Bear market statistics: what the numbers say about depth, duration, and recovery
Bear markets feel like a break from normal investing, but the historical record shows they are a recurring part of market cycles, not a rare anomaly.
Table of contents
- What counts as a bear market
- How often bear markets happen
- How deep and how long they run
- What happens after the decline
- What history says about recovery
- Signals and portfolio responses
- Why bear markets matter beyond stocks
What counts as a bear market
A common market definition treats a bear market as a decline of at least 20% from a market high to a low (Fidelity).
That definition matters because it gives a clear threshold, but the historical evidence shows the path into and out of bear territory can vary a lot from cycle to cycle. Some declines are short and sharp. Others take years to unwind.
Fast facts
- A bear market is commonly defined as a drop of at least 20% from a market high to a low (Fidelity).
- Fidelity says US stocks have entered bear territory about every 6 years on average over the past 150 years (Fidelity).
- UBS says bear markets have happened about once every 7 years on average (UBS bear market guidebook, 2024).
- Since 1945, markets have spent about 31% of the time in a bear market (UBS, 2024).
- Since 1945, markets have spent about 66% of the time at or within 10% of an all-time high (UBS, 2024).
How often bear markets happen
The central takeaway from the long-run data is that bear markets are frequent enough to be expected over a normal investing lifetime.
Fidelity says US stocks have entered bear territory about every 6 years on average over the past 150 years (Fidelity). UBS gives a similar framing, saying bear markets have happened about once every 7 years on average (UBS bear market guidebook, 2024).
That rhythm does not mean the market moves in tidy six- or seven-year intervals. It means that a long-term investor should assume several bear markets will appear over time, even if there are extended stretches of calm in between.
Time spent in different market states
| Market state | Share of time since 1945 | Source |
|---|---|---|
| Bear market | 31% | UBS, 2024 |
| At or within 10% of all-time high | 66% | UBS, 2024 |
| Bull market correction | 3% | UBS, 2024 |
This table is useful because it changes the mental model. The market is not usually in a dramatic crash or a euphoric sprint. Most of the time it is either close to a high or working through a bear-cycle adjustment (UBS, 2024).
What the cycle frequency implies
- Bear markets are normal enough that they should be built into planning assumptions.
- The absence of a bear market for several years is not evidence that one has been eliminated.
- The timing is unpredictable, but the existence of the cycle is predictable.
How deep and how long they run
The depth of a bear market is what makes it emotionally and financially different from an ordinary pullback.
UBS estimates average bear-market declines of 31% in past cycles (UBS, 2024). Fidelity says previous bear markets had a median decline of 33% from recent highs (Fidelity). Bridgewater found bear markets averaged a 37% price decline (Bridgewater, 2018).
Those three figures are close enough to tell a consistent story: once a market is deep enough to qualify as a bear market, the move is often well beyond the 20% threshold that triggered the label.
Typical severity by source
| Measure | Figure | Source |
|---|---|---|
| Common definition threshold | 20% decline | Fidelity |
| Average bear-market decline | 31% | UBS, 2024 |
| Median bear-market decline | 33% | Fidelity |
| Average bear-market price decline | 37% | Bridgewater, 2018 |
Bear markets are not all alike
The UBS cycle data shows the variation clearly:
- The 1947 cycle had a max drawdown of 21.8% (UBS, 2024).
- The 1962 cycle had a max drawdown of 22.3% (UBS, 2024).
- The 1969 cycle had a max drawdown of 29.4% (UBS, 2024).
- The 1973 cycle had a max drawdown of 42.6% (UBS, 2024).
- The 1988 cycle had a max drawdown of 29.6% (UBS, 2024).
- The 2001 cycle had a max drawdown of 44.7% (UBS, 2024).
- The 2008 cycle had a max drawdown of 51.0% (UBS, 2024).
- The 2020 cycle had a max drawdown of 19.6% (UBS, 2024).
- The 2022 cycle had a max drawdown of 23.9% (UBS, 2024).
