Can Eric Glyman count?

You probably haven’t heard of Ramp. It doesn’t make an app, an LLM, or a social media platform. Instead, it sells software that helps businesses manage their expenses. Sounds boring, right?

Nevertheless, Ramp is valued at over $20 billion, and is, as of this writing, the 25th most valuable private company in the world.

Ramp’s CEO, Eric Glyman, and Kyle Harrison, a partner at one of Ramp’s investors, “Contrary”, published an article (a “deep dive”) called “The Efficiency Formula”. The thesis is that the U.S. government is horribly mismanaged. It spends far more than it collects in taxes; it wastes money on multiple “drag baskets”, and, by adopting tools like those provided by Ramp, the US government could be “more efficient and more effective”.

The authors write, “If the US government were a company, its CFO [chief financial officer] would have been fired a long time ago.” Ironically, the authors are guilty of making errors of arithmetic so basic that they should cast doubt on whether Ramp’s CEO can count, let alone be trusted to manage a $20 billion company.

The authors write:

Categories of waste in government spending are what we call “drag baskets.” This includes $100 billion in Medicare and Medicaid fraud, $200 billion in SBA loan fraud, and $236 billion in “improper payments.” These three “drag baskets” alone accounted for at least $536 billion wasted. That’s 7.9% of the government’s entire budget down the drain.

In short, $100 billion (Medicare/Medicaid fraud) + $200 billion (SBA loan fraud) + $236 billion (other fraud) = $536 billion (overall fraud per year). That’s a lot — roughly 8% of the federal budget, as the authors point out.

Each number in this calculation is from a different source. That’s the first red flag. The $100 billion is from a report on healthcare fraud. The $200 billion is from a report published by the Office of the Inspector General of the Small Business Administration (SBA), about fraud associated with pandemic lending programs for small businesses. And the $236 billion is from yet another report, by the Government Accountability Office (GAO).

You might expect that you can’t cleanly add estimates from different sources. You risk double-counting. If you got that far, you’re smarter than a CEO! As it turns out, the $236 billion in improper payments (the third estimate) already includes improper payments for Medicare and Medicaid ($100 billion, close to the first estimate) and for PPP loans (part of the second estimate).

That’s not all. The $200 billion in SBA loan waste isn’t even an annual figure. The SBA made $1.2 trillion in loans during the pandemic to keep small businesses afloat. Of these loans, $200 billion were deemed fraudulent. But that’s clearly not waste or fraud that’s happening every year. Finally, the $236 billion in improper payments the GAO cited includes overpayments as well as underpayments. There’s only $175 billion in clear overpayments: money that Ramp might be able to recoup. All in all, the inefficiency Ramp identifies is probably 2-3x smaller than what was initially quoted. But what’s a few hundred billion between friends?

There are some broader lessons here, besides the fact that (obviously) you can be a dummy and also a multimillionaire. Much of the government’s budget falls into three categories: defense, Social Security, and healthcare (Medicare/Medicaid/etc.). It is true that the government could, and should, be better about minimizing waste and fraud. But even if, somehow, we managed to root out all improper payments, we’d come nowhere close to balancing the budget.

The problem is what we intentionally spend on, not what we accidentally waste. If we want to balance the budget, we have to either raise taxes or cut spending. There is no magical third way, no deus ex machina in the form of a SaaS product for expense management.

Even DOGE figured this out, albeit unintentionally. It failed to make much of a dent in government spending. In areas it did cut, like USAID, it saved trivial amounts of money and endangered the lives of millions. In areas it hardly cut, like defense, it could have saved billions but didn’t (more on that later). Sahil Lavingia, a somewhat famous tech person who worked for DOGE for a couple of months before being fired, remarked that he was “pretty surprised, actually, at how efficient the government was”.

Speed and minimizing waste often go in opposite directions. The government is normally pretty thrifty but moves slowly. It usually imposes burdens on individuals trying to collect money, and has stringent internal processes to prevent government employees from absconding with taxpayer money. These rules can get in the way of helping people with legitimate claims, though. The authors of the SBA loan fraud report wrote, “It is noteworthy that SBA executed over 14 years’ worth of lending within 14 days.” During the pandemic, the government tried to act fast. It made heroic efforts to protect workers’ paychecks and save the economy. It isn’t too surprising that this was accompanied by enormous waste. Ramp’s solution, of imposing more controls, would have saved money. But it might have cost the American people a lot more. It is slightly ironic that a tech company like Ramp would complain that the government was “moving fast and breaking things”.

Bad math is the most obvious problem with “The Efficiency Formula”. But it’s not the only one. The authors show an understanding of economics and the federal government that is probably below the level of most college students. (Glyman studied economics at Harvard, so I’m not sure what his excuse is.)

