The 2008 Financial Crisis: Timeline, Drawdown and Recovery

Key Takeaways

• The Federal Reserve's own account of the episode reports that real gross domestic product fell 4.3 percent from its 2007 fourth-quarter peak to its 2009 second-quarter trough, the largest decline of the postwar era on the data available when that account was written.
• The equity decline was deep but not fast. Computed from daily closes, the S&P 500 fell 56.8 percent over 355 trading days, roughly seventeen months of grinding losses rather than a single crash week.
• Recovery took longer than the decline by a wide margin. The index first closed back at its October 2007 level on 28 March 2013, about five and a half years after the peak.
• The most-cited advance warning, an inverted Treasury yield curve, was visible unusually early. The 10-year yield closed below the 3-month yield on 124 trading days during 2006 and 119 during 2007, and the inversion had ended by 27 August 2007, before the market peak.
• Housing weakness was public information well before equities peaked. New Century Financial, then a leading subprime lender, filed for bankruptcy in April 2007, six months before the S&P 500 high.
• What was genuinely not visible in advance was the size and interconnection of the exposures. That is the part of 2008 that hindsight makes look obvious and that contemporaneous disclosure did not support.

What Happened in the 2008 Financial Crisis?

The 2008 financial crisis is the popular name for a sequence that started earlier and finished later than the year it is named after. The Federal Reserve's historical account of the subprime episode dates the turmoil in financial markets from 2007 to 2010, and the National Bureau of Economic Research (NBER), which is the recognized arbiter of United States business cycle dates, places the recession from December 2007 to June 2009.

The chain of events ran roughly as follows. Mortgage credit expanded to borrowers who previously would have struggled to obtain a loan, funded largely through private-label mortgage-backed securities. Rising house prices made that lending look safe, because a borrower in trouble could refinance or sell. When house prices stopped rising, both of those exits closed at once. Loss rates rose, the bond funding for new subprime lending collapsed, lenders stopped originating those loans, housing demand fell further, and prices fell further still. The Federal Reserve describes this as a self-reinforcing loop, with expectations of further declines feeding the declines themselves.

The damage then moved from mortgages into institutions. In April 2007, New Century Financial Corporation, a leading subprime lender, filed for bankruptcy. Large numbers of private-label mortgage-backed securities were downgraded to high risk and several subprime lenders closed. Fannie Mae and Freddie Mac, which had issued debt to fund purchases of subprime mortgage-backed securities and also held failing prime mortgages, were placed into federal conservatorship in the summer of 2008. What began as a question about the value of a class of bonds became a question about the solvency of the firms holding them, and then a question about whether any of those firms could fund themselves overnight.

Chronology of the market phase

Selected dated events and the S&P 500 close on that date. Index closes computed from daily closing values.

Two features of that sequence matter more than the individual dates. The first is that the equity peak came ten months after the first public subprime bankruptcy and two months before the recession officially began. The second is that the worst single days clustered a full year after the peak, in the autumn of 2008. A reader who imagines the crisis as a crash followed by a recovery has the shape wrong. It was a long decline with an acute funding panic embedded in the middle of it.

What Did the Setup Look Like Before the Crisis?

Three conditions defined the pre-crisis environment, and only one of them was widely treated as a risk at the time.

A long expansion with low measured volatility. The NBER dates the preceding expansion from November 2001 to December 2007, six years of growth. The unemployment rate fell to 4.4 percent in March 2007, its low for that cycle. Nothing in the labor data pointed at a coming contraction.

A rate cycle that had already peaked. The Federal Reserve's record of target changes shows the funds rate reaching 5.25 percent in June 2006 and staying there until September 2007. On the Treasury side, the 10-year yield closed below the 3-month yield on 124 of 250 trading days during 2006 and on 119 of 251 during 2007. An inverted curve of that persistence is the classic recession precursor, and it was fully public.

A housing and credit structure whose risk was not visible from the outside. This is the condition that made 2008 different. The Federal Reserve's account describes private-label mortgage-backed securities as providing most of the funding of subprime mortgages, with the less vulnerable tranches viewed as low risk either because they were insured with new financial instruments or because subordinate securities would absorb losses first. Both of those protections depended on the same assumption, that house price declines would stay local rather than national. Nothing in a public filing let an outside investor test that assumption directly.

If you want to understand why a rate cycle and a credit cycle can look unrelated right up to the moment they are not, the mechanism is set out in the credit cycle and refinancing risk and in financial conditions, credit spreads and liquidity.

How Far and How Fast Did Prices Fall?

