Log in
Yesterday, Amazon Web Services, the cloud services backbone of the internet, had a serious glitch that took down a large chunk of the internet. How much did this outage cost? That’s not clear. But it was certainly a big number, with some estimates in the hundreds of billions of dollars.
Many people experienced the AWS outage in terms of time wasted, Zoom meetings not working, random services breaking, or apps hanging up or not launching. I had trouble dealing with Verizon, their customer service team took 15 minutes to respond to a chat query, and explained they were having “technical issues” on their end. The outage affected everyone from Netflix to Snapchat to Venmo to thousands of other vital products and services. Here’s the CEO of Eight Sleep, which makes internet-connected beds, apologizing for the outage and its affect on the sleep of its customers.
Leaving aside why beds need to be connected to the internet, let’s just stipulate that some sleep got messed up. And all of these costs were incurred because of dysfunction at AWS. We don’t have a name for the externalities induced by the market power here.
In 2022, Cory Doctorow described the cycle of decay of tech platforms, where they lock you in and then decrease overall quality. He deemed it “enshittification.” I think it’s worth offering a cousin to this term, which I’ll call “Corporate Sludge.” Corporate sludge is the cost, or costs, of an excessively monopolized and financialized economy, that do not show up on a balance sheet.
Here’s what I mean. According to Amazon’s internal financials, AWS has a high profit margin. In 2024 it had $107 billion in revenue, and generated $39.8 billion in profit, with is a 37% operating income. A normal product or service, when faced with a catastrophe like the AWS outage, would take a financial hit. Yet here’s the stock of Amazon yesterday.
In other words, the costs of the AWS outage did not show up on the balance sheet directly responsible for it, or in the equity markets supposedly measuring long-term expectations of corporate profits. Economists would call the wasted time a “negative externality,” it’s the equivalent of pollution. And while that cost doesn’t show up anywhere we can affirmatively identify, someone has to pay for it. Those missed meetings, that lost production, it raises costs for virtually everyone, a little. This cost is what economists or government statisticians just don’t see, because it isn’t measured. But that doesn’t mean it’s not real.
Once you start looking, you start realizing that corporate sludge is everywhere. I did an interview with corporate procurement specialist Rich Ham, and he told me that big corporations laid off most of their procurement teams as a cost-cutting measure during the financial crisis. Since Wall Street penalizes CEOs for hiring people, and rewards them for firing, they won’t spend to rebuild those teams. As a result, suppliers of things like uniforms, waste disposal, guards, and pest control massively gouge the corporate world. According to Ham, corporate procurement costs are going up 7% a year. And those increased costs, that corporate sludge, is passed along as higher prices, even as accounting profits aren’t improving. Executives think they are being efficient - headcount is down - but then they wonder why everything costs so much.
I suspect, though it goes unmeasured, that a lot of the increase in inflation that no one can quite explain is a result of corporate sludge. It’s a bit like ‘dark matter,’ a fudge factor astronomers created to describe why their theories of the properties of matter can’t explain the expansion of the universe. Similarly, I don’t believe economists have a good explanation for inflationary increases of the last few years. They certainly understand that supply chains broke, but they can’t describe why prices didn’t go back down once they were repaired. And I don’t think anyone can really explain why the economy seems to be booming, while ordinary people are unhappy. My fudge factor to explain it is corporate sludge - there’s massive inefficiency everywhere as a result of hidden market power that doesn’t show up on balance sheets, but shows up as time wasted, anxiety, and extra costs where you don’t expect it.
Health Care Sludge
I’ve been thinking about the concept of corporate sludge for about a year, after the debate over health insurance in the U.S. prompted by Luigi Mangione’s alleged killing of a health care CEO. At the time, there was a lot of back-and-forth about popular anger at insurers. Economic commentator Noah Smith wrote a piece that bothered me a lot, explaining why we were focused on the wrong bad guy. Here’s the headline:
He explained that American anger is misplaced. Insurance companies, he wrote, are generally efficient pass-throughs that got the blame for covering for greedy doctors. What was Smith’s evidence? Well, it was a balance sheet analysis. While UnitedHealth Group’s revenue is on the order of $400 billion, he explained, the company had a thin profit margin of just 6.11%. It’s actually efficiently run, not greedy. What kind of a villain or monopolist passes most of its revenue on to someone else? It’s a rational argument from Smith, and yet, not persuasive. People hate United Health Group, because we know that analyzing accounting profits excludes the real costs of its behavior.
The experience of health care is full of corporate sludge, from having to dispute weird bills to being steered to medication that may or may not be correct, to doctors spending their time fighting with bureaucrats over reimbursements and audits. In February 2024, UnitedHealth Group’s Change subsidiary got hacked, and it shut down cash flow to doctors, pharmacists, and hospitals, in some cases for months. Like the AWS outage, that didn’t affect UHG’s profit. But it certainly had a cost.
In other words, to look only at accounting profits is to miss the genuine costs of monopoly or financialization, which is the corporate sludge embedded in an economic system where the actual stakeholders who must use the system, patients and clinicians in the case of health care, have little power. And it’s not just UHG. In our current model of tight-fisted financiers choosing how to allocate health resources, somehow U.S. hospitals spend twice as much on administrative costs as they do on direct patient care. Yet that doesn’t show up anywhere as accounting profit; hospitals are constantly explaining how strapped they are for cash.