That spread matters. A market can cross the formal bear threshold with a relatively contained decline, or it can spiral into a much deeper drop that removes half the market value.
How long the prior bull market lasted
UBS says the average prior bull market length across postwar bear cycles was 9.0 years (UBS, 2024), or 9.8 years including the 2020 cycle (UBS, 2024). The average time between market cycles was 10.3 years across the sample (UBS, 2024), or 11.2 years including 2020 (UBS, 2024).
Here is the cycle-by-cycle view:
| Peak cycle | Prior bull market length | Peak to trough | Recovery to new high | Max drawdown |
|---|---|---|---|---|
| 1947 | 13.9 years | May 1946 to Nov. 1946 | Oct. 1949 | 21.8% |
| 1962 | 15.1 years | Dec. 1961 to Jun. 1962 | Apr. 1963 | 22.3% |
| 1969 | 6.4 years | Nov. 1968 to Jun. 1970 | Mar. 1971 | 29.4% |
| 1973 | 2.5 years | Dec. 1972 to Sep. 1974 | Jun. 1976 | 42.6% |
| 1988 | 12.9 years | Aug. 1987 to Nov. 1987 | May 1989 | 29.6% |
| 2001 | 12.8 years | Aug. 2000 to Sep. 2002 | Oct. 2006 | 44.7% |
| 2008 | 5.1 years | Oct. 2007 to Feb. 2009 | Mar. 2012 | 51.0% |
| 2020 | 10.8 years | Dec. 2019 to Mar. 2020 | Jul. 2020 | 19.6% |
| 2022 | 1.8 years | Dec. 2021 to Sep. 2022 | Dec. 2023 | 23.9% |
The table shows that prior bull-market length is not a direct warning signal. A long run can still end with a modest bear cycle, and a shorter run can be followed by a severe one.
What happens after the decline
A bear market is not just a fall in price. It is often accompanied by broader stress in earnings, risk premiums, and the real economy.
Bridgewater found earnings fell in 75% of bear markets (Bridgewater, 2018), and those earnings declines averaged 28% in bear markets (Bridgewater, 2018). Bridgewater also found the real economy averaged a 5% decline versus potential in bear markets, compared with only a 1% decline versus potential in corrections (Bridgewater, 2018).
That distinction helps explain why bear markets can feel so different from simple valuation resets. The market is often responding to a broader deterioration in business conditions, not just investor mood.
The role of earnings and risk premiums
- Earnings fell in 75% of bear markets in Bridgewater’s study (Bridgewater, 2018).
- Corrections averaged a 15% price decline, far less severe than bear markets (Bridgewater, 2018).
- Corrections also saw milder damage to the real economy than bear markets (Bridgewater, 2018).
- Risk premiums declined by 18% on average after the peak in bear markets, versus 10% in corrections (Bridgewater, 2018).
This is one reason bear markets tend to be longer and more consequential. They are often tied to a more complete repricing of growth, profits, and discount rates.
What history says about recovery
The most important historical point is not that bear markets happen. It is that they have historically recovered.
Fidelity says the market has historically recovered from every bear market (Fidelity). UBS says bear-market drawdowns are over within 1 year on average (UBS, 2024), and then it takes about 2 more years on average to recover those losses (UBS, 2024). UBS also says bear-market recoveries usually take 3 to 5 years to fully recover (UBS, 2024).
Recovery by cycle
| Peak cycle | Trough date | New all-time high date | Years to full recovery | Gains erased by drawdown |
|---|---|---|---|---|
| 1947 | Nov. 1946 | Oct. 1949 | 3.4 | 1.2 years |
| 1962 | Jun. 1962 | Apr. 1963 | 1.3 | 2.9 years |
| 1969 | Jun. 1970 | Mar. 1971 | 2.3 | 5.4 years |
| 1973 | Sep. 1974 | Jun. 1976 | 3.5 | 9.7 years |
| 1988 | Nov. 1987 | May 1989 | 1.8 | 1.5 years |
| 2001 | Sep. 2002 | Oct. 2006 | 6.2 | 5.3 years |
| 2008 | Feb. 2009 | Mar. 2012 | 4.4 | 11.6 years |
| 2020 | Mar. 2020 | Jul. 2020 | 0.6 | 2.2 years |
| 2022 | Sep. 2022 | Dec. 2023 | 2.0 | 1.6 years |
A few patterns stand out from this history:
- Fast recoveries are possible, as the 2020 cycle shows with a 0.6-year return to a new all-time high (UBS, 2024).
