Symmetric vs Asymmetric Inflation Targets

A symmetric inflation target treats overshoots and undershoots as equally undesirable—if the target is 2 %, both 3 % and 1 % are equally bad. An asymmetric target tolerates one direction more than the other, usually accepting higher inflation to avoid the costs of unemployment. This design choice has large real consequences for how aggressively central banks tighten, how much slack they allow in labor markets, and the distribution of gains and losses between savers and workers.

The Design Question

When a central bank publishes an inflation target, it must decide: how much is the cost of being a little above target versus a little below? If the target is 2 %, is inflation of 2.5 % as bad, worse, or better than 1.5 %?

Under a symmetric target, they are equally bad. The central bank should raise rates as fast to bring inflation down from 2.5 % to 2 % as it would lower rates to bring inflation up from 1.5 % to 2 %. There is no favorite direction.

Under an asymmetric target, the central bank explicitly says: “We tolerate overshoots more than undershoots” (or vice versa). This means the central bank will move more aggressively to prevent inflation from falling below target than to prevent it from rising above. As inflation rises toward the ceiling, the central bank hesitates; as it falls below the floor, the central bank cuts hard.

This is not a minor technical detail. It shapes the entire path of interest rate policy and, through that, employment and financial stability.

Symmetric Targets: Equal Treatment

A purely symmetric target is conceptually clean. The central bank publishes: “Our inflation target is 2 % annually,” implying that 1 % is as costly as 3 %.

The economic rationale is simple. Low inflation (1 % or below) imposes costs on the economy:

• Workers’ real wages grow unpredictably, discouraging long-term contracts.
• Debtors gain less from inflation eroding their real debt burden.
• A fall in inflation can signal weak demand and rising unemployment.
• Firms are reluctant to cut prices (menu costs, labor morale), so nominal wage stagnation becomes real wage cuts.

High inflation (3 % and above) imposes different costs:

• Savers lose purchasing power; interest on savings is eroded.
• Workers must constantly renegotiate wages.
• Long-term investment becomes harder to plan (inflation surprises).
• Firms face uncertainty in pricing.

If you believe these costs are roughly equal at ± 1 % from target, a symmetric target makes sense. You accept one unit of risk of undershoot and one unit of risk of overshoot.

In practice, most central banks now publish symmetric targets. The Federal Reserve says it targets 2 % inflation; the European Central Bank says 2 % as well. The Bank of England has 2 % CPI inflation as a point target. These look symmetric on paper.

Asymmetric Targets: Built-In Bias

An asymmetric target explicitly weights one direction as worse than the other. The most common form is higher tolerance for inflation than deflation. A central bank might say: “We try to keep inflation near 2 %, but we will accept inflation of 3.5 % to avoid deflation or very high unemployment.”

The rationale reflects a historical lesson from the Great Depression and Japan’s “Lost Decade.” Deflation—falling prices—is particularly costly because:

• Nominally, debts do not fall; so deflation raises the real burden of debt, spurring defaults.
• Real wages rise even if nominal wages fall or stagnate, pressuring employers to cut jobs.
• Consumers and firms postpone purchases, expecting lower prices tomorrow; demand collapses.
• Once deflation expectations take hold, raising inflation expectations again is very costly (requires years of loose policy).

Unemployment is also costly:

• Lost income, reduced future earning potential, psychological damage, and increased mortality and illness.
• A year of unemployment is worse than a year of 0.5 % higher inflation for most workers.

Given these asymmetric costs, a central bank might target a 2 % midpoint but make clear that it will tighten much more aggressively if inflation threatens to fall below 1 % than if it rises above 3 %. This is an asymmetric target.

Japan’s Implicit Asymmetric Approach

Japan’s central bank did not formally adopt an asymmetric target until recently, but its behavior was asymmetric for decades. The Bank of Japan raised rates very reluctantly in response to inflation, but cut aggressively whenever growth slowed. This bias—tolerate inflation, fight deflation—was eventually written in. In 2013, the Bank of Japan adopted an explicit 2 % inflation target but added language suggesting it would be more tolerant of overshoots than undershoots in the transition period.

This asymmetry reflected a political reality: the Japanese public and government were terrified of deflation (having lived through 20+ years of it) and preferred even moderate inflation to the risk of deflation. An asymmetric target formalized this preference.

Worked Example: Rate Paths

Suppose the target is 2 % and current inflation is running at 2.5 %. Under a symmetric target, the central bank should raise rates by some amount today; under an asymmetric target (higher tolerance for inflation), it might hold or raise less, waiting to see whether inflation recedes on its own.

Now suppose inflation falls to 1.5 %. Under a symmetric target, the central bank should lower rates to bring inflation back up to 2 %, because 1.5 % is as bad as 2.5 %. Under an asymmetric target with higher inflation tolerance, the central bank might cut much faster, trying to get back to 2 % (or even toward 2.5 %) before the risk of deflation sets in.

Over a full cycle, the asymmetric central bank keeps interest rates lower on average, inflation slightly higher on average, and unemployment slightly lower. The symmetric central bank is more balanced but leaves more unemployment risk.

The Phillips Curve and Policy Trade-offs

This choice connects to the Phillips Curve: the empirical relationship between unemployment and inflation. A central bank that is asymmetric (tolerates inflation, fears deflation) will run the economy hotter, keeping unemployment lower, in exchange for a bit more inflation. A symmetric central bank, indifferent between inflation and low inflation, will tolerate higher unemployment to keep inflation down.

Which is better? Economists disagree.

The symmetric argument: Inflation, once elevated, is hard to bring down without a recession (Volcker shock, 1980s). It is better to head off inflation early, keep expectations anchored to 2 %, and avoid the costly disinflation later. A few percentage points of unemployment now beats a deep recession in 10 years.

The asymmetric argument: Unemployment is painful immediately and irreversibly damages workers’ careers. High inflation is disruptive but reversible (a recession and rate hikes can bring it down). Moreover, monetary policy cannot permanently lower unemployment below the “natural rate”—it can only trade off temporary gains for future pain. So it is better to err on the side of keeping labor markets tight and inflation moderate, rather than leaving unemployment high to provide a buffer against inflation risks.

The 2020s Debate

The COVID-19 pandemic and the 2021–2023 inflation surge reopened this debate sharply. The Federal Reserve had allowed inflation to rise well above 2 %, arguing (initially) that the overshoot was temporary. Critics said the Fed was behaving asymmetrically—tolerating inflation to protect jobs—even though its published target was symmetric.

When the Fed finally raised rates sharply in 2022, unemployment remained low for many quarters, validating some of the asymmetric intuition (you can tolerate high inflation without destroying the labor market). But the lag in inflation’s response suggested that if the Fed had tightened sooner, it could have kept inflation much closer to 2 % without the same long disinflation effort.

This experience suggests that even if a central bank publishes a symmetric target, its actual tolerance depends on the state of the economy, the Fed’s credibility, and policymakers’ beliefs about the costs of unemployment versus inflation. Formal target design is important but not determinative of behavior.

Financial Stability Dimension

A third consideration is financial stability. A central bank running an asymmetric, higher-inflation regime keeps real interest rates lower for longer. This can feed asset bubbles—rising stock and real estate valuations that eventually collapse. A symmetric central bank, willing to tolerate low inflation and raise rates when needed, may prevent bubbles at the cost of lower average growth.

This has led some economists to propose flexible asymmetric targets or side constraints. For example: “Our inflation target is symmetric at 2 %, but we will not allow credit growth to exceed 10 % annually, even if inflation is low.” This tries to balance the employment benefit of asymmetry with the financial stability cost.

See also