Soft landing vs hard landing: the historical record

Why this comparison matters

When a central bank raises rates to fight inflation, two outcomes are possible: growth slows just enough to bring prices down (soft landing), or it slows so much that output and employment contract (hard landing). The two terms describe the same policy through opposite endings. The confusion lies in treating a hard landing as the default and a soft landing as a fluke – the historical record is more balanced than that framing suggests. A companion piece: the Eco3min mapping of asset-class correlations across regimes.

What a soft landing is

A soft landing is a slowdown in growth that stops short of recession, where inflation returns toward target while unemployment stays broadly contained. The term gained currency under Alan Greenspan, credited with engineering one in 1994-95 when the Fed roughly doubled the funds rate from about 3% to 6%, then cut three times in 1995 without a downturn. Powell has cited 1965, 1984 and 1994 as the canonical examples. It is rare but not unique.

→ The complete explanation: What is a soft landing and has it ever been achieved?

What a hard landing is

A hard landing is when tightening, or the conditions around it, ends in an NBER-dated recession: falling output, rising unemployment, contracting credit. Blinder identifies three post-1965 recessions – 1973-75, 1980 and 1981-82 – that were direct results of Fed tightening; the 2008 and 2020 downturns had non-monetary triggers. The distinction matters because not every recession that follows a hike cycle was caused by it.

→ Related reading: How accurate are recession indicators?

The key differences

Trigger. A soft landing requires demand to cool gradually; a hard landing involves a sharper contraction, often amplified by credit stress or an external shock layered on top of high rates.

Starting conditions. This is where the real divide sits. The three classic soft landings began with inflation already low or below 5%, per the Congressional Research Service (2023). The three monetary hard landings followed the most aggressive cycles, when inflation was already entrenched – the higher the plane is flying when descent begins, in Blinder’s phrasing, the harder the landing tends to be.

Frequency. Soft landings are not the exception they are often assumed to be. Blinder counts five of eleven Fed tightening cycles since 1965 as soft or soft-ish, where GDP fell less than 1% or no NBER recession followed for at least a year.

How they play out across regimes

The landing depends less on the size of the rate move than on the inflation regime it confronts. When tightening begins from moderate inflation and no supply or geopolitical shock intervenes – 1965, 1984, 1994-95 – growth has tended to decelerate without breaking, and soft landings followed. When the starting point is an entrenched inflationary regime requiring extreme tightening, as in 1973-75 and 1981-82, the descent has historically overshot into recession. The 2022-24 cycle is the live test: the Fed lifted rates from near zero to 5.25-5.5% against inflation that peaked at 9.1% in June 2022 (BLS), yet US real GDP grew 2.9% in 2023 (BEA) with no recession – a sharp cycle that, so far, resembles a soft-ish landing more than the 1980s playbook the starting inflation might have implied.

The landing is set on takeoff: not by how hard the brakes are pressed, but by how high inflation was flying when the descent began.

→ Framework: What is the Sahm rule and how reliable is it?

The common confusion

The frequent error is to treat recession-warning indicators as verdicts rather than probabilities. The 2-year/10-year Treasury curve inverted in 2022 and stayed inverted for roughly 500 days – the longest in the modern era – and the 3-month/10-year spread remained inverted until December 2024, yet no recession arrived. The Sahm rule, triggered in July 2024 when the unemployment three-month average rose 0.53 points above its low (CRS, 2024), also flashed without a downturn. Both have strong records, but a strong record is not the same as an inevitability, and the same indicators that called past hard landings missed this cycle.

→ More on the false signal: Why does the yield curve invert before recessions?

Practical observation

Go deeper

Related guides

• Policy rates vs long-term rates: why they move differently
• 2-year vs 10-year yield: reading the curve

Frequently asked questions

How often has the Fed achieved a soft landing?

Counting depends on definition. Using a strict bar – no recession at all – soft landings are rare, with 1965, 1984 and 1994-95 the cleanest cases. Using Alan Blinder’s broader test, where GDP falls less than 1% or no NBER recession follows for at least a year, five of the eleven Fed tightening cycles since 1965 qualify as soft or soft-ish. The takeaway is that the outcome is more frequent than the popular framing of a near-impossible feat implies, even if a fully clean landing remains the exception.

Why did the 2022-24 cycle not end in a hard landing despite an inverted curve?

The yield curve inverted in 2022 and stayed inverted into 2024, the longest such episode in the modern era, and the Sahm rule triggered in 2024 – both classic hard-landing signals. Yet US real GDP grew 2.9% in 2023 (BEA) and employment held. Several factors are cited: households and firms had locked in low borrowing costs before 2022, the labor market entered the cycle exceptionally tight, and disinflation came partly from supply normalization rather than demand destruction. The episode is a reminder that indicators carry probabilities, not certainties.

What separates a soft landing from a hard one?

The decisive variables, on the historical record, are the inflation level when tightening begins and whether an external shock intervenes. Cycles that started from moderate inflation without an oil, war or financial shock – 1965, 1984, 1994-95 – tended to land softly. Cycles that confronted entrenched inflation requiring extreme tightening, or that collided with a supply shock, tended to overshoot into recession. The size of the rate increase matters less than the altitude, in Blinder’s analogy, from which the descent starts. This page belongs to a larger run, collected in where the other comparisons sit.

Last updated — 12 July 2026