How Lawsuits Affect Stock Prices: 23 Real Cases Tested

On the day the market first heard about the scandal, the median stock in my sample fell 8.5%. A year later it was 20.6% below its pre-news price. That is the part of how lawsuits affect stock prices that surprised me most: the first day was not the worst of it. In most of these cases the damage kept building for months after the headline, while on the day the company finally agreed to pay, the stock rose a median of 1.2%.

I went into this expecting the opposite. The usual line in investor circles is that fines are noise, the market overreacts, and patient owners get the drop back. So I took 13 of the best-known corporate legal disasters of the last 15 years (Volkswagen, BP, Boeing, Wells Fargo twice, Equifax, Facebook, PG&E, Johnson & Johnson, Deutsche Bank, Bayer, Tesla, Apple) and followed each stock’s price from the close before the news. Then I took 10 of the final settlements for the same stories and did the same thing from the day the number became official.

Every date below comes from a regulator’s release, a company filing, or contemporaneous reporting, and every return comes from dividend-adjusted daily closes. Nothing here is estimated or remembered.

How I measured it

This is a classic event study, kept simple so anyone can check it.

Horizons are day 0, one week (5 sessions), one month (21), and one year (252). The two groups answer different questions: the first headline is the day the market learned something new, and the final settlement is the day the price tag became official for a story everyone already knew.

How lawsuits affect stock prices on the first headline

One definition matters before the numbers. “First headline” is the day the legal problem became public, not the day a lawsuit was filed in court. Only two of the 13 are court events (the SEC suing Musk and the first Roundup verdict). The rest are a regulator acting, a company disclosing, a newspaper investigation, or an accident that the lawsuits followed. That is deliberate: the market trades the moment it learns about the exposure, and the court docket usually comes later.

Here are all 13 first headlines. The day-0 moves line up with what was reported at the time (J&J closed down 10.0% on the Reuters talc story, Bayer 10.3% on the Roundup verdict, Equifax 13.7% on the breach), which is a useful check that the dates are right.

Two things jump out. First, the drop kept going. The median was -8.5% on day 0, -12.4% after a week, -11.4% after a month and -20.6% after a year, so more than half of the one-year damage arrived after the first session. Even buying at the close of day 0, after the first drop, would have lost money over the next year in 8 of 13 cases, by a median of 7.2%. Apple (+56%) and Deutsche Bank (+30%) rewarded that trade, and Equifax, J&J and Wells Fargo’s first fine made modest gains.

Second, some of the worst outcomes started with almost no reaction at all. BP fell 0.6% the day after the rig exploded, because nobody yet knew the well would leak for close to three months. PG&E lost 2.0% on the day the Camp Fire started; it filed for bankruptcy less than three months later. Wells Fargo actually closed up on the $185 million fine, and the story only grew from there. The first price is the market’s guess at a number it cannot see yet.

Is that just normal single-stock noise?

A fair objection: one stock swings a lot in a year, so maybe -20% is ordinary. I tested that directly. I took the same 13 stocks, picked random days at least a year away from any of the events, and measured the same horizons, 5,000 times.

On ordinary days these same companies rose a median 14.3% over a year, and not one of the 5,000 random samples came anywhere near a one-year median of -20.6%. So the weakness is real, and it is not a quirk of which companies I picked. It could still be a quirk of which stories I picked, which I come back to in the limits section.

The settlement day: the cheque was already priced

The second group flips the question. By the time a company signs, everyone knows the story. What is left to learn is the number and whether the threat is over.

Six of the ten moved less than 2% on the day. Deutsche Bank’s settlement came in at roughly half the $14 billion the DoJ had demanded, and the stock moved 0.3%. Facebook had already told investors in April 2019 to expect up to $5 billion, so the $5 billion headline was confirmation, not news.

The three big moves are the interesting ones, because they were not about the amount. Tesla jumped 17% because the settlement meant Musk would not be removed as CEO. PG&E jumped 16% because the $13.5 billion deal put a figure on liabilities that had pushed it into bankruptcy. BP rose 5% because the $18.7 billion settlement closed off years of open-ended claims. In each case the market was paying for a ceiling, not celebrating a bill.

What separated a shrug from lasting damage

I expected size to be the answer, with mega-caps shrugging and smaller names getting hurt. The data does not support that. Apple shrugged, but Johnson & Johnson, then one of the largest companies in the world, lost about $40 billion of market value on the talc report and needed a full year to get back to its pre-news price. Boeing is a giant too, and it was 45% lower a year after the second crash.

Two questions sorted the cases much better.

Recovery times make the same point. The first column is the first day the stock closed back at or above its pre-news price; the second says whether it was still there a year after the headline.

“Not yet” means not by 29 September 2026, my last price. Wells Fargo’s fake-accounts fine is left out because the stock closed up on day 0, so there was no drop to recover from. Boeing, Bayer and PG&E have not climbed back to their own pre-news price, seven to eight years on.