The Federal Reserve's own summary states that the S&P 500 fell 57 percent from its October 2007 peak to its trough in March 2009. Computed from daily closing values, the exact close-to-close figure is 56.8 percent. Those are the same number reported at different precision, not a disagreement.

Peak-to-trough decline by index. Computed from daily closing values; close to close, price only.

Notice how similar those three numbers are. In 2008 the technology-weighted index fell about as much as the broad market and about as much as the industrial average. That is unusual, and it is the fingerprint of a crisis transmitted through funding rather than through a single overvalued sector. Compare that with the dot-com bubble, where the Nasdaq Composite fell roughly twice as far as the Dow.

The pace of the decline

Seventeen months of decline does not mean seventeen months of steady losses. Computed from daily closes, the four worst single sessions of 2008 were 15 October at 9.03 percent, 1 December at 8.93 percent, 29 September at 8.81 percent and 9 October at 7.62 percent. All four fell in the ten weeks after the Lehman Brothers filing. The Cboe Volatility Index reached its highest close of the year on 20 November 2008 at 80.86.

The Treasury market recorded the same panic from the other side. On 2 January 2008 the 3-month bill yielded 3.26 percent and the 10-year note 3.91 percent. On 15 September 2008, the day Lehman Brothers filed, the 3-month was 1.02 percent. By 31 December 2008 the 3-month had reached 0.11 percent and the 10-year 2.25 percent. Money was not being reallocated between risky and safe assets in an orderly way. It was being parked in the shortest, most certain instrument available at almost any price.

Which Warning Signs Were Visible in Advance, and Which Only in Hindsight?

This is the section most retrospectives get wrong, so it is worth being precise. Some of the 2008 evidence was genuinely public and genuinely early. Some of it is only legible because we know how the story ended.

Signals classified by whether they were usable at the time.

The honest version of the yield curve story is more uncomfortable than the popular one. The signal fired, it fired for a long time, and it fired far too early to act on with any precision. An investor who moved to cash on the first inverted close of January 2006 spent nearly two years watching the S&P 500 rise before being proved right. That is not a refutation of the indicator. It is a statement about what kind of information it is. The mechanics and the base rates are covered in the yield curve, term premium and recession signals.

The genuinely unavailable information was structural. A reader in 2007 could know that subprime lending had grown, that some lenders had failed, and that house prices had turned. What no public disclosure allowed was a reliable estimate of how much of the loss would land on which balance sheets, or how much short-term funding those balance sheets depended on. Retrospectives that describe the crisis as foreseeable are usually describing the first set of facts and quietly borrowing certainty from the second.

Hindsight check. Ask of any 2008 warning sign: could a person acting only on information published before October 2007 have sized a position from it? For the yield curve, the answer is a qualified yes with terrible timing precision. For the interconnection of counterparty exposures, the answer is no. Treating both as equally available is the core hindsight error, and it is the reason so many crisis narratives produce confidence rather than caution. Cognitive biases in trading covers why the effect is so persistent.

Why Did a Housing Problem Become a Market Problem?

United States residential mortgage losses, on their own, were not large enough to threaten the global financial system. What made 2008 systemic was the way those losses were funded and distributed.

Losses were concentrated in leveraged institutions rather than spread across end investors. When a diversified household loses money on a bond fund, it consumes slightly less. When a leveraged institution loses money on a bond portfolio, it must either raise capital or shrink its balance sheet, and shrinking means selling. Selling pushes prices down, which creates the next loss for the next institution.

Funding was short and collateral-dependent. A large part of the securitized credit system financed long-dated assets with very short-term borrowing secured against those same assets. Once the assets became hard to value, the collateral became hard to accept, and the funding disappeared over days rather than quarters. Solvency and liquidity stopped being separable questions.

Correlation assumptions failed at exactly the wrong moment. Structures whose safety depended on losses staying regionally uncorrelated failed when the decline turned out to be national. This is not a subtlety specific to 2008. It is the general property that diversification measured in calm periods overstates diversification available in stressed ones, which is treated directly in how correlations change across regimes.

The practical consequence for a portfolio holder is that in 2008 the things that usually offset each other stopped doing so. Broad equities, credit, real estate and most active strategies fell together. Short-dated Treasury securities and cash did not. That is a narrower set of effective hedges than most pre-crisis portfolios assumed they held.

How Did the Federal Reserve Respond?

The Federal Reserve's published record of target rate changes shows the full arc. Rates came down slowly for a year, then very quickly.

Federal funds target rate changes, September 2007 to December 2008, from the Federal Reserve Board's record of open market operations.

