What this covers

Josh walks Matt through GE's history and what is left of it. The first stretch is forensic. GE Capital used the parent's AAA rating to borrow cheaply through wholesale funding instead of deposits, so GE behaved like a bank without bank regulation. Because industrial sales are booked long before cash arrives, finance looked more profitable than manufacturing. DuPont Analysis shows why return on equity fell once post-crisis rules forced more capital. The 2008 funding run left GE dependent on the government and Berkshire Hathaway. Josh also argues, contentiously, that Jack Welch used the finance arm to paper over weak industrial quarters, with GE Capital near 60% of earnings at its peak.

The second stretch is a ledger of Jeffrey Immelt-era decisions. It covers the 2004 WMC subprime purchase and its $1.5 billion civil penalty, more than $14 billion in oil-services acquisitions, and the Predix industrial software platform. It then turns to Alstom, bought for $10 billion in cash and later hit with a near $22 billion impairment under John Flannery. Josh calls Alstom the defining mistake and says the gas-power downturn was foreseeable, since Siemens had been cutting capacity. He also criticizes tens of billions in share repurchase spending near the stock's peak.

The last stretch looks at the surviving businesses. Aviation sells engines at steep discounts, often more than 50% off list, to earn multi-decade parts and service revenue, and Josh expects its operating leverage to surprise people. Renewables gets the same hardware-plus-service lens but with far less service income and a geographic moat that is eroding. Healthcare converts roughly all earnings to free cash flow. He closes on Larry Culp's decentralization of P&L responsibility.

Sharpest takeaway

Josh Aguilar argues GE's decline was largely self-inflicted: Welch and Immelt used GE Capital leverage and centralized, hubris-driven capital allocation to mask weak industrial economics, and the remaining focused businesses can only re-rate if they are separated and run with decentralized, disciplined capital allocation.

  • GE Capital's leverage and wholesale funding made GE bank-like without bank regulation, and the 2008 crisis and post-crisis capital rules exposed it
  • Immelt's acquisitions (NBC sale, oil investments, WMC, Alstom) and buybacks destroyed shareholder capital, and the Alstom impairment shows the cost
  • Breaking up GE only creates value if each focused business is allocated capital better than the market, and decentralization under Culp is the path

The argument · threads12 threads · 76 claims
0.71

Ill-timed buybacks, acquisitions, and big bets wasted shareholder capital and overreached.

12 pointscentrality 4/5
  • GE spent tens of billions of dollars on share repurchases in Immelt's final three years, at prices nearly double the level the stock later fell to, and that capital could have been used to deleverage and avoid selling exceptional businesses like GE BioPharma.

    “GE spent tens of billions of dollars on share repurchases in Immelt's final three years and nearly doubled the levels the stock price ultimately fell to.”

  • Immelt spent more than $14 billion over ten years acquiring companies that helped drillers pump and transport oil, and those investments were ill-timed and later largely exited through the sale of Baker Hughes interests.

    “Immelt also spent more than $14 billion over 10 years on ill-timed oil investments when he acquired companies that helped drillers pump and transport oil.”

  • Immelt had an uncanny ability to buy high and sell low, or otherwise waste shareholder capital, despite having to deal with 9/11 and the Great Recession.

    “Immelt almost had an uncanny ability to buy high and sell low, or just plainly waste precious shareholder capital.”

  • Immelt's push into turbines, including Alstom, was made as GE heralded the golden age of gas, while Siemens had been right-sizing capacity, so the later decline of the gas power market was not completely unforeseeable.

    “The declines of the gas power market weren't completely unforeseeable. Siemens had actually been right-sizing capacity while GE went in the opposite direction.”

  • Immelt spent mid single-digit billions on the industrial IoT platform Predix, which failed because it tried to be all things to every industrial, and parts were later sold to Silver Lake.

    “Immelt also spent mid single digit billions in its failed industrial ILT platform”

  • GE sold NBC in 2011 for an enterprise value under $40 billion, which compares unfavorably with AT&T's roughly $110 billion purchase of Time Warner, though the deals are not apples to apples.

