Ninety Days of Cash

In the summer of 1997 Apple had, by most accounts, less than ninety days of cash left. It was losing something like a billion dollars a year, its stock was in the gutter, and the story people tell now, the return of Steve Jobs, the iMac, the iPod, the most valuable company on earth, was not yet even a rumour. On the 6th of August, Microsoft put 150 million dollars into the thing, took non-voting stock, and promised to keep making Office for the Mac. Bill Gates, of all people, helped keep Apple’s lights on.

Hold that image, because it contains the whole lesson and almost everyone draws the wrong one from it. The genius came later. What mattered first was that Apple didn’t run out of money before the genius could show up. A turnaround is a bet on survival first and recovery second, and if you get the order wrong it doesn’t matter how right you are about the recovery. The graveyard is full of companies with brilliant restructuring plans that ran out of cash in month four.

This is the category Peter Lynch called turnarounds, and it’s the one where his sunny temperament served him least well and his discipline served him most. He sorted the wrecks into types: the ones riding a government bailout, the “who would have thought” surprises, the honest company caught in a single overblown tragedy, the good business trapped inside a bad one and waiting to be spun out, and the restructuring that sheds the junk a company bought in better days. Different clocks, different catalysts. But the question he actually asked was never “how cheap is it?” It was “how, specifically, do the earnings get better?” Lower costs, higher prices, more volume, new markets, or cutting the bleeding operation loose. If you can’t name the mechanism, you don’t have a turnaround, you have a hope.

And he was unusually hard-nosed about one thing: the shape of the debt. Long-term bonds are survivable, because bondholders have to sit down and negotiate. Short-term bank debt that the lender can call is what actually kills companies, because a bank can accelerate the loan and tip a liquidity squeeze into a bankruptcy overnight. Same total leverage, completely different odds of survival. He also loved a troubled utility, on the theory that regulators and the public won’t let the lights go out, and he mapped their recovery in four acts you could follow on the price-to-book: disaster hits and the stock falls 40 to 80 percent, then crisis management and a dividend cut, then stabilization, then recovery and the dividend comes back. Buy on the dividend cut, he said, and wait for the good news, rather than straining to catch the exact bottom.

Here is where the professionals who do nothing but this for a living push harder than Lynch did, and mostly in one direction: down into the capital structure. The distressed-investing school, the world of Marty Whitman’s Safe and Cheap and funds like Oaktree, doesn’t start with “will it survive?” It starts with “if it doesn’t survive in its current form, which piece of paper controls what’s left?” There’s a fulcrum security, usually a layer of debt, that ends up owning the reorganized company when the dust settles. And under the absolute-priority rule that governs a US bankruptcy, the common stock, the thing a retail investor is most tempted to buy because it’s trading at two dollars and feels like a lottery ticket, is typically wiped to zero before any of that recovery reaches it. To these investors the equity isn’t the turnaround vehicle at all. The debt is. The stock is just the most visible and most dangerous way to play.

Then there’s the flat warning from the other side, and it deserves to be stated without flinching. Most turnarounds don’t turn. The base rate of failure for deeply distressed companies runs well over half, and cheap-with-deteriorating-fundamentals is the textbook falling knife: it keeps falling. The reason famous comebacks feel common is that the failures delist and vanish from the data, taking their cautionary tale with them. Apple and post-2009 General Motors are burned into memory; the dozens of near-identical bets that quietly went to zero are not, because nobody writes the retrospective on a company that no longer exists.

I’m not going to crown a winner among these. Lynch’s typology is a genuinely useful diagnostic, the distressed pros are right that the capital structure decides who gets paid, and the sceptics are right that the base rate is brutal. They’re describing the same animal from three distances.

What has clearly changed since Lynch’s day is the plumbing, and not in the small investor’s favour. Modern bankruptcy is faster and more professional: debtor-in-possession financing, prepackaged filings negotiated before the company even goes into court, and a deep private-credit industry that specializes in “loan-to-own,” where the sophisticated lenders take control the moment things wobble and public shareholders are diluted or erased before they can react. The window between “the numbers are visibly improving” and “the stock has already re-rated” has narrowed to weeks. And Lynch’s beloved safe utility is not as safe: deregulation and the energy transition changed the math, and in 2019 Pacific Gas and Electric filed for bankruptcy under the weight of wildfire liability, which is not a fate the old regulatory-safety-net thesis had a slot for.

So if you want something to actually hold onto, put it all on the survival side of the ledger first. Three gauges, none of them exotic. Interest coverage: can the company pay its interest out of what it earns, and is that number climbing toward comfortable rather than sinking under one? The maturity wall: not the total debt but when it comes due, because fifty million dollars due in ninety days is far more lethal than five hundred million due in 2028. And the cash runway: months of cash at the current burn rate, with anything under a year and no committed financing being a live, binary event, not an investment. Only once those clear do the recovery signals earn your attention: gross margin turning up for a few quarters (harder to fake than net income), debt falling steadily, inventory no longer outrunning sales. Lynch left one sharp tripwire here worth keeping: debt that has dropped for five straight quarters and then starts rising again is usually the sound of the fix coming undone. When you see it, leave.

The math is what forces the discipline. The payoff is convex, a possible total loss against a possible multiple of your money, and convex payoffs punish you for staring at the upside. If the chance of a zero is even one in three, a five-bagger dream can still be a losing bet. So invert the question the way the distressed desks do. Not “how much could I make if this works,” but “what is this worth if it files, and can it stay alive long enough for the repair to matter.” The dream takes care of advertising itself. The downside needs you to go looking for it.

Apple got its ninety days extended by a cheque from its biggest rival, for reasons that had as much to do with an antitrust case as with charity. Most companies in that spot never get the reprieve. They’re the ones missing from every inspiring story, which is exactly why the inspiring stories are such poor guides.