The Federal Reserve's own retrospective summarizes the move as a reduction from 5.25 percent in September 2007 to a range of 0 to 0.25 percent in December 2008, with much of the reduction concentrated in January to March 2008 and September to December 2008. It also notes that policy did not stop at the rate. Once the funds rate reached its lower bound, the Committee turned to forward guidance, first calendar-based and later tied to unemployment and inflation thresholds.

Two details are worth carrying forward. First, the first cut came three weeks before the market peak, so easing did not mark the bottom. Second, the funds rate reached zero in December 2008 and the equity trough did not arrive until March 2009, roughly three months later. Policy action and price bottoms are not the same event, and the gap between them was measured in months. Federal Reserve policy rates and forward guidance covers how the transmission actually works.

How Long Did the Recovery Take?

Recovery time depends entirely on what you count. Below is the price-only measure, which is the strictest and the one most people mean when they ask how long it took to get back.

Time from peak close back to that same close. Computed from daily closing values; price only, dividends excluded.

Three qualifications belong with that table, and leaving them out is how recovery statistics become misleading.

Dividends shorten it. These are price indexes. An investor holding a fund that reinvested dividends recovered earlier than these dates, because the reinvested income lowered the price level needed to break even. We do not state a total-return recovery date here because we did not verify one from a source this session.

Inflation lengthens it. Nominal recovery is not purchasing-power recovery. A portfolio that returned to its 2007 nominal value in 2013 had lost real value over the period.

Recovery of an index is not recovery of a portfolio. An investor who sold near the trough never got the recovery at all. An investor who was adding money throughout bought the whole decline at lower prices and recovered much sooner. An investor drawing down a portfolio in retirement experienced something different again, and considerably worse, which is exactly the mechanism described in sequence of returns risk.

The real economy recovered on its own timetable, and slower. The unemployment rate was 5 percent in December 2007, reached 9.5 percent in June 2009, and peaked at 10 percent in October 2009, four months after the recession officially ended. It was still 9.3 percent in December 2010. Labor market repair lagged the equity market by years.

What Was Specifically Different About 2008?

Every episode in this library has a feature that does not repeat. For 2008 there are four.

The transmission ran through bank and shadow-bank funding. Most equity bear markets are repricings of expected earnings or discount rates. This one was a question about whether counterparties would still be there tomorrow. That changes which assets protect you, because in a funding crisis correlations converge and only the shortest, most certain instruments hold up.

Losses landed on institutions that could not absorb them without selling. The leverage sat where forced selling was mechanically required, which converted a credit loss into a market-wide liquidation.

Nominal policy rates started high, so there was room to cut. The funds target went from 5.25 percent to zero. That is 525 basis points of conventional easing. In 2020 the same institution had roughly 150 basis points available before hitting the floor. A crisis that begins from a low rate level cannot be met with the same conventional response, and this is why 2008 is a poor template for the policy path of later episodes.

Inflation was not a constraint. The Federal Reserve was able to ease aggressively because prices were falling, not rising. That is the precise opposite of the constraint faced in the 2022 rate shock, where the same institution was raising rates into an equity and bond decline. Any lesson of the form "the Fed will step in" carries an unstated inflation assumption.

Common Myths About the 2008 Crisis

"Everyone knew it was coming." Some people identified specific mispricings in securitized mortgage credit and were correct. That is not the same as the information being generally available. The scale of counterparty exposure was not disclosed in a form that supported an outside estimate, and the timing of the unwind was not predictable even for those who called the direction. Survivorship in the retelling does most of the work here: the correct calls are remembered and the equally confident incorrect ones are not.

"The crash happened in September 2008." The S&P 500 peaked in October 2007 and had already fallen about 20 percent before the Lehman Brothers filing. September and October 2008 produced the most violent days, not the beginning of the decline.

"Buying the dip worked." It worked from March 2009. It did not work from January 2008, from July 2008, or from September 2008, each of which looked like a dip at the time and was followed by substantially lower prices. The statement is only true with the trough date supplied in advance, which is precisely the information nobody had.

"Stocks always come back within a few years." The S&P 500 took five and a half years to recover on price. In the dot-com episode the Nasdaq Composite took just over fifteen. In 1929 the Dow took until November 1954. Recovery times are not a stable parameter and should never be planned around as though they were.

"Diversification failed." Diversification across equity sectors and across most credit largely failed, because those exposures share a common factor that dominates in a funding crisis. Diversification into short-dated government debt and cash did not fail. The lesson is about which diversification, not whether.

What a Reader Can Actually Carry Forward

The point of studying 2008 is not to be able to recognize the next one. The next one will be transmitted differently, and pattern-matching to 2008 is how investors mispriced 2020 and 2022. The point is to extract the things that were true regardless of the specific trigger, and to be explicit about the things that were specific to this episode.

