Lights Out: Pride, Delusion, and the Fall of General Electric

by Thomas Gryta

More than forty thousand men and women had worked there at GE’s peak. It was a tenth of that size by 2017.

Flannery had taken to uttering a new mantra around the company’s shiny new offices in Boston: “No more success theater.”

Flannery was comfortable with numbers, especially financial statements, and that was where he began his search in Schenectady. It didn’t take long to see the problem: as Flannery paged through the financials, he realized that GE Power had somehow run out of cash. This discovery wasn’t just shocking—it was unthinkable.

“It was like they drove off a cliff,” Flannery later told an observer, “and there were no skid marks.”

Combing through the books of the sprawling conglomerate in preparation for taking on the job of his life, Flannery looked into the biggest and most important industrial business of them all—the unit that was the reason for GE’s existence—and found a deep, empty hole where there should have been cash.

It was so large that it was given the same brand of financial credit and trust as the US government itself.

One manager tried to subvert the system by putting an employee who’d recently died in the bottom 10 percent of the ranking list in order to save another employee’s job.

Workers came around to liking Welch for good reason: the stock kept rising. From 1980 to 2000, GE’s earnings rose to $12.7 billion from $1.5 billion, while revenue more than quintupled, rising to $129.9 billion. The stock price rose more than forty times its value in the period.

The oversight role of the board was minimal. After all, Welch’s meteoric success provided very little for the board to complain about.

GE’s annual report mentioned the word “internet” just eight times in 1998. One year later, the word appeared fifty-six times. But then the buzzword would disappear just as quickly as the ’90s internet stock market bubble did. The word “internet” was back down to ten appearances in the 2001 annual report, having rippled through the company’s description of itself to the share-owning public and then disappeared again.

“My second day as chairman, a plane I lease, flying with engines I built, crashed into a building that I insure, and it was covered with a network I own,” he said just weeks afterward.

One of those tricks was the Edison Conduit, a massive special-purpose entity that was technically independent of GE and not on its balance sheet. The problem was that the Edison Conduit was controlled by GE and guaranteed by GE Capital, meaning that GE carried all the risk, while investors didn’t even know it existed.

One strong year and a few weaker years doesn’t look great to shareholders—it is much better to spread the profits around more consistently to give the impression of a well-oiled machine. That impression created by GE was a true one, but not in the way that people thought. This is the essence of earnings management.

The company had enjoyed broad trust from investors for years, winning plaudits for its investor relations and special notice from one trade publication in the 1990s for being “shareholder-oriented.” Through that period, GE didn’t even hold conference calls to discuss quarterly results, as competitors did. Its earnings releases were typically short and light on detail. If anything, GE tended to leave investors with questions about how the company had hit its numbers. But the performance of the stock left little room for complaints.

company. It was only with the implosion of Enron—its suddenness and the shock of the fakery revealed—that many investors began to think more skeptically about all those smooth quarters and to wonder whether the roaring success of the Welch years could really be sustained.

“GE has been shrouded in mystery for a number of years,” Gross told CNBC. “Institutional investors have wondered why a company can continue to produce 15 percent earnings growth year after year, quarter after quarter.” The answer, in Gross’s telling, was that the success might not be real.

Capital’s massive size also enabled it to jump into new markets in a significant way, making major plays with the hundreds of millions it had to put to use every day to make a profit on the back of GE’s sterling credit. Subprime mortgages were just such a market in 2006.

Channeling the cold-bloodedness for which Welch had been renowned, Immelt wanted no part of a business that relied on outmoded expectations. The company quietly began looking around for buyers for the massive Appliance Park business in Louisville, the factory complex in which GE had supplied, and helped to create, the post–World War II American middle-class home.

Its plastic pellets were turned into products all over the world. It made the plastic cases for Apple’s famous iMac G3, the multicolored desktop computer that had heralded the triumphant return of Steve Jobs as CEO.

FEW FATES WERE worse than missing your numbers at General Electric.

But in the end, Capital was always a problem. It was utterly complex and filled with risk, and its tentacles reached everywhere in the company. In addition, the broader market had lost its former patience for letting big industrial companies mess around in the risky business of finance.

It was reasonable to think that Immelt didn’t know much about how the greasy gears of Capital worked. In reality, it isn’t clear that many people anywhere understood those complexities.

The first quarter of 2008 showed once and for all that it had long been GE Capital that was the force keeping the earnings streak on track—and that GE’s earnings were unpredictable when Capital itself was struggling.

In fact, precisely what PIMCO’s Bill Gross had warned about in 2002 was now coming to a head. He had said that GE’s dependence on commercial paper was dangerous because the company wasn’t fully backing up its needs with a contingency plan, as many financial services funds did, or were required to do.

you can’t unring a bell. If major sellers of commercial paper could go bankrupt overnight, that changed every investor’s assessment of the risks involved.

The meatball logo retained much of its power, still calling out from the aisles of hardware stores and appliance retailers, but thanks to quiet licensing deals made over the years, the logo could be found on gadgets that GE hadn’t actually made for decades, or had never made. (Among these products were a GE-branded computer mouse and a faux security sign meant to fool burglars into believing that a house was protected by a fictional GE alarm system.)