The authors claim that federal government should follow the “golden rule of budgeting: make more money than you spend”. (As it turns out the federal government should not be run like a business, and, in particular, it’s not trying to maximize the amount of money it “makes”.) They write that “raising the debt ceiling” amounts to “kicking the can down the road”. (Perhaps they’d prefer the alternative of defaulting on Treasuries and throwing the bond market into crisis?) They claim, hilariously, that "The average American business is being held to a higher standard than the US government.” And that “The size of federal government spending in relation to GDP has grown by an order of magnitude over the past 100 years. In 1930, the percentage of spending to GDP was 3%; by 2023, it reached 22%.” Hmm, I wonder what was happening in 1930?

One section that made me chuckle was this:

Prescription drug costs illustrate the problem. While essential for two-thirds of Americans, prices have increased almost 40% in the last decade, partly due to pharmacy-drug company relationships that favor “branded drugs over less expensive generics.” Medicare and Medicaid lack transaction context, but card programs could enable alerts for branded vs. generic prescriptions and show which pharmacies are clearly favoring brands over generic options.

The idea that prescription drug costs would go down meaningfully if we had “card programs” that gave us alerts about which pharmacies at which to pick up our prescriptions is laughable. It’s the “Nudge-” style “libertarian paternalism” that hardly ever works. Do you know how other countries manage prescription drug costs? Their healthcare systems negotiate directly with drug companies to bring prices down. Medicare, until relatively recently, wasn’t even allowed to do that — thanks to George W. Bush.

All of this points to a bigger problem: you can’t talk about federal spending and budget deficits unless you talk about politics, and, in particular, the Republican Party. But the authors of “The Efficiency Formula” studiously avoid the topic. Again — it would be nice if there were some apolitical way to eliminate budget deficits, which is probably why pointing to waste and fraud, or mentioning behavioral nudges, is so appealing. (”[Our suggested] improvements don’t require us to cut back on Medicare, Medicaid, or SBA loans, which are critical programs that support millions of people in their health and livelihood.”) But, sadly, that isn’t the world we live in.

The authors complain a little about wasteful defense spending but never point out who supports it. (To be fair, it’s both parties, but one more than the other.) The authors write that “businesses are cyclical, with revenue and earnings rising and falling with changing market conditions. However, US government spending has consistently increased regardless of market conditions”. I’m curious if the authors know which U.S. presidents have presided over massive increases in budget deficits. Would they be surprised to learn that they’re all from the same political party as many of Ramp’s investors?

One last point. I’ve been slightly imprecise when talking about “waste, fraud, and abuse”. There are plenty of examples of government spending that is wasteful but not fraudulent. It’s telling that the authors have very little to say about government boondoggles in defense spending, but plenty to say about Medicaid, Medicare, and pandemic programs. It probably has something to do with the fact that one of Contrary’s portfolio companies is Anduril, a defense contractor.

The thing about fraud and waste is that it’s bad for the public but good for the private sector. Medicare fraud is bad for Medicare recipients but good for Republican Senator Rick Scott, who, as a healthcare CEO, presided over the largest Medicare fraud in history. SBA loan fraud is bad for taxpayers but good for the small businesses who swindled the government into paying for their Bentleys, Ferraris, and Lamborghinis. Defense waste is bad for our military, but good for Anduril.

And so it is for Ramp. Remember why “The Efficiency Formula” was published. It was a sales pitch: introduce Ramp and save the taxpayer millions of dollars. The problem is that the federal government already has a payments system that, by all accounts, performs fairly well.

Gruenbaum and acting GSA administrator Stephen Ehikian entered the agency with a strong belief that SmartPay and other government payment programs were rife with fraud or waste, causing huge losses, sources within GSA say — an idea echoed in Ramp’s January memo.

Yet both GOP and Democratic budget experts, as well as former GSA officials, describe that view as ill-informed. SmartPay, which provides Visa and Mastercard charge cards to government employees, enables the federal workforce to purchase office supplies and equipment, book travel and pay for gas.

The cards typically are used to fund travel and purchases up to $10,000.

“SmartPay is the lifeblood of the government,” said former GSA commissioner Sonny Hashmi, who oversaw the program. “It’s a well-run program that solves real world problems … with exceptional levels of oversight and fraud prevention already baked in.”

Jessica Riedl, a GOP budget expert at the conservative Manhattan Institute think tank, said the notion that there was significant fraud in the charge card technology was far-fetched. She had criticized waste in government credit card programs before the latest SmartPay system was implemented in 2018.

“This was a huge problem about 20-25 years ago,” she said. “In the past 15 years, there have been new controls put into government credit card purchases.”

See, if it’s for other people, it’s waste and fraud. But if it’s for us, it’s good business.