What generalizes

• Recovery time is a planning input, not a footnote. The relevant question is not whether a portfolio recovers but whether the holder can wait. A five-year price recovery is survivable for a saver with income and a long horizon, and can be permanent damage for someone drawing down. Build the horizon into the allocation before the drawdown, because it cannot be added afterwards.
• Leverage decides whether a loss is a loss or a liquidation. The same 30 percent decline is an unpleasant year for an unlevered holder and a forced exit for a levered one. This applies at every scale, from a global bank to a margin account. See risk management.
• Correlations measured in calm periods overstate the protection available in stressed ones. Test an allocation against the correlations of a crisis, not the correlations of the last three years. Stress testing and scenario analysis covers how to run this properly.
• Policy support and price bottoms are separate events. The funds rate hit zero in December 2008. The equity trough was in March 2009. Waiting for the announcement is not the same as waiting for the low.
• A signal that is right about direction can be useless about timing. The 2006 curve inversion was correct and roughly two years early. An indicator you cannot size a position from is context, not a trigger.

What does not generalize

• The specific 56.8 percent depth. Drawdowns in this library range from about 21 percent to about 89 percent. There is no typical figure to plan around.
• The five-and-a-half-year recovery. The COVID decline recovered in six months. The Nasdaq Composite after 2000 took fifteen years. Same market, same country, wildly different durations.
• The aggressive policy response. It was available because inflation was not a constraint and rates started at 5.25 percent. Neither condition is guaranteed.
• "Housing is the risk." The 2008 fault line was in one specific credit structure. Fighting the last crisis is the most reliable way to be surprised by the next one.

The one question worth asking now

Rather than "am I positioned for the next 2008," a more useful exercise is: if the assets you hold fell by half and stayed down for five years, what would you be forced to do? If the honest answer involves selling, the allocation is too aggressive for the horizon regardless of what any historical average says. That question is answerable today, from your own cash flows, and it does not require predicting anything. It is the part of 2008 that is genuinely portable.

References

Every figure on this page was verified against the following sources, each retrieved on 23 August 2026:

• Federal Reserve History: The Great Recession: the December 2007 to June 2009 recession dates, the 4.3 percent real gross domestic product decline from the 2007 fourth quarter to the 2009 second quarter, the unemployment path from 5 percent to a 10 percent October 2009 peak, the approximately 30 percent decline in home prices from the mid-2006 peak, the 57 percent S&P 500 decline, the fall in household and nonprofit net worth from approximately 69 trillion dollars in 2007 to approximately 55 trillion in 2009, and the description of the funds rate path and subsequent forward guidance.
• Federal Reserve History: The Subprime Mortgage Crisis: the 2007 to 2010 dating of the market turmoil, the role of private-label mortgage-backed securities in funding subprime lending, the April 2007 New Century Financial bankruptcy, the subsequent downgrades and lender closures, and the summer 2008 conservatorship of Fannie Mae and Freddie Mac.
• National Bureau of Economic Research: US Business Cycle Expansions and Contractions: the December 2007 peak and June 2009 trough, and the November 2001 start of the preceding expansion.
• Federal Reserve Board: Open Market Operations: every federal funds target change and resulting level in the 2007 and 2008 table above.
• US Department of the Treasury: Daily Treasury Par Yield Curve Rates: the 3-month and 10-year yields quoted for 2006, 2007 and 2008, and the count of inverted trading days, computed from the published daily series.
• US Bureau of Labor Statistics: Unemployment Rate, Series LNS14000000: the monthly unemployment rates quoted, including the 4.4 percent March 2007 low, the 10 percent October 2009 peak and the 9.3 percent December 2010 reading.

Figures deliberately not stated. This page does not give a total-return recovery date, a dollar figure for aggregate financial-sector losses, or a count of bank failures, because no source verified in this session supplied them. Where the shape of an event is known but the number is not, the shape is described and the number is left out rather than estimated.

Method note: index peak, trough, decline and recovery figures labeled as computed were derived by Swoopr Investment from daily closing values of the named index, retrieved from the Yahoo Finance historical chart API on 23 August 2026. Drawdowns are measured close to close, not intraday, so the intraday low of any episode is lower than the trough shown. Recovery means the first daily close at or above the prior peak close, price only, with no dividends reinvested. Figures labeled total return are computed instead from dividend-adjusted closing prices and are stated as such wherever they appear.

This is educational content about a historical episode. It is not investment advice, it is not a forecast, and nothing here should be read as a claim about how any future market decline will behave.

Frequently Asked Questions