    “GE sold NBC back in 2011 for an enterprise value of under $40 billion, which while not apples to apples compared very unfavorably versus other cable deals like when AT&T bought Time Warner for roughly $110 billion.”

  • GE bought Alstom for $10 billion in cash, which Josh calls the defining mistake of Immelt's tenure, and then took a near $22 billion impairment on the deal during John Flannery's tenure.

    “GE bought Alstom for $10 billion out of cash, but took a near $22 billion impairment on the deal during the end of John Flannery's tenure.”

  • GE shareholders earned about a 30% total annual return from 1996 to 2000 and about 23% annualized over more than two decades, while 2000 sales rose 16% to nearly $130 billion.

    “GE shareholders benefited from a 30% total annual return on their investment. And in over two decades, they would've enjoyed a 23% total annualized return.”

  • GE Capital's short-term borrowings were roughly a third of its assets, a large share that made it vulnerable to funding shocks.

    “GE's Capital short-term borrowings were an appreciable chunk of its assets, about a third or so from my recollection.”

  • Acquiring Alstom to chase the golden age of gas was GE's most consequential mistake, because the decline in gas power was foreseeable and Siemens had right-sized capacity while GE expanded.

    “The declines of the gas power market weren't completely unforeseeable. Siemens had actually been right-sizing capacity while GE went in the opposite direction.”

  • GE's failed merger attempt with Honeywell reflected a desire to get bigger, and Honeywell's market value today is nearly 16% higher than GE's.

    “GE didn't know it at the time, but it wanted to get bigger with what ultimately was its failed merger attempt with Honeywell whose market value today is nearly 16% than GE is today.”

  • GE's Predix industrial IoT platform failed because it tried to be all things to every industrial, while later IoT offerings succeed by focusing on the domain expertise of the businesses they serve.

    “it wanted to be all things to every industrial, whereas a lot more of these IoT offerings nowadays”

0.65

GE's decline was largely self-inflicted rather than caused by outside shocks.

4 pointscentrality 5/5
  • The NBC sale and other divestitures were not made from a position of strength; Immelt was trying to refocus GE on industrial roots and make it a software-industrial company, a vision that was early and bigger than reality.

    “His big initiative was really, can we make this a software industrial type company? Which is actually all the rage nowadays, but he was just early and just his vision was bigger than the reality”

  • GE once paid dividends of over $5.4 billion a year, while in 2021 it paid a token dividend of $535 million, and GE did not clear $2 billion in free cash flow last year.

    “the company that used to pay out dividends of over 5.4 billion a year. Today GE paid a token dividend in 2021 of 535 million.”

  • GE's asset sales were not made from a position of strength; the sales were driven by problems under Immelt that were not fully appreciated until the end of his tenure, under Flannery and Culp.

    “Was it from a position of strength? Clearly not.”

  • Although GE faced outside shocks such as 9/11, the Great Recession, and the pandemic, there was plenty of self-inflicted pain, meaning much of its decline was controllable.

    “There's been 9/11. There's been the Great Recession. There's been the global pandemic. But there's plenty of instances of self-inflicted pain.”

0.64

Insulation from bad news, centralized management, hubris, and suppressed dissent drove bad strategies.

9 pointscentrality 4/5
  • GE's problems can be pinned on two areas, culture and capital allocation, which worked in tandem; Welch's issues were short-termism at the cost of rational long-term decisions and a failure to mitigate downside risk, and both were carried through GE Capital.

    “One is culture, and two is capital allocation. To a certain extent, they worked in tandem with one another. Welch's issues were really two-fold, short-termism at the cost of rational long-term decision making.”

  • Immelt had an aversion to hearing the bear case on anything he wanted to pursue, which fostered overly rosy projections, underappreciation of the power segment's problems, and aggressive revenue recognition, and that insulation from reality drove the company's financial problems.

    “Immelt really, depending on who you believe, had an aversion to hear the bear case of anything he really wanted to pursue.”

  • Jeff Immelt was CEO of GE for about 16 years, materially longer than the median tenure of industrial CEOs today, and Josh believes Immelt gets too much of a pass from other market participants.