- Deeper or more economically damaging bear markets usually take longer to repair, as the 2001 and 2008 cycles show (UBS, 2024).
- The return to a new high can happen long after the trough, which means the emotional bottom and the wealth-recovery point are not the same thing.
Post-bear performance
S&P Dow Jones Indices found the S&P 500 gained an average of 15% over three years after entering a bear market in a study covering 12 bear-market slumps from 1929 to 2019 (S&P Global, 2023).
They also found that over one year after entering a bear market, stocks gained 9% on average, with only 5 losses out of 13 cases (S&P Global, 2023).
A few additional data points sharpen the picture:
- The post-2020 rally surged 59% in 12 months after the COVID bear market (S&P Global, 2023).
- The worst three-year post-bear result in that study was a 75% plunge after the 1929 crash (S&P Global, 2023).
- The best three-year post-bear bounce in that study was a 63% surge through 1985 (S&P Global, 2023).
Signals and portfolio responses
Bear markets are not just a history lesson. The statistics also show how markets, policy, and portfolio tools interact during stress.
UBS says a dynamic 200-day moving average strategy holds 100% S&P 500 normally (UBS, 2024). When the S&P 500 falls below its 200-day moving average, UBS says that strategy shifts to 50% S&P 500 and 50% intermediate Treasuries (UBS, 2024).
That kind of rule-based response does not eliminate drawdowns, but it shows how market signals can be used to change exposure rather than simply endure volatility.
Defensive and tactical data points
- Long-duration Treasuries rallied during 70% of equity selloffs (UBS, 2024).
- Long-duration Treasuries averaged a 6% to 8% gain during those equity selloffs (UBS, 2024).
- UBS estimates tax-loss harvesting can add about 0.5% to after-tax annual portfolio returns (UBS, 2024).
- UBS says a five-year structured note example can provide 80% of MSCI Emerging Markets upside (UBS, 2024).
- UBS says that same note can protect against the first 10% of index loss if held to maturity (UBS, 2024).
These figures point to a broader lesson: bear markets are where portfolio design matters most. Exposure, hedging, rebalancing, and tax management all become more visible when prices are falling.
Why bear markets matter beyond stocks
Bear markets affect more than portfolio balances. They can influence policy expectations, sentiment, and cross-asset pricing.
NBER finds that a 10% stock market decline predicted a 32 bp federal-funds-rate cut at the next FOMC meeting since 1994 (NBER). The same 10% stock drop predicted a 127 bp cut after one year (NBER). NBER also says lagged stock returns explain 38% of the variation in growth-expectation updates since 1994 (NBER).
That means equity declines can feed into macro expectations rather than sit in isolation. NBER also says around 80% of FOMC stock-market mentions framed the stock market as driving the economy (NBER), which underlines how central markets can become in the policy conversation.
Why the statistics matter for planning
- The average bear market is not just a valuation dip; it is often a broader cycle with earnings and economic consequences (Bridgewater, 2018).
- The market has historically recovered, but the path back can be fast or painfully slow depending on the cycle (Fidelity; UBS, 2024).
- Some bear markets are brief, but the deeper ones can erase many years of prior gains (UBS, 2024).
- Defensive positioning, disciplined rebalancing, and tax-aware moves can matter when volatility spikes (UBS, 2024).
The data does not eliminate uncertainty. It does, however, make one thing clear: bear markets are recurring, measurable, and survivable when viewed through a long enough time horizon (Fidelity; UBS, 2024).