Only Apple and Deutsche Bank got back quickly and stayed there, and in both the threat was bounded from the start. Tesla and Facebook are the trap in this table: both were back within weeks and then fell again. Tesla’s slide came from production and demand worries, not the SEC. Facebook fell 19.0% in one day on 26 July 2018 when the company guided to slower growth and higher spending, including on privacy and security, and it was 12.7% below its pre-news price a year after the scandal.

What each headline actually revealed

Knowing how lawsuits affect stock prices on average only helps if you can recognise the next bad case early, so here is what the market learned on each day 0, sorted from the worst one-year outcome to the best.

What to look for in the next headline

Put side by side, the bad cases share a few features that the mild ones lack. None of them is a rule on its own, but each one showed up again and again at the bottom of the table:

• The liability is counted in units that can still multiply. Cars, customer accounts, user records, victims, plaintiffs. On day 0 of Volkswagen the number was 482,000 cars; days later it was 11 million. If the headline gives a count that is still being revised, the cost is not known yet.
• People were hurt or killed. PG&E, Boeing and BP are three of the four worst outcomes. Physical harm brings wrongful-death claims, political pressure and regulators with the power to stop operations.
• A regulator restricts the business, not just the bank account. A grounding, an asset cap, a product withdrawn. That hits future earnings, which is worth far more than any one fine.
• Evidence of concealment. Volkswagen’s defeat device, J&J’s internal test results, Facebook knowing since 2015, Equifax waiting weeks. “They knew” invites punitive damages and many more lawsuits.
• The first verdict in a mass lawsuit. Bayer’s $289 million was small; the more than 5,000 cases behind it were the real news.
• The problem is the product. Diesel cars that cheat tests, a plane that crashes, a credit bureau that leaks data. Customers can walk away from those.

And the mild cases had the opposite traits: one fixed amount the balance sheet could absorb (Apple), an opening negotiating number with precedents pointing lower (Deutsche Bank’s $14 billion became $7.2 billion), or a threat aimed at one person rather than the business that was settled in two days (Tesla was up 7.6% a month later; its weak year came from production problems, not the SEC).

The checklist is not a formula. Equifax and J&J both showed concealment and both fell 10% or more on day 0, yet a year later they were only a few percent below their pre-news price. The flags tell you the first drop may not be the bottom, not how far it will go.

What I take from it as a long-term owner

I usually argue for ignoring the daily noise. In my time in the market test, sitting in the S&P 500 since 1950 turned $10,000 into $4.59 million, against $2.89 million for a 200-day timing rule. A major legal scandal turned out to be one of the few situations where “it’s just a headline” was the wrong default, at least in this sample.

• Do not assume day one priced it in. When the liability is open-ended, the first drop was the smaller part of the damage in most of these cases.
• Ask whether the cost has a ceiling. A fixed fine that the balance sheet can absorb is a different animal from thousands of product lawsuits or a grounded fleet.
• Ask whether the product or licence is hit. That is the difference between a cost and a smaller business.
• Do not buy the settlement on reflex. The day itself was mildly positive, but a year later the median stock was up only 3.3% and Bayer was down 25%.
• Position size is the real protection. PG&E fell 87% in a year. In a diversified portfolio that is a bad year for one line. In a concentrated one it is the whole year. The ATH staircase rule handles this mechanically: it holds a name for two months at a time, and a stock that stops setting new highs does not qualify again.

This also fits what I found when I tested technical analysis for stocks: over years, what moves a single stock is the business, and a scandal that damages the business is fundamental news, not a chart pattern to fade.

Limits of this test

• Small sample. 13 first headlines and 10 settlements. Medians are more trustworthy than means here, because one extreme case (PG&E’s bankruptcy) moves any average.
• The market moves too. These are plain stock returns, so they include whatever the whole market did. Most of these years were good ones: the S&P 500 or the DAX rose a median 10.2% in the year after the first headlines, so the losses above are not the market’s fault. Boeing’s one-year window ends on 10 March 2020, in the first weeks of the COVID crash, which exaggerates its fall.
• Sector news overlaps. An oil or bank selloff in the same months shows up here as company damage.
• Day-0 rule. Where a story leaked in pieces (Facebook’s FTC fine, Volkswagen’s settlement talks since April 2016), part of the move happened before the date I used.

The settlement dates are checked against primary sources, including the EPA’s Volkswagen release, the FTC’s Equifax release and Boeing’s 8-K on the DoJ agreement.

Conclusion: how lawsuits affect stock prices in practice

A major lawsuit or scandal was not a free dip in this sample. The first headline cost the median stock 8.5% on the day and 20.6% over a year, with 10 of 13 below their pre-news price, while the final settlement was mostly already in the price. The cases that recovered quickly had a bounded cost and an untouched business. The ones that never recovered had neither.

So my rule is not “ignore the lawsuit” and not “sell on the lawsuit”. It is two questions: is there a ceiling on this, and does it hit what they sell? If both answers are comfortable, the drop may well be noise. If either one is “I don’t know”, history says the market does not know either, and it tends to find out the hard way.

FAQ: how lawsuits affect stock prices