    “Jeff Immelt was CEO of General Electric for about 16 years, which is materially longer than the median tenure of industrial CEOs nowadays.”

  • GE was the poster child of centralized management and that is why it does not work in Josh's view, the opposite of Berkshire's decentralized trust-based model.

    “I would actually say that GE was like the poster child of centralized management and really why it doesn't work in my view.”

  • The value proposition of a conglomerate is tax-free movement of capital, but that only matters if management can allocate capital better than the market, and GE's long-run stock performance shows it has not done so.

    “The whole value proposition of the conglomerate is that you can have tax free movement of capital, but that only matters really if you can allocate capital better than the market.”

  • The main lesson from GE is how far hubris can take an organization, and Buffett's institutional imperative explains how the arrogance of not seeing downside risk leads companies to pursue bad strategies.

    “The lessons learned, I'd really say kind of amount to how far hubris can take you. And Buffett talks about this with the institutional imperative.”

  • A company cannot consistently beat earnings expectations every quarter, as GE aimed to under Welch, without serious long-term adverse consequences.

    “there's really no way a company can consistently beat earnings expectations like GE aimed to do every quarter under Welch without some real long-term adverse consequences.”

  • The key lesson from GE is how far hubris can take a company, and the defense is an intellectual environment where alternative views and downside risks can be surfaced before decisions are made.

    “The lessons learned, I'd really say kind of amount to how far hubris can take you.”

  • GE was historically known for homegrowing managers, but that model does not work at large scale.

    “GE was really well-known for homegrowing their managers.”

0.56

GE Capital's wholesale-funded, bank-like leverage made its competitive advantage fragile during the crisis.

6 pointscentrality 4/5
  • Using DuPont analysis, financial firm ROE is a function of asset returns and asset-to-equity leverage, so when GE was forced to hold more capital after the crisis, its leverage fell and its stellar returns on equity declined.

    “returns on equity for financial services firms are a function of two variables... And gone where GE's stellar returns on equity as it was forced to hold more and more capital.”

  • GE operated like a bank without bank regulation, using its AAA credit rating to borrow cheaply through wholesale funding (short-term market borrowing) rather than deposits, which gave it a competitive advantage that was fragile.

    “GE was like a bank, but it certainly wasn't regulated like one, and it could use its leverage and it's then AAA credit rating to obtain its cost of funds cheaper, which it did by obtaining what's called wholesale funding.”

  • Leverage can cause a financial firm to blow up overnight; firms in financial services have feast-and-famine histories, and value investor Seth Klarman has reminded investors of this risk.

    “With leverage, you can blow up overnight. In financial services, firms have a history of both feast and famine.”

  • Regulators concluded non-bank financial firms like GE Capital could be systemically important, so under Dodd-Frank GE began being regulated like a bank around mid-2016 as it shed most of its loans.

    “The regulators quickly figured out that it just wasn't financial institutions like banks that were systematically important, but non banks like GE Capital also formed part of this group and GE started to become regulated like a bank under the newly created Dodd-Frank legislation”

  • GE bought WMC, a subprime mortgage lender, in 2004, sustained losses before shuttering it, and was forced to pay a $1.5 billion civil penalty under the FIRREA statute for fraud.

    “GE also bought into WMC under Immelt. That was a subprime in all day mortgage lender in 2004.”

  • During the 2008 financial crisis, lenders fled the short-term loan market, and because GE Capital had few deposits to fall back on, GE lost its source of funds in a situation akin to a bank run, leaving it dependent on the government and on Berkshire Hathaway.

    “During famine, as was the case with the great recession in 2008, the market for sure termed loans right up as lenders fled, and GE simply didn't have much in the way of deposits to fall back on.”

0.56

Deconglomeratization will continue, favoring focused companies, so a breakup should re-rate the stock.

6 pointscentrality 4/5
  • Market trends are moving toward deconglomeratization, and the more winning formula is focused conglomerates, like Raytheon Technologies spinning off Otis and Carrier, since investors can diversify for themselves and separate businesses avoid competing for capital.

