. If economic conditions change, the central bank must choose between honoring its guidance (and implementing the wrong policy) and abandoning its guidance (and losing credibility). This is a version of the rules-versus-discretion dilemma that Friedman spent decades analyzing, applied to a new instrument.

The second feature is the 2 percent inflation target itself. Where did 2 percent come from? The honest answer is that it was somewhat arbitrary. New Zealand’s original target range was 0 to 2 percent, later widened. Other countries converged on 2 percent through a combination of imitation, convenience, and the argument that a small positive target provided a buffer against deflation and the zero lower bound on nominal interest rates. There is no deep theoretical reason why 2 percent is optimal rather than 1 percent or 3 percent.

Friedman, characteristically, would have preferred zero inflation, on the grounds that any positive inflation rate imposes unnecessary costs on money holders and distorts relative prices. The 2 percent target reflects a pragmatic compromise between the monetarist preference for price stability and the Keynesian concern about the risks of too-low nominal interest rates in a world where the real equilibrium rate might be low.

What the Story Tells Us

The journey from monetarism to inflation targeting is not a story of one school defeating another. It is a story of practical adaptation, in which an intellectual revolution changed the questions that central banks asked, even as the specific answers evolved beyond what the revolutionaries had envisioned. Friedman changed the framework: he made inflation the central bank’s problem, he made expectations the key mechanism, and he made rules the benchmark against which discretion was judged. The fact that the rules ended up being about interest rates rather than money supply, and that the models ended up being New Keynesian rather than monetarist, does not erase the intellectual debt.

If anything, the COVID-era inflation episode reinforced one of Friedman’s deepest convictions: that monetary and fiscal authorities cannot sustainably create purchasing power out of thin air without eventually confronting inflationary consequences. The mechanisms were different from what a strict monetarist model would have predicted. The money supply numbers were messier, velocity was less predictable, and the supply-side disruptions added genuine non-monetary factors. But the basic monetarist intuition that you cannot flood the economy with money and expect prices to remain unchanged proved remarkably durable.

The question for the next generation of central bankers is whether the inflation-targeting consensus, built on monetarist foundations but no longer recognizably monetarist in its methods, is adequate for a world of climate shocks, geopolitical fragmentation, digital currencies, and fiscal dominance. Friedman, who delighted in being proven right about the big picture and wrong about the details, would probably have enjoyed the argument.