Soft Landing

A soft landing is a disinflation that arrives without a recession: the central bank raises interest rates enough to slow the rise in prices, and the economy slows without contracting. The word "landing" carries the analogy, an aircraft descending under control rather than dropping, and the alternative it is set against is a hard landing, in which the same tightening tips the economy into a downturn.

The phrase is often treated as commentators' shorthand, and it is not. It appears in the Federal Reserve's official record as a description of what policy was trying to achieve. Pressed at the September 2023 press conference, the Federal Reserve's then-Chair, Jerome Powell, said: "A soft landing is a primary objective, and I did not say otherwise. I mean, that's, that's what we've been trying to achieve for all this time." Earlier in the same conference he described it as an outcome rather than a plan, saying he had "always thought that the soft landing was, was a plausible outcome—that there was a path, really, to, to a soft landing."

The defining feature of this term is an asymmetry, and it is the reason the page exists. A Federal Reserve chair has named a soft landing as a primary objective, and the formal documents that would carry criteria for it do not mention it at all. Contrast that with a recession, which is also not defined in statute but does have a body that assigns it: the National Bureau of Economic Research reviews indicators, sets dates for peaks and troughs, and announces them, so a recession is a claim that can eventually be checked. Nothing comparable exists for a landing. The word does not appear in the Committee's Statement on Longer-Run Goals and Monetary Policy Strategy, which is where the Federal Reserve sets out its objectives and is the document that specifies the 2 percent inflation goal numerically. It does not appear anywhere in the Federal Reserve's July 2026 Monetary Policy Report to Congress, all 77 pages of it. It does not appear in NBER's business-cycle dating material. It is a word policymakers use in conversation about an objective their formal documents describe in other terms.

The price side and the joint outcome are different claims, and conflating them is the common error. Bringing an inflation rate down is disinflation, and that has its own entry; it can be observed directly in published data, and it can happen alongside a recession, before one, or without one. A soft landing is the conjunction: the disinflation plus the absence of a contraction. So "inflation is falling" is evidence about one half of the claim and says nothing about the other, which is why a commentary that reports falling inflation and concludes a soft landing is under way has skipped a step.

A soft landing is not a costless one, and the same account says so. In July 2023 Powell noted that the disinflation to that point had come "without any real costs in the labor market," and then qualified it in the same answer: "the historical record suggests that there's very likely to be some softening in labor market conditions," and, on the shape of the good outcome, "consistent with having a soft landing, you would—you would have some softening in labor market conditions, and that's still likely as we—as we go forward with this process." So on that account the expected version of a successful landing includes a weaker labor market than the one it started from. A definition of soft landing that requires no rise in unemployment is stricter than the one those remarks describe, and it is one of several a reader may encounter.

The consequence for a household plan is that this is not a positionable view. A forecast is only useful if it can be resolved, and this one cannot be, in either direction: there is no announcement to wait for and no criterion to apply afterward. A household that shifted its allocation in anticipation of a hard landing would have no event to tell it when the anticipated thing had or had not happened, and the entry on panic selling covers what acting on that kind of view tends to cost. The information a plan can actually use from this territory is narrower and more useful: whether an emergency fund covers a job loss, whether spending is committed or discretionary, and whether the portfolio's bond allocation matches the years the money is needed.

The mechanism a soft landing describes is the ordinary transmission of monetary policy, run at just the right pressure. The Committee raises its target range; borrowing gets dearer; households and businesses defer purchases and projects; demand cools relative to what the economy can supply; and the rise in prices slows. A soft landing is that sequence stopping before the cooling turns into a contraction. The entry on monetary policy covers the transmission, including the long and variable lags that make the calibration hard.

A hypothetical episode, which is not a description of any actual one, showing why the term cannot be scored. Over two years, suppose the following happens.

• Inflation falls from 8 percent to 3 percent.
• The unemployment rate rises from 4 percent to 5 percent.
• No recession is dated for the period.

Was that a soft landing? Three defensible criteria give three answers. If the criterion is "no recession," yes, clearly. If it is "inflation back at the 2 percent goal," no, because 3 is not 2 and the Committee's commitment is to the goal rather than to the direction. If it is "no rise in unemployment," no, and by a full percentage point. Nothing published chooses among the three. There is no body to appeal to, no release that settles it, and no date on which the answer arrives.

That is the practical content of the term, and it is a negative worth knowing. A soft landing is a coherent thing to aim at, which is why a Federal Reserve chair described it as a primary objective. It is not a thing that gets declared, and any source that declares one is applying a criterion it chose itself.

What the term is good for

• It names a real and coherent policy objective in one phrase, and a Federal Reserve chair used it that way in the official record.
• It keeps two things joined that are easy to separate. Inflation coming down is only half of the outcome anyone actually wants.
• It makes the alternative explicit. A hard landing is the same policy path with a contraction at the end of it, which is a genuine risk rather than a rhetorical one.

Where it misleads

• The documents that would carry criteria for it do not mention it, so two commentators can reach opposite verdicts on the same period and neither can be shown wrong.
• It is often used as though it meant no economic pain at all, when the Federal Reserve's own framing of it expected some softening in labor market conditions.
• It is applied in the present tense, unlike a recession, which is dated only in hindsight, so it invites confident statements about a period that is still running.
• It cannot be resolved, which means it cannot support a portfolio decision. There is no event that pays off a correct call on it.