    “There's this trend towards deconglomeratization, and I don't think it'll end over time.”

  • The breakup should allow GE to be far more focused, given each business has very different investment profiles and capital allocation needs, and it is an important catalyst to re-rate the stock in line with peers.

    “The breakup we think really should allow GE to be a lot more focused, given each of these business' very different investment profiles and capital allocation needs”

  • The valuation debate is dominated by fear: investors have capitulated on GE's problems, worry that the spin-offs will not execute, and doubt that these businesses will ever earn the normalized earning power that bulls expect.

    “The conversation of the debate and the valuation debate is a lot more dominated by fear.”

  • Organizations need an intellectual environment where someone can present the alternative point of view and surface downside risks before any decision, and focus is worth more than the benefits of a centralized research center.

    “You're going to want an intellectual environment where if someone can present the alternative point of view and surface what are the downside risks to making any sort of decision.”

  • GE today focuses on three businesses: aviation, healthcare, and energy, with energy broken into power, renewables, and digital grid, and healthcare and energy slated to be spun off in early 2023 and early 2024.

    “At this point, GE is focused on three businesses: aviation, healthcare and energy”

  • The trend toward deconglomeratization will continue; the market caps of conglomerates will not grow, and the future is more focused conglomerates and pure-play companies, not GE-style all-in-one firms.

    “There's this trend towards deconglomeratization, and I don't think it'll end over time.”

0.45

Aviation's installed-base model creates a wide moat and strong operating leverage.

7 pointscentrality 3/5
  • GE's remaining businesses in aviation, healthcare and energy all compete in oligopolistic markets and share a strategy of building a large installed base of equipment to sell higher-margin parts and service revenue.

    “An advantage that GE has is that it competes in oligopolistic markets. To a certain extent, all these businesses try to focus a little bit on the same thing, and that is build a large installed base of equipment to sell higher margin parts and service revenue.”

  • Aviation's original equipment drivers are global passenger traffic and distance traveled (RPKs), while aftermarket demand depends on cycles flown, and RPKs are driven by middle-income growth in developing markets.

    “RPKs themselves are driven in our view by factors like middle income growth in developing markets.”

  • The LEAP engine, made in a joint venture with Safran, is single-sourced on Boeing's 737 MAX and dual-sourced on the Airbus A320neo family, where it competes with Pratt & Whitney's GTF, and no other company competes as effectively at delivering engines reliably at scale.

    “That engine is single sourced, meaning it's the only engine that Boeing uses in its 737 MAX, and it's dual sourced in the Airbus A320neo family”

  • Renewables has hardware plus service revenue but far less service revenue than aviation, because aviation's installed-base investment justifies multi-decade parts and service income while renewables does not.

    “There's hardware plus service revenue but it's not nearly as much service revenue. In aviation, the investment justifies the parts and service revenue. You've got multiple decades' worth of that.”

  • GE sells aviation engines at a loss to make it up in highly profitable parts and service revenue that provides multi-decade benefits, which is the key reason aviation deserves a wide moat; customers routinely pay 70% below list price, and discounts hover over 50% off list.

    “This is a true razor and blade business. GE sells these engines at a loss to make it up in highly profitable parts and service revenue, which provides it with multi-decade benefits.”

  • Aviation has a high degree of operating leverage, with a good degree of fixed costs, so incremental revenue drops through to margin and people will be surprised by how much the business re-rates.

    “That business has a high degree of operating leverage. There's a good degree of fixed costs and you'll see incremental margin drop down to the bottom line.”

  • Aviation RPKs have historically tracked GDP per capita, but COVID lockdowns decoupled them, creating pent-up demand for global passenger travel and a shift from products toward experiences such as eating out, which should benefit air travel.

    “the commercial aerospace and RPKs, or revenue passenger kilometers, tend to be coupled with GDP per capita trends. And that's been decoupled and that's because of the COVID lockdowns”

0.40

GE's past peak reflected GE Capital and management churn, not managerial brilliance.

