Why do private equity deals fail?

Private equity deals fail when earnings fall and the deal's structure cannot take the fall. In the failed buyouts we have reviewed, earnings fell for one of four reasons. In most cases the cause was visible in the documents before the price was paid.

The business was in long-term decline. It was bought at the top of a cycle. Someone else set its prices. Or the plan needed something the business had never done.

Four features of the deal then made the fall fatal. Those features appear in winning deals too. The four causes do not.

The record is stable enough to be uncomfortable. Roughly one buyout in ten goes bankrupt. About one in four loses money. More than half fall short of the plan they were bought on. A minority of deals carries the average return of the asset class.

The companion piece covers why capable people keep making the same mistake. This page covers what the mistake looks like.

What a failed deal is

The equity in a buyout is what is left after the debt. It is the earnings, times the multiple a buyer will pay at exit, less what is owed.

So a deal fails in only a few ways. Earnings fall. The multiple falls. Or the claims ahead of the equity are too large, or too rigid, to survive either.

A deal that returns the money but misses the plan fails more quietly, and more often. The plan set the price. If the plan was wrong, the price was wrong.

We reviewed three dozen failed buyouts, some from our own work and most from the public record. Nearly every one had a cause plus an amplifier. The cause made earnings fall. The amplifier stopped the business living through it.

The four causes

1. A business in decline, bought as if it were a dip

No cause came up more often in the deals we reviewed. The channel, the technology or the customer's habit was already moving. The plan assumed it would stop.

Toys R Us was bought in 2005, when shoppers were already moving to discount chains and online. It spent the next twelve years paying interest while its shops went unfixed. It filed for bankruptcy in 2017 and closed its American stores the next year.

Tribune, the newspaper group, was taken private at the end of 2007. Within a year it was bankrupt. EMI was bought the same year, as recorded music moved to digital. Its lender took the company in 2011.

In Australia, the owner of the Nine television network was bought in 2006 and 2007, as advertising began to move online. Its lenders took control in 2012. REDgroup, the owner of Borders and Angus & Robertson, went into administration in 2011. Book buyers had moved online.

Dick Smith shows how visible the warning can be. When Woolworths sold the chain in 2012, it called the business non-core. Consumer electronics were moving online and prices were falling. The warning was in the seller's own announcement.

2. Peak earnings, bought at a peak price

The business is cyclical, or lives off a commodity price. Earnings are near their high. So is the multiple. The low case in the model still shows growth.

TXU, the Texas power company, was bought in 2007 in the largest buyout on record. The case depended on natural gas prices staying high. They fell, and the company filed for bankruptcy in 2014. Samson Resources, an oil and gas producer bought in 2011, went the same way four years later. Energy prices had dropped.

Harrah's, the casino group later renamed Caesars, was bought in January 2008, at the top of the gaming cycle. Its main operating company filed for bankruptcy in 2015. Chrysler was bought in 2007, as American car sales were about to collapse. It went bankrupt in under two years.

In each case the base year was close to the best year the business had ever had.

3. Someone else sets the price

Revenue or margin depends on one payer, one regulator, one government or one customer. After the deal, they change the terms.

Envision Healthcare, an American doctor-staffing company, was bought in 2018. It was known for billing patients outside their insurer's network. A federal law then banned surprise bills. Its largest insurer cut what it paid. Envision filed for bankruptcy in 2023.

Southern Cross Healthcare ran care homes across Britain, paid for mostly by local councils. When the councils squeezed their fees, it could not pay its rent. It collapsed in 2011.

Healthscope, the Australian hospital group, reported revenue growth "ahead of market" before its 2019 buyout. Its own results slide showed fewer people holding the private insurance it depends on. The growth had come from new beds, with no gain in share. The buyer modelled an ageing population. The business was selling to insured patients, and that group was shrinking.

In every case we reviewed, this risk was written down in the documents. The buyer judged it unlikely, and never priced it.

4. The plan needs something the business has never done

Growth, a turnaround or a building programme, at several times the pace of anything in the record. The levers may be real. Nobody has checked whether these people, with these systems, can pull them.

Chrysler is the public example. Its 2007 buyer had never run a car maker. It appointed a chief executive from a home-improvement retailer. The plan was a turnaround that the company's previous owner, a car maker, had failed to deliver in nine years.

This cause is the hardest to see from outside. It shows up rarely in the press and often in our own files. The test is simple. Has the business done this before, and is the result in its revenue today?

Whether the plan follows from the record is a practice in the value-creation area of a deal check. The five stages read:

• Red flag: The plan's growth has no relationship to what the target has delivered, and nobody can say what changed.
• Doubtful: The plan outruns the run-rate, and the only explanation offered is the levers.
• Unproven: The plan is bridged to history at the top line only; the drivers underneath have not been reconciled.
• Supported: The plan is bridged driver by driver to the last three years, and every step above the run-rate has a named cause.
• Proven: The plan has been re-based to the drivers we believe, the base year tested for one-offs, and the price rests on that case, not management's.

The four amplifiers

A cause makes earnings fall. An amplifier decides whether the business survives it.

The floor was sold. The buyer sells the property to landlords and rents it back to help pay for the deal. A cost that could flex becomes one that cannot.

