Why Markets Often Rise on Bad News

Markets can rise on bad news because a stock price reflects expected future cash flows discounted at a rate that depends on interest rates and on the extra return investors demand, and a weak report can move the second part more than it damages the first. It can also be old news by the time it is published, or it can resolve an uncertainty that had been weighing on prices. None of this is guaranteed: the same headline can send stocks down when it points to a recession.

Key takeaways

Prices respond to the surprise relative to what was expected, not to the headline itself, so a bad number that is better than feared can lift stocks.
Weak economic data tends to raise expectations of lower interest rates. Peer-reviewed work finds that this rate channel dominates stock reactions to unemployment news in expansions, while the earnings channel dominates in contractions.
Stock lows have come before the official end of the last two US recessions: March 2009 versus a June 2009 trough, and March 2020 versus an April 2020 trough.
On May 8, 2020, the day a record 20.5 million payroll drop was reported, the S&P 500 closed up about 1.7%.
A rally on bad news is not a signal of anything lasting. Short covering and relief can lift prices briefly without changing the outlook.

Weak data and the odds of easing

A share is worth the cash it is expected to pay out, discounted back to today at a rate tied to prevailing interest rates. When data come in weak, two things can change at once: expected profits fall, and expected interest rates fall. The first lowers the value of a share; the second raises it. Which wins depends on the circumstances.

Researchers John Boyd, Jian Hu and Ravi Jagannathan studied this in an NBER working paper later published in the Journal of Finance (2005). They found that, on average, news of rising unemployment tends to lift stocks during expansions and lower them during contractions. The reason is that the announcement bundles two signals. A weaker labor market points to lower interest rates, which helps stocks, and to weaker future earnings and dividends, which hurts them. For stocks as a whole, the interest-rate signal dominates in expansions and the earnings signal dominates in contractions. Because expansions are the more common state, the pattern shows up often enough to earn the nickname "bad news is good news."

On Fed decision days, a Federal Reserve Board working paper by Ben Bernanke and Kenneth Kuttner (2004, later published in the Journal of Finance) puts the average reaction to an unanticipated 25-basis-point cut at roughly a 1% rise in broad stock indexes, mostly through the extra return investors demand for holding stocks. Only the part of a move that was not already priced in counts.

A simple illustration

Illustrative example with round numbers, not market data. Suppose that before a jobs report, traders put a 20% chance on a 25-basis-point rate cut at the next Fed meeting. A weak report pushes that chance to 60%. The expected cut changes from 0.20 × 25 = 5 basis points to 0.60 × 25 = 15 basis points, a shift of 10 basis points toward easier policy. Stocks can rise on that shift even though the report itself describes a slower economy, provided investors judge that the lower-rate effect outweighs the earnings hit. If the same report instead made a recession look likely, the earnings side would grow and the net effect could flip.

The same logic applies to jobs and inflation releases generally: a number matters partly because of what it implies for the Fed's next move.

News is judged against expectations

A reported figure is only half of a market reaction. The other half is the consensus forecast that was already built into prices. Prices tend to move on the gap, not the level.

What was expected	What was reported	Likely reading
Weak	Weak, but less bad than expected	Positive surprise; stocks can rise
Weak	Weak, as expected	Little new information; small move
Strong	Strong, as expected	Little new information; small move
Strong	Weak	Negative surprise for earnings; may be offset if it raises easing odds

The table is a framework, not a prediction. A report can be better than the consensus on one measure and worse on another, and the market decides which matters more that day.

Relief rallies and short covering

Traders say "buy the rumor, sell the news" about good news that is priced in before it arrives. For feared events, some traders flip it: sell the rumor, buy the news. Both are market sayings, not documented rules, but the second captures something real: when a widely feared event arrives, the uncertainty is gone. If the outcome is no worse than the fear, prices can rise on resolution alone. That is a relief rally.

Positioning adds to it. A short sale is the sale of shares the seller does not own, usually borrowed, with the plan to buy them back later. The SEC's investor bulletin on short sales notes that if the price rises, the short seller loses money buying shares back to replace those borrowed. When bad news is less bad than feared and prices tick up, short sellers who buy to close their positions add to the demand. The effect can lift prices briefly and then fade, because it comes from trades being closed rather than from changed views about the business.

Stocks look ahead of the economy

Stock prices are set by expectations of the future, so they tend to turn before the economy does. The National Bureau of Economic Research (NBER) business cycle dates show how large the gap can be.

