--- title: "Market Correction vs Bear Market: What Separates the Two Labels" description: "A correction and a bear market are separated by a threshold the press agreed on. Here is what each label measures, and what neither of them can tell you." author: "David Cohen" category: "Headlines Decoded" date: 2026-10-18T09:20:01.510Z canonical: "https://investormuse.com/blog/market-correction-vs-bear-market-difference" --- # Market Correction vs Bear Market: What Separates the Two Labels ![correction tape, phrases, office](https://cdn.pixabay.com/photo/2016/01/02/08/01/correction-tape-1117994_1280.jpg?w=1200&q=75) > A correction and a bear market are separated by a threshold the press agreed on. Here is what each label measures, and what neither of them can tell you. Two headlines, two moods. "Stocks slip into correction territory" sounds like weather. "Wall Street enters a bear market" sounds like a diagnosis. The gap between them is a number somebody picked, and knowing where that number came from changes how much attention the headline deserves. Both labels describe the same thing: how far a broad index has fallen from its most recent high. Neither describes why it fell, how long it will last, or what the companies inside the index are doing. ## One Official Threshold, One Newsroom Convention Only one of the two has an official definition in the United States. The SEC's investor glossary states that [a bear market generally occurs when a broad market index falls by 20 per cent or more over at least a two-month period](https://www.investor.gov/introduction-investing/investing-basics/glossary/bear-market), with pessimistic sentiment alongside it. The matching entry for the other direction says [a bull market generally occurs on a rise of 20 per cent or more in a broad market index over at least two months](https://www.investor.gov/introduction-investing/investing-basics/glossary/bull-market). Notice what the regulator's definition contains that the headline version never does. There is a time condition attached. A fall of a fifth over eight weeks or longer is a bear market by that wording. A fall of a fifth across three sessions and a recovery in the fourth week is something else, and the glossary gives it no name. The ten per cent figure attached to the word correction has no equivalent official status. The SEC glossary carries entries for both bull and bear markets and no corresponding entry for a correction at all. The ten per cent line is a convention that the financial press and the industry settled on, useful as shorthand and carrying no more authority than that. Treating it as a technical threshold gives it a precision it was never granted. There is a second vagueness built into both labels, and it explains why outlets announce the same milestone on different days. The regulator's wording refers to [a broad market index](https://www.investor.gov/introduction-investing/investing-basics/glossary/bear-market) without naming one, so the answer changes depending on whether you use the S&P 500, a total market index or a technology-heavy one. It changes again depending on whether you measure from a closing high or an intraday high, and whether you count dividends. Those choices can move the trigger date by days and the threshold crossing by a percentage point or two. A label that depends on four unstated methodological decisions is not a measurement in the engineering sense. It is a reporting convention with a number in it, and asking which index and which peak somebody used is a fair first question whenever one of these headlines appears. ## A Measurement, Not a Diagnosis Both labels are produced by subtraction. Take the index's highest recent close, take today's close, work out the percentage gap. That calculation is indifferent to everything that actually matters: whether earnings are growing, whether credit is available, whether the fall is concentrated in three enormous companies or spread across five hundred. The 1987 Berkshire Hathaway letter is useful here, because it retells Ben Graham's allegory of a business partner who quotes a price every day and whose moods swing between euphoria and despair. The letter's warning about that partner is the line to keep: he is [there to serve you, not to guide you](https://www.berkshirehathaway.com/letters/1987.html), and it is his pocketbook rather than his judgement that you should find useful. A drawdown label is a transcript of that partner's mood on a particular morning. It is information about the quote, not about the business, and the two get conflated every time a threshold is crossed on a slow news day. ## The Economy Keeps a Separate Scorecard Market labels and economic labels are often used interchangeably, and they are produced by completely different processes. In the United States, recessions are dated by the National Bureau of Economic Research, whose committee [treats depth, diffusion and duration as somewhat interchangeable criteria](https://www.nber.org/research/business-cycle-dating) and relies mainly on real personal income less transfers and nonfarm payroll employment. It dates a peak and a trough after the fact, not in real time. Its published chronology is worth knowing because it puts a floor under speculation about duration. The NBER's table of expansions and contractions, last updated in March 2023, shows [the average contraction from 1945 to 2020 running 10.3 months against an average expansion of 64.2 months, with the 1854 to 2020 average contraction at 17.0 months](https://www.nber.org/research/data/us-business-cycle-expansions-and-contractions). The