5 pointscentrality 3/5
  • GE's peak was less about managerial brilliance than about GE Capital, and if management had truly been that brilliant, GE would not have run into the situation it faced later and lost its market value.

    “I think it's really mostly on GE Capital, and I think it's less to do with managerial brilliance.”

  • Under Larry Culp, GE is moving toward decentralization by making businesses responsible for their own P&Ls, taking on previously shared expenses, and owning lean implementation, safety, quality, delivery and cost.

    “GE under Larry Culp is moving under that direction, making sure that the businesses are really responsible and they become the center of gravity”

  • GE's practice of moving managers around frequently, which Munger criticized, loses domain management competence, unlike Roper Technologies, which keeps managers in place for long periods, and GE still makes this mistake.

    “the minute that you rotate these folks around, you lose that management competence in one specific area. And that's a mistake that GE still makes, I think, to this day.”

  • Industrial equipment is capital intensive and involves contract assets, so GE recognizes a sale long before it collects the cash, which made financing businesses look more profitable than manufacturing.

    “frequently involves booking of contract assets, which in accounting speak means that GE generally recognizes a sale of equipment long before it collects the rated cash.”

  • In 2000, GE was the most valuable company at nearly $600 billion, grew sales 16% to nearly $130 billion, and generated over $15 billion in operating cash flow; today GE generates about $74 billion in revenue, and its market value is a fraction of the peak.

    “GE was worth nearly $600 billion or so.”

0.35

Welch used GE Capital to smooth earnings, implying low-quality earnings and possible financial engineering.

3 pointscentrality 3/5
  • Welch used GE Capital as a vehicle for earnings management, spackling over the industrial businesses' weak quarterly results so GE could consistently beat earnings expectations, which implied low-quality earnings and at worst financial engineering.

    “Welch used GE Capital as a vehicle for earnings management by essentially spackling over the industrial businesses' week quarterly.”

  • A book alleges Jack Welch advised executives to keep a nickel of earnings per share in their pocket, which at best signals low-quality earnings and at worst amounts to financial engineering, a judgment Josh leaves to listeners.

    “There's a book out there that alleges Jack Welch always advises direct reports or execs to keep a nickel in your pocket of earnings per share”

  • GE's success in 2000 was mostly propped up by GE Capital, which constituted nearly 60% of earnings at its height, and GE would not have achieved its results without GE Capital growing to the size it did, a fact investors missed for a long time.

    “I really think it was mostly the latter, meaning that it was propped up quite a bit by GE Capital. You would not have had the success without GE Capital becoming the size it eventually became.”

0.26

Renewables faces structural competition and losses that management underappreciates.

5 pointscentrality 2/5
  • Renewables has a divide-and-conquer geographic dynamic because blades, towers and nacelles are costly to ship overseas and developers and financiers prefer homegrown turbine makers, though this protection is starting to erode.

    “The reason that's been the case is because the primary components like your blades, your towers, and your nacelles, they're tough to ship overseas for outside competitors.”

  • GE's renewables business moved from negative mid-single-digit operating margins to negative mid-teens in the last couple of quarters, driven partly by PTC dynamics and price competition, and Josh expects it will not break even until the middle of the decade.

    “Recently it had been earning the negative mid-single digit margins on the op margin side. In the last couple quarters we're now looking at negative mid teens.”

  • Josh is very skeptical of management's long-term renewables target and believes they underappreciate competitive risks, especially from Chinese competitors such as Goldwind and Ming Yang.

    “I'm really, really skeptical of management's long term target here, and I think they under-appreciate some of the competitive risks here.”

  • Scott Strazik, placed in charge of the power business, is expected to stem the bleeding, aiming for break-even and eventually a mid-single-digit operating margin by rightsizing backlog and taking costs out of the business.

    “And it's now looking like it'll be like a mid-single digit op margin business.”

  • GE is moving toward energy transition markets, skating to where the puck is going, while still servicing nuclear and fossil installed bases, and it inherited its wind and access to technology from acquisitions in the early 2000s.