2009 and 2020

The NBER dates the 2007–2009 recession as running from December 2007 to a trough in June 2009, 18 months. The S&P 500's bear-market low came on March 9, 2009, according to Yardeni Research's bull and bear market tables (Standard & Poor's and Haver Analytics data). That was three months before the economic trough. The NBER announced the June 2009 trough on September 20, 2010, about 18 months after the stock low. It added that the trough did not mean conditions had returned to normal.

In 2020, the NBER dates a two-month recession from February to April. FRED data for the S&P 500 show a closing low of 2,237.40 on March 23, about 33.9% below the February 19 close. The index closed at 2,447.33 the next day, up about 9.4%, and at 2,912.43 on April 30, about 30.2% above the March 23 close. That is the month the NBER later identified as the economic trough, in an announcement on July 19, 2021, roughly 16 months after the stock low.

The day record job losses were reported

The Bureau of Labor Statistics employment report released on May 8, 2020 showed the unemployment rate at 14.7% for April and total nonfarm payroll employment down 20.5 million, both the worst readings in the history of those series. The S&P 500 closed at 2,929.80 that day, up from 2,881.19 the day before, a gain of about 1.7% (FRED). By then the index was already about 31% above its March 23 close, so the report landed after much of the damage had been priced in.

The policy backdrop mattered too. On March 15, 2020 the FOMC lowered the federal funds target range to 0 to 1/4 percent, and on March 23, 2020 it said it would keep buying Treasury and agency mortgage-backed securities "in the amounts needed to support smooth market functioning." That statement came out at 8 a.m., before the open, yet the S&P 500 still closed 2.9% lower that day, at its low. The 9.4% gain came on March 24. Prices cannot separate the Fed's actions from other forces, so the timing is consistent with a policy effect without proving one.

When bad news is simply bad

Bad news stays bad when it signals that earnings will fall by more than lower rates can offset, which is the contraction case in the unemployment research above. From its October 9, 2007 close of 1,565.15 to the March 9, 2009 close of 676.53, the S&P 500 fell about 56.8% (Yardeni Research tables), through many rounds of weak economic reports. Over the same dates the effective federal funds rate fell from 4.91% to 0.20%, so easier policy did not offset the earnings damage.

Reasoning from the two channels, several conditions make a rally on bad news more likely to be a poor signal:

The Fed cannot or will not respond, for example when inflation is the main concern.
The data point to a broad earnings decline rather than a soft patch.
Credit stress is spreading, so lower rates do not reach borrowers.
The rise was driven by closing short positions, which fades once those trades end.

Frequently asked questions

Does weak data always lift stocks?

No. In the research above, the lift shows up on average in expansions rather than contractions, and nobody knows the regime for sure in real time. The NBER's dating-procedure FAQ says its committee waits "long enough to avoid any doubt" before naming a peak or trough, so investors reacting to a report are judging the state of the economy without an official answer.

Can a large rate cut come with falling stocks?

Yes. The FOMC cut its target range to 0 to 1/4 percent on Sunday, March 15, 2020, and the S&P 500 fell from 2,711.02 to 2,386.13 the next trading day, about 12.0% (FRED). A cut that large and that sudden can also tell investors how worried the Fed is, so the average reaction found in research is not a rule for any single day.

Can good news push stocks down?

The same rate channel can work in reverse. Strong data that makes rate cuts look less likely, or rate increases more likely, can weigh on stock prices even though it points to healthier earnings. Whether it does depends on which effect investors judge to be larger that day.

What is a short squeeze?

It is a rise in a stock's price that pushes short sellers to buy back shares to limit their losses, and that buying pushes the price up further. Because a short seller's loss grows as the price rises, there is no fixed ceiling on it. A squeeze can follow bad news that turns out less severe than feared.

Why do the official recession dates arrive so late?

The NBER's committee has no fixed timetable and allows for standard data revisions before it dates a turning point. According to the same FAQ, announcements have come as soon as four months after a peak (February 2020) and as late as 21 months after a trough (March 1991, announced in December 1992).

A checklist for reading headlines

Find the consensus forecast for the release and compare it with the actual figure, not with last month's number.
Ask what the report implies for interest-rate expectations, and whether the implied move helps or hurts valuations.
Check whether the economy is in an expansion or showing signs of contraction, since the same news can have opposite effects.
Look at what happened over several days and weeks, not only the first hour, to separate short covering from a lasting change.
Treat any single-day reaction as a description of that day, not as evidence of what comes next.