But were they a “product of pure ecomagination”? The Evolution locomotive was, if anything, a product of the EPA, whose tightened emissions standards had compelled railroads to find locomotives less damaging to the environment—and thereby created for GE the incentive to invest in designing and producing them.

It was like a Coke commercial with jet fuel and ecological wish-casting at its heart. In another, which the company called “Model Miners,” semidressed men and women strut about in headlamps in a coal mine, sensually brandishing jackhammers and picks, to the strains of “Sixteen Tons,” the American folk classic about miners sinking into debt.

GE was now fresh and exciting. The company would eventually convince multiple news outlets to declare that GE was no less than a “124-year-old startup.”

Immelt viewed the job in part as a reward for Bornstein’s critical work staving off the financial panic. “You can tell a lot about somebody when they’re in the shitstorm,” Immelt said.

All GE executives should be focused on one central, critical job: selling.

“additive manufacturing makes a shitload more sense than Six Sigma did.”

But ultimately, though Immelt’s flacks claimed that he never considered it, it all came down to the price of a General Electric share.

By March of 2016, after GE had spent years pressing the case, Bloomberg Businessweek obliged with the headline of Jeff Immelt’s dreams: “How GE Exorcised the Ghost of Jack Welch to Become a 124-Year-Old Startup.”

A GE compliance official hired to help deal with the incoming attention from the Fed supervisors asked what Capital’s standard protocol for extending a new line of credit required. What chain of command signed off on the deal? What committee reviewed it? These were questions that any bank would have answered easily. But at Capital, the new official eventually learned, there was no concrete protocol.

The former executive listening to Neal’s version of events in Crotonville that day squirmed in his seat. My God, he thought to himself. You have learned nothing.

Big changes were coming for Capital. The market would not yield to Immelt’s desire for a revaluation. As long as Wall Street believed that America’s great industrial company was in fact a massive bank with a couple of turbine businesses bolted on the sides, nothing about Jeff Immelt’s reputation or the company’s stock price would change meaningfully.

Some found the timing curious. Was GE selling at the right time to get the most value for these businesses it had built over decades? GE didn’t care—it wanted out.

GE’s move was extremely self-assured. It projected the image of a company that had aggressively used its financial wing for a generation to manage its earnings and now was telling the world, backed by its Wall Street reputation, that it could shed that piece of itself and not just fail to suffer but actually perform better than it had in a decade.

It was in servicing industrial machinery—not in selling it—that all the profit lay.

The innovations were dutifully pitched for small features in the business press by the division’s public relations staff. These pitches were, one employee said, “very friendly and bubbly and full of shit.”

Now its sell-side market analysts were getting suspicious. Out in the power markets, the company was aggressively bidding for any new business it could win, often offering prices to build new coal- and gas-fired power plants so low that it would prove impossible to turn a profit once all the work was complete. But those long-term profits were of little concern to Alstom’s sales teams—they needed to win those deals for the down payments alone: the deposit paid by a customer on signing such a contract was virtually the only cash coming in the door.

In that moment, Adam Smith was listening to the logic of the deal collapse. Alstom had been worth examining because it was financially imperiled; the value of pursuing it in the first place had been to capture its assets at a bargain, strip what GE wanted, and then sell the rest, like an old car, for the parts. Now the company was talking about making sure Bouygues saw a return on its investment in a company that, absent a buyer, seemed to be headed for bankruptcy or a possible government bailout. Going into the deal with that kind of logic, Smith thought, GE would be seriously overpaying on almost any price it offered.

But in fact, some did attempt to raise objections, though they did so advisedly, cautiously, and briefly. A member of the Power business team that worked on the deal said that by then it had become clear that when any acquisition was a top priority of corporate, or of Immelt’s, any dissent was unlikely to alter GE’s course.

Brent crude had closed out the previous month at more than $105 a barrel, only a little off its summer peak. But even as GE managers were speaking, a long slide had begun. By the end of October, oil had fallen below $100 a barrel. By the end of the year, it was under $80. A year later, a barrel of Brent crude cost less than $50—less than half the price at which GE had just assured its investors that all its plans for the oil business made financial sense.

As with all of GE’s industrial business lines, there was supposed to be security in the vast backlog of orders taken over the years by the oil and gas unit. But all those orders for pumps and compressors and drill bits weren’t going to do GE much good if the customers, pinched by the oil price collapse, couldn’t actually pay for the machinery they’d bought.

While the marketing and sales folks were getting ready to sign software contracts, there were still lingering and ominous internal questions about whether GE’s goals were realistic, or even possible.

By their count, the deal concessions GE was being asked to make had now grown so large that GE’s original logic for the transaction no longer made sense. The cost to close the deal seemed to exceed the overall benefit to the company. And that, they realized, would trigger a breakup provision in the sale agreement, allowing GE to back out of what they increasingly viewed as an albatross of a deal.

For the Power business, it was pedal to the metal. It was launching its next-generation natural gas turbine—in essence, a cousin of GE’s jet engines, but vastly larger, and capable of turning an equally large generator to produce electricity. The newest machine was air-cooled and operated at temperatures higher than the melting point of the materials of which it was made. No one questioned that it was a marvel of industrial engineering.

Immelt’s locker-room chumminess and absolute power at the company made such comments sting. And even if it was never delivered explicitly, the lesson sunk in, down through the levels of the company. There was no market for hard truths or bad news. Not as far as the guy at the top was concerned.