Yet, this money is going somewhere. I noted this dynamic a few weeks ago when I profiled a monopoly in unnecessary hospital surveys, which is the result of a merger between survey companies Qualtrics and Press Ganey, as well as a regulatory mandate by the Affordable Care Act. Ryan Smith, the billionaire owner of the Utah Jazz, founded Qualtrics, so one way to understand corporate sludge is that it’s shifting unnoticeable amounts of money from each of us to people who in turn buy sports teams. There are innumerable economic termites in health care, like billing codes that the American Medical Association has a copyright on, or electronic record keeping systems, or HIPAA-compliant note-taking software. And then there are also just big unnecessary billing departments fighting with other big unnecessary billing departments.
Noah Smith would look at the profit margin of hospitals, insurers, distributors, PBMs, et al and ask, “where’s the problem?” What he’d find is that nearly everything in health care has low margins or can be made to seen that way. So just by that analysis, Smith would see nothing but an efficiently run health care system. Yet the U.S. spends three times as much as every other country on its system, gets worse results, and generates a lot of health care millionaires who are good at pricing arbitrage.
The Keys Aren’t Near the Lamp Post
In other words, one of the more important reasons to look at corporate sludge as a meaningful challenge in the business world is that it helps expose something economists don’t measure, which is private inefficiency. A firm with market power can harvest that market power as accounting profits, or as additional administrative bloat or a higher cost supply chain. While economically these are similar, in terms of accounting and rhetoric, they are not.
For instance, one of the arguments that price gouging didn’t matter in food inflation during Covid was that grocery stores have thin operating margins. If there’s so much gouging, where’s all that extra profit? As I noted before, cutting a corporate procurement staff means a supermarket chain pays more for waste disposal or uniforms, but a profit margin analysis would miss that the price of milk is higher because a vendor is screwing the grocery chain.
Moreover, looking purely at the margins of a Kroger or Walmart misses how these companies organize an entire supply chain, and incentivize the sale of more expensive food in general. For instance, most large grocers make a fair amount of money through “slotting fees,” where branded food companies pay for display space for their wares, which is a form of price discrimination. Slotting fee contracts, or “trade spend,” make it very hard for smaller companies that sell cheap, fresh food to get into supermarkets, because they can’t afford to get on the shelf. It’s the ultra-processed food that permeates.
Since the government stopped enforcing laws against price discrimination in the 1980s, the food industry learned “how to sell larger quantities of low-nutrient processed foods merely by manipulating their placement.” One consequence is that “rates of obesity in the United States doubled,” but another is that cheap regionally produced food producers disappeared, because they couldn’t get onto the shelf. So prices across the supply chain went up.
If you just look at the operating margins of the big grocers, you’d miss the transformation of our food industry from a low cost locally based high quality distribution system to a high cost globalized low quality one. You can argue all you want about economies of scale in food processing, but it’s just absurd to believe that the overhead of a major corporation like Kraft makes it more efficient than a local food producer. I mean, it’s been a few years since Covid, and the supply chains have cleared, yet prices didn’t come back down. No one can explain why. The answer is corporate sludge, hidden inefficiencies that are a result of market power.
Electric Sludge
Another example full of sludge comes in the details of the operations of investor-owned electric utilities, which are increasing prices dramatically and blaming it on the build-out of data centers. While data center growth is meaningful, utility prices in a lot of places were increasing before the massive AI investment, and states without data center growth are also seeing much higher prices. Moreover, there is an odd discontinuity in price increases, with some utilities hiking costs and others keeping them lower. What explains the difference? In America, most of our utilities are investor-owned, but some are owned by cities or are structured as cooperatives. Mark Ellis, a utility analyst, sent me this graphic of the change in electricity pricing for investor-owned vs publicly-owned utilities.
Publicly owned utility rates have increased faster than investor-owned ones in only a few states.
What’s going on here? Well, investor-owned utilities get regulators to let them raise prices based on a supposed need to send high profits to Wall Street, whereas publicly owned ones don’t. But the costs are much higher than what we see go to investors. Just looking at some of the public filings of utilities shows there’s also a lot of bloat. For instance, while the publicly owned utilities have a few lawyers on staff, private utilities of significant size seem to employ the equivalent of an internal mid-size law firm, paying dozens of lawyers up to a million dollars apiece for legal, regulatory, and lobbying work. This kind of gold-plating is no doubt happening down the line, from replacing poles when it’s unnecessary to do so, to bringing on unnecessary well-paid administrative staff to promote silly diversity initiatives.
All of these costs are real, and must be paid.
Now, getting back to AWS, most people, when looking at the situation, observed that too much of the internet is based on a few chokepoints. Fast Company had an article titled The AWS outage reveals the web’s massive centralization problem, European leaders expressed alarm on their dependency on U.S. big tech giants, and a giant Reddit thread - AWS Services are down, This Is Why Monopolies Should Be Banned - drew thousands of comments. (Ironically the thread itself went down for a period because of the outage.)
But none of these people said the problem with AWS is that it has a high profit margin. And no one focused on market share, or whether customers can in some theoretical world switch to another cloud provider. That’s because normal people can see what’s going on, even if economists can’t. Corporate sludge is everywhere.
Thanks for reading! Your tips make this newsletter what it is, so please send me tips on weird monopolies, stories I’ve missed, or other thoughts. And if you liked this issue of BIG, you can sign up here for more issues, a newsletter on how to restore fair commerce, innovation, and democracy. Consider becoming a paying subscriber to support this work, or if you are a paying subscriber, giving a gift subscription to a friend, colleague, or family member. If you really liked it, read my book, Goliath: The 100-Year War Between Monopoly Power and Democracy.
cheers,
Matt Stoller