Healthscope is the clearest case. An independent expert had named the group's advantage as its sites, "extremely difficult to replicate". The 2019 buyer sold twenty-two of them as part of the deal, and rented them back.

When receivers were appointed in 2025, the hospitals kept running. The advantage survived. The company that sold it did not.

Mervyn's took the same step. The American department store was bought in 2004 and was bankrupt in 2008. Its property was moved into a separate company, and its rents went up.

Red Lobster sold its restaurant sites to fund its 2014 buyout. It filed for bankruptcy ten years later. Its chief executive told the court that many of its leases were priced above the market. Southern Cross failed for the same reason: rents that rose every year, against fees that did not.

The cash is already out. The owners pay themselves dividends from new borrowing in the good years. When the bad year comes there is no cushion.

Simmons Bedding paid its owners a large dividend from new debt. It went bankrupt in 2009 when the housing market turned. Debenhams paid out more than its owners had put in before it returned to the stock market in 2006.

The price was set by a contest. A long auction or a bid battle, won at a premium.

Healthscope had cut its guidance three times in three years. A contest between two bidders ran for a year. The winning bid was fourteen and a half times earnings, a forty per cent premium.

The announcement said the business had "returned to growth". Guidance was cut again before the deal completed. EMI was also bought at auction, at the peak of the 2007 market.

The debt was sized to the plan, not the record. This one shows up in almost every failure. So on its own it explains little.

Why capable people miss it

In most of these deals nobody was deceived. The buyers were capable and the diligence was thorough. The cause was written down. What went wrong was a decision to treat a warning as a discount.

The warning is priced as a discount. A business that has been shrinking costs less than one that has been growing. Cheap feels like room for error.

That room vanishes if the price is set on a plan that reverses the shrinking. The discount was for the decline. The plan spends it on the recovery.

The risk section is read as boilerplate. A company says in its own documents that a contract, a landlord or a regulator could end it. Readers treat that as lawyers' wording. In the deals we reviewed, it was a description of what then happened.

The costs the deal creates are left out of the downside. A low case that still shows growth is a low case nobody has run. Put the rent and the debt in, against the trend the company has already described, and see whether the equity survives. A committee can ask for no cheaper piece of arithmetic.

Why the difference matters

Debt kills only the deals whose earnings fall.

Hilton was bought in 2007 at a full price, with heavy debt, months before a financial crisis. Commentators often call it the most profitable buyout on record. It survived because none of the four causes was present. Travel kept growing, and the business had already done what the plan asked of it. We cover it, and two others, in a separate piece.

So the order of the questions matters. Look for the cause first. If one is present, the amplifiers decide how bad it gets. If none is present, the amplifiers are a matter of price.

A different kind of failure

Sometimes the fund makes money and the next owner loses it.

Dick Smith was bought cheaply in 2012. It went public thirteen months later on a forecast. The fund was fully out within two years, at many times its money. By January 2016 the retailer was in receivership.

Myer's private equity owners floated it in 2009, and sold their entire holding in the float. The shares have never traded above the price the public paid. Debenhams returned to the stock market in 2006 with many of its shops already sold and leased back. It went into administration in 2019.

For the fund, those deals worked. For anyone buying a business from a fund, they are the reason to run the same four questions on the way in.

What catches them

None of this needs a bigger data room. It needs the questions asked in the right order.

Test the segment the business sells into. Ask where today's earnings sit in the cycle. Find whoever else can set the price, and read what the documents already say about them. Bridge the plan to the record, driver by driver, before anyone argues about levers. Then run the downside on the structure as it will be after the deal, rent included.

A deal check covers the first four. Five areas, twenty-odd practices, a sentence per stage, ten minutes. It stops short of saying whether to do the deal, and it leaves the price alone.

It gives one of four readings of the case. Keep going. Slow down and finish the homework. Slow down, there are real doubts. Or stop.

The amplifiers belong to the pricing and structuring work that follows.

We tested the assessment on 105 public buyouts, in an exploratory, AI-assisted retrospective test. It raised a warning on 43 of the 51 that later failed, and on 14 of the 54 that succeeded. The limits are on the method page: some answers rest on what the AI agents already knew, and the warning was tuned on these deals.

Frequently asked questions

Is the failure usually the business or the price? Both, in that order. In the deals we reviewed the business had a cause, and the price and structure left no room for it. A full price on a business with no cause has often worked. A full price on a business in decline almost never has.

Do most failures show at entry? In our experience, and in the public-record deals we have scored, yes. Three of the four causes were visible in published documents at the time. Failed deals score well below successful ones at entry on the same documents.

What about deals that fail from bad luck? A pandemic or a recession inside the hold is bad luck. A structure that cannot survive one is a decision.

Does this apply to a strategic acquirer? The four causes are the same. A strategic buyer adds another question. Are the synergies real, and who is being paid for them?

How do I know if my deal is heading this way? Score it honestly before the bid. The check takes ten minutes and says whether to keep going, finish the homework, face real doubts, or stop. It is private, and it costs less than finding out from the lenders.

• The deal checks: buying as a fund, buying as a strategic acquirer, and selling
• How a Stages of Excellence check works
• Half of buyouts miss the plan they were bought on