The Real Deep State
We are, it turns out, governed by the swamp.
It’s 2002 and drug companies have a real problem on their hands.
They are spending more, more, more on sales representatives to visit doctors, to hawk new wares, to leave samples, and to sell more drugs. It was getting to the point where hiring more sales people just no longer made sense..
Martin Elling saw a solution. He wrote:
Current sales efforts encourage those who are already high prescribers to prescribe more. It would, though, be more useful to assess the lifetime value of a doctor’s prescription habits. Not one company we surveyed does so regularly, but the information is available: when a patient fills a prescription, the order is stored in a database that can be matched for drug, producer, and physician. The data can be used along with information on the current stage of a physician’s career (just started practicing? close to retirement?) to target physicians who are most likely to prescribe more of a given drug over time, no matter how much or how little they prescribe at the moment.
It’s a clever idea. Using data analytics to target doctors who are under-prescribing, but who may write more scripts with a bit of a push, could be more useful than squeezing more from pill-happy docs. Target enough resources to those doctors, and perhaps they might start prescribing more drugs, more expensive drugs, new drugs.
Elling wasn’t a physician, chemist, pharmacist, or a scientist at all — he was a lawyer and consultant. This article, it wasn’t in a doctor’s trade magazine or the Wall Street Journal: It was in McKinsey Quarterly, the in-house publication of consulting giant McKinsey & Company.
That kind of forward-thinking earned fans. Drug giant Purdue signed up with McKinsey in 2004, spending more than $80 million with the firm over the next 15 years.
The money was well worth it. McKinsey promised to “Turbocharge Purdue’s Sales Engine” for opioid OxyContin. And turbocharge they did: Purdue sold some $35 billion worth of the drug. That engine didn’t stop running after McKinsey and its top executives pled guilty to criminal and civil charges for “misbranding” the drug in 2007, erroneously and dangerously claiming it to be safe and addiction-resistant. McKinsey and Purdue were just getting started.
In 2013, McKinsey drew up a plan to hawk OxyContin: Sell more drugs to high-prescribing doctors, get them to prescribe more, and — worse yet — try cutting out the doctor altogether and prescribe directly to consumers. When critics piped up, alarmed at marketing an incredibly potent drug like candy, McKinsey came up with a plan to quash their objections.
McKinsey, always eyeing new cashflows, wasn’t going to just help Purdue. It has no loyalty in that respect. It advised Johnson & Johnson in a Powerpoint, the firm’s favored method of attack, that targeting doctors treating back pain would be a great way to turbocharge opioid sales.
At the same time, McKinsey was consulting for the very body that was supposed to be keeping an eye on prescription abuse: The Food and Drug Administration. Sometimes, the same consultants were acting as double agents: Proffering advice for the drug companies on how to evade regulation that would limit the sale of their blockbuster drugs just as they were advising the U.S. government.
While McKinsey is not uniquely responsible for the opioid crisis, it nevertheless created the pipeline of powerful prescription opioids that flooded North America over the past twenty years and kept us blind to the growing opioid addiction crisis all the while. For their role in that murderous conspiracy, they paid out nearly $600 million in 2021. They admitted no wrongdoing. Martin Elling was fired for discussing destroying records to hide their involvement.
Dominic Barton, apparently, had no idea. As the global managing director for McKinsey from 2009 to 2018, he told the Globe & Mail he “simply had no knowledge of the work undertaken for Purdue by my former colleagues.”
No matter. McKinsey’s allure is so strong that, in 2017, Barton was tapped by Prime Minister Justin Trudeau to be chair of the Canadian federal government’s blue chip panel on turbocharging the economy: The Advisory Council on Economic Growth. He served in that role, while also heading up McKinsey, until Trudeau gave him an even bigger job: Canadian ambassador to China. (Beijing: Another star client of McKinsey.)
McKinsey has done well under Trudeau. At least $77 million in contracts and supply agreements with various government departments since Trudeau took office, by my count.
This week, on Bug-eyed and Shameless, I want to talk about McKinsey and Canada: And, more broadly, what it means to outsource the work of government to a private consultancy.
A lot of work that I do in this newsletter revolves around conspiracy theories and disinformation about who really runs the government. The wrong answer, as we know well, is often some combination of “globalists,” “Jews,” George Soros, “the woke mafia,” Klaus Schwab, Bill Gates, and so on.
But I think it’s important that we don’t allow that paranoia distract us from the fact that there is, increasingly, a deep state here in the West. But it’s not some Davos-based thinktank or some shadowy cabal. It’s the consulting firms that are, increasingly, hoovering up massive amounts of public money to outsource the work of our civil service, often while mired in obscene conflicts of interests. This work isn’t about creating a new world order or destroying our democracy, but there is good reason to think that it is weakening the very foundation of our state and moving critical decision-making outside of government where scrutiny becomes more difficult. This transition has largely been kept from the public, never approved by our democratic institutions, and core questions remain about whether we’re getting actual value from these deals — or whether we’re sleepwalking into another crisis like the one McKinsey has, decades too late, sought penance for.