    “I think they are just skating to where the puck is going, so to speak. There's demand with the energy transition.”

0.21

GE Healthcare converts cash efficiently and offers dividend and acquisition catalysts.

3 pointscentrality 2/5
  • Underappreciated catalysts for GE Healthcare include a potential dividend, since healthcare companies typically pay out 35 to 60% of earnings, and bolt-on acquisitions of early-stage startups in a dynamic space.

    “some underappreciated catalysts include the potential for a dividend payment, since healthcare companies typically pay out maybe 35 to 60% of their earnings out in dividends.”

  • GE Healthcare is the best free cash flow converter in GE's portfolio at about 100% of earnings converted to free cash flow, with high-teens operating margins through the cycle and lower mid-single-digit organic growth.

    “It's also the best free cash flow converter at about 100% of free cash converted to earnings.”

  • Under current management, capital is prioritized toward organic investment such as sustainable aircraft engine development with Safran in the mid-2030s, healthcare digital and AI investments, and offshore wind, while a roughly $3 billion buyback authorization should be used given the stock's low valuation.

    “I think in terms of share repurchases, the authorization's for about $3 billion. I'd like to see them use that, just given the stocks price to value ratio at this point.”

0.21

Hardware-plus-service economics do not always hold, since cycles and service demand vary by industry.

3 pointscentrality 2/5
  • The hardware-plus-service model does not always work: LED lighting resolved the light bulb maintenance cycle, while aerospace engines are complex and heavily regulated by the FAA, creating multi-decade parts and service demand.

    “Think about LED lighting, right? It resolved the light bulb maintenance cycle.”

  • Aviation is the longest-cycle business, taking longest to fill orders, while healthcare is a shorter mid-cycle business in which unfilled orders take about a year to fill based on backlog burn rates.

    “aviation, longest cycle business of the bunch, meaning it takes longest to fill orders. Healthcare, shorter, mid cycle business, meaning it takes about a year to fill unfilled orders”

  • About half of aviation services are sold via time and materials, and half through flight hour services agreements (power by the hour), which give GE a steadier cash flow stream but require it to maximize on-wing time.

    “we think about half are sold via time and materials, while the other half is sold through what's called flight hour services agreements.”

0.18

Specific GE segments face distinct growth, competition, and cost dynamics in their markets.

7 pointscentrality 1/5
  • GE's power business should benefit from the energy transition as a bridge technology, because intermittency of renewables is a problem that batteries cannot correct at scale.

    “should really benefit from the energy transition as a bridge to gap technology as the world moves to renewables, given the intermittency issues, which is a problem batteries can't correct”

  • Chinese wind competitors such as Goldwind and Ming Yang are an underappreciated risk to GE's renewables business and will likely keep it under pressure.

    “I'm really, really skeptical of management's long term target here, and I think they under-appreciate some of the competitive risks here.”

  • In normal times aviation grows at a mid-single-digit organic rate with about 20% operating margins through the cycle, while Josh's model expects a higher near-term growth rate driven by pent-up demand.

    “Normal times, I'd say this business is probably a mid single digit organic grower, with 20% op margins through the cycle”

  • GE and Safran are developing a sustainable engine with an open fan and no nacelle, targeted for the middle of the 2030s, which will absorb substantial R&D spending.

    “towards the middle of the 2030s decade, developing the sustainable engine, removing the nacelle”

  • GE Healthcare should participate meaningfully in bolt-on acquisitions of early-stage startups, as shown by its BK Medical deal, because the space is dynamic with rising competition from new entrants.

    “I expect them to increasingly do this over time.”

  • Narrowbody jets are increasingly important because they can travel longer distances, such as the Airbus A321 XLR's 4,300 nautical miles, roughly London to Miami.

    “The Airbus A321 XLR, which is Airbus's extra long jet, can travel 4,300 nautical miles, which is the distance from London to Miami.”

  • The LEAP engine lists at about $14 million, but airlines and airframers routinely pay around 70% below list, with discounts that analysts estimate still hover over 50% off list.

    “assume the list price of a LEAP engine is about $14 million.”