same table records the February 2020 peak and April 2020 trough, a contraction of two months and the shortest in the series. Two things follow. Contractions in the post-war period have been far shorter than the expansions around them, and a recession is confirmed by a committee months after it began. So anybody attaching an economic explanation to a market label on the day the label is triggered is guessing, however confident the phrasing. The forward-looking indicators people reach for instead have their own problems, covered in [what a yield curve inversion does and does not signal](/yield-curve-inversion-recession-explained). ## What Neither Label Tells You in Advance Three questions matter during a fall, and the label answers none of them. - How deep it goes. A decline crosses the ten per cent line on its way to every outcome, including a full recovery three weeks later and a multi-year slide. The threshold is a waypoint, not a forecast. - How long the recovery takes. That depends on the depth and on the earnings path afterwards, and the arithmetic is unforgiving in a way most people underestimate, as [the maths of recovering from a large drop](/stock-drops-20-percent-recovery-math-volatility) sets out. - Whether anything is actually broken. A fall driven by a repricing of interest rates is a different event from one driven by a credit freeze, and the percentage gap looks identical in both cases. ## Two Months Is Doing More Work in That Definition Than Twenty Per Cent The whole correction-versus-bear-market debate fixates on the wrong half of the definition. Everyone argues about whether the threshold should be ten or twenty per cent. Almost nobody notices that the SEC's wording also requires the move to persist for at least two months, and that condition carries more information than the percentage does. Follow the reasoning. Depth alone is produced by a single bad fortnight, and a single bad fortnight can be caused by a liquidity squeeze, a crowded trade unwinding, or an algorithmic cascade that leaves no trace in company results. Duration is much harder to manufacture. A fall that persists for months is a market steadily revising its view of future earnings rather than reacting to a shock, and a revision of that kind is the thing worth taking seriously. A two-month-old decline of twelve per cent is a more meaningful signal than a three-day decline of twenty-two. Which inverts the usual advice. Instead of asking how far the index has fallen, ask how long it has been falling and whether the reason has changed during that time. The depth question gets answered automatically as the duration accumulates. The duration question never gets answered by the depth. One finding would overturn this. If a short, violent fall reliably caused the economic damage it appeared to anticipate, then depth would be the leading indicator and duration would be confirmation arriving too late to matter. The NBER's record of post-war contractions does not support that, and the clearest counter-example sits in its own table: the two-month contraction of 2020 that accompanied the fastest market fall in living memory. The long view on these episodes is laid out in [the history of market crashes and cycles](/stock-market-crashes-cycles-history-guide). ## If You Remember One Thing A correction and a bear market are two readings on the same ruler, and the ruler measures price rather than business. The regulator's version of the deeper label includes a time condition that the headline version drops, and that condition is the part worth watching. How long these episodes have historically taken to repair is a separate question with a much better evidence base, worked through in [how long bear markets have lasted and what recovery looked like](/bear-markets-history-how-long-recovery). *This article is for educational and informational purposes only. It is not financial advice, and Investor Muse is not a registered investment advisor. Historical figures describe the past and do not predict future results. Speak with a qualified professional before you make decisions about your money.* ## FAQ ### What is the difference between a market correction and a bear market? A bear market has an official description: the SEC's investor glossary puts it at a fall of 20 per cent or more in a broad market index over at least a two-month period. A correction refers to a shallower fall, conventionally around 10 per cent, but the SEC glossary carries no entry for it. The 10 per cent figure is industry and press shorthand rather than a defined threshold. ### Does a correction always turn into a bear market? No. Every decline passes through the shallower threshold on its way to any outcome, including a quick recovery. The label marks a point already reached and carries no information about what happens next. ### Is a bear market the same thing as a recession? No. A bear market describes index prices. A recession describes economic activity and, in the United States, is dated by the National Bureau of Economic Research using depth, diffusion and duration, with peaks and troughs identified after the fact rather than in real time. ### How long do economic contractions usually last? The NBER's chronology of expansions and contractions, last updated in March 2023, shows an average contraction of 10.3 months between 1945 and 2020 against an average expansion of 64.2 months. The February to April 2020 contraction lasted two months, the shortest in a series that begins in 1854. --- Source: https://investormuse.com/blog/market-correction-vs-bear-market-difference