Then, later, for paying subscribers: A peek into the voluminous January 6 depositions. And: A word about the only useful instalment, thus far, of the Twitter Files.
After the extent of McKinsey’s role in pushing OxyContin across the United States because clear, there was a race to see who could hold the consultant to account. The New York Attorney General led the charge, but most states joined the fight — ultimately leading to the half-billion dollar settlement. A number of members of Congress followed suit, hauling in McKinsey executives to testify. Last April, Congress held hearings entitled: “McKinsey & Company's Conduct and Conflicts at the Heart of the Opioid Epidemic.” Various government agencies and democratic bodies across the U.S. have launched investigations into McKinsey for a suite of other reasons.
Across the Atlantic, a sprawling investigation in France is looking at McKinsey’s dual duties as a favored contractor of the Macron government and as a campaign advisor to Macron’s presidential elections. The Senate has launched a significant investigation into a suite of consulting firms, including McKinsey, and their relationship with the Élysée, particularly when it came to the pandemic. That commission “revealed a sprawling phenomenon.” (The actual French word, tentaculaire, doesn’t quite have a non-creepy English translation.)
The introduction to the 385-page study. offers a perfectly succinct encapsulation of the problem: “Consulting firms intervened at the heart of public policy, raising two principle questions: Our vision of the state and its sovereignty over private firms, on one hand; and the proper use of public funds, on the other.'“
Questions about whether or not McKinsey was actually helping or hurting seemed immaterial to the Macron government. “At the request of the health minister, McKinsey took over coordination between the state and one of its agencies, Public Health France,” the study reads. This liaison role meant McKinsey was organizing two daily briefings between two government agencies. This “omnipresence” wasn’t always welcome: Public health staff bemoaned being called into a meeting by a private firm to demand progress at 3pm on plans that were devised at 9am.
What was McKinsey and their ilk actually bringing to the table? We imagine consultants coming in and improving efficiencies, cutting red tape, breaking down silos — disrupting. Yet the French Senate found that some of their strategies involved team-building exercises using lego, or imagining civil servants were the crew abroad a pirate ship. Bureaucrats complained that, on paper, they 10 hours of meetings with the consultants per month, but in reality it was much higher because they never ended on time. Surely you would expect consultants obsessed with efficiency would reduce meetings, not increase them. “I regularly got the feeling we were being infantilized,” one civil servant told the study.
Each consultant cost the French state €1,528 per day. And their actual value-add seemed, at best, murky.
The U.S. and France aren’t the only places where McKinsey has earned sweetheart deals while maintaining a questionable ethical record. The company admitted to a wide-ranging corruption scheme in South Africa, feigned shock that its report identifying Saudi Arabia’s most vocal online dissidents might have been misused by their clients in Riyadh, and continues to insist that it had no role in enabling China’s genocide of the Uyghurs despite documents suggesting the contrary.
Canada is only just coming alive to the problem. The Globe & Mail has hit on Barton’s role in the company during the heyday of its opioid-pushing program and revealed the explosive growth in outsourcing to the firm, whilst the opposition and media in Quebec City have drawn attention to the fact that Premier Francois Legault outsourced significant parts of Quebec’s pandemic response to the U.S. firm. But there has been no broad study of the consultant’s increasing status as the Canadian deep state.
That seemed to change when Radio-Canada published a report this week about the growing reliance of Ottawa on the firm. (Largely, it should be said, repeating the Globe’s reporting from early last year.)
That report (which actually undercounts the value of McKinsey’s contracts under the Trudeau government by about $11 million) has since garnered headlines in La Presse and in various English outlets. Opposition Members of Parliament are now mounting calls for a Canadian inquiry. My Substack colleague Paul Wells puts a fine point on a lot of the issues here in his dispatch from this week:
If we’re ever going to fight the illegitimate, paranoid, and conspiratorial distrust in government, we’re going to have to occasionally roll up our sleeves and dig deep into where distrust in government is warranted. So let’s do that.
Let’s start with a simple question: What’s so bad about governments outsourcing to consultants?
And that comes with a relatively simple answer: Not, necessarily, anything.
Consultants exist because, sometimes, institutions need a bit of outside expertise on a matter where that internal capacity doesn’t exist. Or they need a fresh set of eyes to dispassionately understand what’s working, and what’s not. Pushed, you can guarantee that the Trudeau government will offer exactly that as the reason why McKinsey has gotten so much work from Ottawa these past few years.
But there’s obvious downsides. Sometimes an outside perspective is just an inexperienced one. Sure a business degree and a $500 suit are great, but that doesn’t mean you’re going to pick up a file and understand it with more depth than than the team of 20 who have, combined, a century of experience working on it. The more you rely on outside help, the less you develop that expertise inside your own civil service. You ritualize and weaponize those consultant reports — civil servants launder their opinions through consultants to convince their political masters that a certain course of action is the right one, and political staff seek out consultants’ opinions when they don’t like the advice they’re getting from the civil service. That basically adds to the bureaucracy, it doesn’t reduce it. Oh, and it’s also wildly expensive.
That is the general pro-and-con list. The reason McKinsey has earned so much scrutiny is because they come with their own raft of externalities.
Since the Radio-Canada story came out last week, I’ve been speed reading through When McKinsey Comes to Town, by Walt Bogdanich and Michael Forsythe, the authoritative book on the firm. It lays bare, in painful detail, just how the firm puts profit before ethics, bakes in bad policy, and plays both sides of the table.
Bogdanich and Forsythe break down McKinsey’s role in enabling the Trump administration’s brutal immigrant detention policies through their lucrative contract with Immigration Customs Enforcement (ICE).
McKinsey looked at the ‘daily bed rate’ for different facilities and found that some were far cheaper than others. Why not move detainees to the cheaper facilities and fill those beds first? On the surface, it made sense, but the reality was that those beds were cheaper because many were in city and county jails. Rural counties coveted ICE contracts that paid them to house people in otherwise vacant cells, generating much-needed revenue.
“They were cheaper because they are shitty sheriff’s jails, and they just want the money.,” one former Department of Homeland Security official said, calling them “garbage beds.”
[…] ICE was paying [private prison contractor] CoreCivic $157 million a year for it to run Dilley, the biggest family detention facility in the ICE archipelago. McKinsey’s slides showed the ‘should cost’ was drastically lower: $40 million, close to a 75 percent cut. Realistically, McKinsey surmised, the agency could push for cuts to the contract of up to $90 million. Where was the fat? The details [obtained by ProPublica] are redacted, but the cuts came in areas such as “staffing,” “food & supplies,” depreciation, and “other.”
In another chapter, the authors lay out how effective McKinsey became at winning sole-sourced, uncompetitive, or seemingly rigged contracts to manage state health services — in the case of Missouri, that led to tens of thousands of people being kicked off Medicaid even as healthcare costs remained high. All the while, huge new contracts were going to other McKinsey clients.
In Illinois, suspicion about McKinsey was so acute that one lawmaker, Greg Harris, held hearings and hauled up Felicia Norwood, the state’s Medicaid director.
“Is there no internal capacity to do any of this work without going out and spending tens of millions of dollars ore?” Harris asked.
Norwood said no. “The work definitely can’t be done by those employees.” Strategizing organizational change was deemed beyond their capabilities.
[…] McKinsey, for one, was not particularly interested in having anyone look over its shoulder, even for its government work. At one hearing, Harris broached the firm’s penchant for secrecy with Norwood. He cited a provision in McKinsey’s contract “that neither the state nor McKinsey can refer to each other or attribute any information to the other party in an external communication including news releases pertaining to this contract.”
“Is this is a standard provision in Illinois contracts?” Harris asked.
Norwood said she thought it was but couldn’t say for sure.
[…] Harris said some deliverables appeared vague, such as helping support and prepare leadership “for provider engagement sessions at appropriate cadence to be defined.” At this point, some people began to laugh, he said.
“Do we have cadence defined?” Harris asked with a touch of sarcasm.
“Is that your question?”
“Yeah. I’m just going through these, just wondering what we have gotten for the million dollars.”
“We will go back and get something for you that actually specifically outlines everything we’ve paid and the deliverables that we’ve received,” Norwood promised.
Six months later, Norwood left her state job for one of McKinsey’s biggest clients.
So what did Canada get for $77 million?
Answering that question is as tricky as defining cadence, it seems. Canada’s procurement process continues to be incredibly opaque: There has not been, best I can tell, any public tender documents that you normally expect to see in procurements like this — requests for proposals, statements of work, etc. The final McKinsey reports have never been made public, and McKinsey tends to insist they never will be. Anecdotally, I’ve heard that the Government of Canada is reticent to release documents produced by consultants via the Access to Information system.
Concerns about McKinsey’s role in enabling ICE’s child detention network didn’t ward away their Canadian cousins, the Canadian Border Services Agency, from inking $6 million in competitive contracts with McKinsey from 2016 to 2018, and then re-upping for a new sole-sourced $2 million deal in 2022.
It’s not even clear what, exactly, these contracts are for. Inside a 2019 CBSA briefing document, there are some fleeting mentions of a McKinsey report helping to innovate IT systems at the border agency. McKinsey, however, isn’t an IT company. But, according to a particularly inscrutable quote from said report, McKinsey encouraged CBSA to “implement structure to encourage innovation and hasten delivery of advanced analytics, automation…" The quote trails off.
There’s a bit more transparency behind a $28 million deal with Citizenship and Immigration Canada. The ministry provided a breakdown of what, exactly, McKinsey was providing for the money. Some of those objectives certainly strike the right note of what a consultant should be doing: Helping the bureaucracy build capacity within, so it can reduce dependance on third parties.
Other gigs are eyebrow-raising. Take McKinsey’s role in working with the Department of Defence, worth some $13 million in recent years. The Canadian Armed Forces has been rocked by a deluge of allegations of sexual misconduct in recent years. In 2021, the Forces began an overhaul of its complaint process, with an eye to making the entire process friendlier. To do that, it brought in McKinsey.
“As I understand it, McKinsey has proposed a possible new complaints mechanism to provide a unified approach to complaints,” reads the findings of a 2022 external review. “This new complaints mechanism includes a ‘single disclosure team’ for receipt of disclosure and reports of misconduct. This point of disclosure is meant to apply to sexual misconduct, sexual harassment, hateful conduct and grievances.”
In other words, all complaints would go to a single desk. The reviewer, former Supreme Court justice Louise Arbour, saw some obvious problems with that. Sexual misconduct, she wrote, “should not be conflated with or treated in the same manner as other forms of workplace related complaints.” Keeping these complaints internal also robs Canadian Forces members of an opportunity to go to tribunals or the courts. “The system proposed by McKinsey is of little use with respect to sexual harassment and misconduct.” Arbour blasted a variety of aspects of McKinsey’s report, revealing it as little more than a reshuffling off the deck.
Not dissuaded by the Phoenix pay system catastrophe, Ottawa still trying to make the system work: Except now it has McKinsey onboard. At a cost of nearly $28 million, McKinsey has “provided expertise required to streamline processes and standardize work at the Pay Centre, so as to increase efficiency and reduce processing times for pay transactions, as well as to build greater capacity through new ways of working.” In other words, Ottawa is paying a third-party to tell civil servants how to do their jobs because a third-party screwed up and failed to pay civil servants for doing their jobs. Got it.
Thus far, McKinsey’s work for Ottawa has been limited to about a half-dozen departments. But that looks set to change. The amount of reliance on McKinsey is rising and expanding.
McKinsey has been added to a list of “Artificial Intelligence suppliers,” meaning they qualify for a “streamlined procurement process.” Government agencies can award McKinsey up to a $9 million contract without taxes.
McKinsey looks set to leverage its questionable record managing France and Quebec’s public health strategies during the pandemic at the federal level in Canada.
In November, the Public Health Agency of Canada put out a solicitation, asking for “innovative technologies and tools to communicate information in new and unique ways that speak to Canadians. This includes, but is not limited to, creation of graphics, layout of reports, creation of other products such as infographics and videos.” It wasn’t looking for a design company: It was looking for a consultant. Seventh on the list of 15 was McKinsey.
The contract was specifically for Chief Public Health Officer Theresa Tam’s team. What’s shocking is that Tam had a dedicated communications shop, inside the Public Health Agency of Canada, that would normally be responsible for those kinds of projects. That team was shut down, supposedly for “operational reasons,” a public health source tells me. Word inside the department was that the team would be replaced with consultants.
Worth noting that Tam’s senior public health advisor is a McKinsey alumni. (After, incredibly, a year-long stint on Health Canada’s opioid response team.)
”Our processes are archaic, super cumbersome and unnecessary. The advice we give is usually good; our services less so,” the source told me. But just because people are frustrated “doesn’t mean they should be spending millions of dollars on external services when they’re already spending millions of dollars on the public service. They should be making the current system work, not outsourcing.”
And that’s exactly it. This is not a problem of too few hands: The core Canadian civil service has grown by more than 42,000 over the past decade. If they’re not up to the job, then maybe the hundreds of millions of dollars going towards consultants could surely be used within the civil service to level up skills, improve morale, and attract new talent.
But this is the insidious nature of public sector consulting: They want you hooked: On opioids, on their fabled ability to ‘disrupt,’ on them. And they want as little public discussion about that role as possible.
Below the paywall: A little preview of some upcoming Bug-eyed and Shamelesses, a brief note about a top-secret project I’m working on, and a recommendation to read the only edition of the Twitter Files worth caring about.
Keep reading with a 7-day free trial
Subscribe to Bug-eyed and Shameless to keep reading this post and get 7 days of free access to the full post archives.