McKinsey & Company’s ethical problems are baked-in to the organization

I spent about a year working at McKinsey & Company as a Principal Data Engineer. In that role, I was responsible for training and technical implementation for clients who were building a data science capability. Given that I was there for only a year, you can infer that this job was not a good fit for me. One of the reasons I quit — but not the only reason — was that I felt I could not work for a company that was so unapologetic about its role in fueling the opioid crisis in the United States. This was personal for me. I happen to have chronic pain, and I’ve met plenty of other chronic pain sufferers who have become addicted to opiates. Fortunately, I avoided that particular problem, but this was mostly good luck. If things had gone slightly differently for me, I could easily have become an addict myself. As more information kept coming out (primarily through the New York Times) about McKinsey’s actions, this bothered me more and more.

While this news about McKinsey was coming out, I was very perplexed. McKinsey has a strong set of values that are hammered home all the time. I never got the sense that anyone was being hypocritical about them. The people I worked with were highly principled, and I have a hard time imagining any of them doing something so despicable. Nonetheless, it happened. How?

Despite leaving the firm after only a year, I have no bad feelings about McKinsey at all— in many ways, it was the best job I’ve ever had. I worked with some brilliant people, and everyone I worked with was perfectly professional and ethical. Opportunities for professional development are world-class; in fact, I tell people that I got at least five years of experience in my one year there. Furthermore, they take their obligations to their clients very seriously, and as far as I could see, they truly live their values. But all that having been said, it’s fair to bear in mind is that I only worked there for a year, and one year is not nearly long enough to really understand all the complicated dynamics of a sprawling organization like McKinsey. So you should read this note with that fact in mind.

I should also mention that McKinsey is famously secretive (for reasons I’ll explain). So I won’t be saying anything here that’s not common knowledge about how McKinsey & Company operates, and I certainly won’t be saying anything about the clients I’ve worked with.

Organizational facts

So how does a giant company like McKinsey manage to operate effectively at such a huge scale? They do a few things, and they do them very well. First, McKinsey deploys a huge number of small teams and gives them a great deal of autonomy. Each team has a lot of leeway to tailor its work to the specific client. This is obviously necessary because centralized oversight is just impossible at that scale.

Decentralizing causes a tough problem, however. It becomes difficult to maintain standards among a huge number of autonomous teams. So the second thing McKinsey does is to establish a “McKinsey way” of doing practically everything. There’s a McKinsey way to perform an analysis, a McKinsey way to structure a presentation, a McKinsey way to write an email, and so on. So the decentralization is managed by establishing norms that are internalized by each team. That way, they get consistency without a burdensome level of oversight.

The third tactic that McKinsey deploys in order to operate at such a scale is that they work for competitors in the same industry. It would be normal in a small consulting firm for a client to demand that the firm not work for its direct competitors. But there are literally not enough industries in the world to support a giant like McKinsey if they aren’t able to work with competitors in the same industry. The only way that McKinsey can make this acceptable to its clients is to maintain strict confidentiality about each and every client engagement. Without confidentiality, there would be no McKinsey as we know it because there wouldn’t be enough clients to support the firm.

McKinsey & Company takes client confidentiality seriously. You cannot even reveal which client you’re working with to other McKinsey consultants. You may only refer to your client by a code name. If you’re in McKinsey’s office, and you’re discussing your current client with someone else, you must use the code name — even if the other person is also working with the same client. It’s well known that McKinsey employees get into the habit of referring to the company as “the firm.” This is deliberate, so that anyone overhearing a conversation won’t know that it’s McKinsey they’re talking about. A direct result of this culture of secrecy is that teams have almost no visibility into what other teams are doing.

Finally, responsibility is not only pushed down to individual teams for execution on client projects, but also for ethical decisions. This is the feature of McKinsey that is especially interesting to me. As a McKinsey consultant, you are never required to work for a client with whom you have ethical disagreements. So if you don’t want to work for EvilCorp, you don’t have to. There is no penalty for refusing to work with a specific client if you’re doing so on moral grounds.

Ethical breaches fly under the radar

What is not well-understood is that these organizational decisions have unintended consequences, and those consequences create an environment in which ethical lapses should be expected. In my opinion, for reasons I’ll explain shortly, it is too simplistic to think that McKinsey is evil. In my experience, the people who work there are like anyone else — no better, no worse. But they are operating in an environment where ethical problems find fertile ground; and again, this is not because McKinsey is evil but because of how it is organized.

Let’s consider a fictional example. I’ll choose something outrageous just so it’s clear I’m not talking about a specific client. Suppose that EvilCorp hires McKinsey to improve its production of space lasers. Let’s stipulate that EvilCorp is going to do this for horrible reasons.

Given the terrible nature of EvilCorp’s project, let’s suppose that the vast majority of employees decide that they are not willing to work on it. As I said, they can refuse to do this work without incurring any penalty. McKinsey can afford to have this humane and sensible policy, however, because there are so many highly qualified people it can draw on to staff a client engagement. Among the huge number of consultants in the firm’s pool, they can surely find someone to work on the EvilCorp project.

Now let’s suppose that when the team is deployed to EvilCorp’s headquarters, it turns out that their project is even more nefarious than anyone had thought. As I mentioned above, it’s up to the individual team members to raise any ethical concerns in the course of a client engagement. They can certainly do so; and I take McKinsey’s leadership at their word that any such concern will be taken seriously. The trouble is, it’s unlikely that any such ethical concern will be raised because the people who were sent to EvilCorp were assigned to that project precisely because they have no ethical qualms about working for EvilCorp in the first place! To put the point another way, the people who were most likely to raise ethical concerns have been eliminated from the project before it even started because they would have refused to join in the first place. So it should not come as a surprise that when EvilCorp’s project slides deeper into unethical territory, nobody is particularly bothered.

Now let’s think about McKinsey & Company’s recent troubles with their efforts to market opiates. Let’s just stipulate that it’s obviously unethical to do the things they did to “turbocharge” opiate sales. Let’s put aside the specific things they recommended (e.g. paying CVS for every overdose) and just think about the process of marketing opiates in the first place. In my opinion, opiates have a perfectly legitimate set of uses for medical purposes (e.g. certain kinds of severe post-operative pain), but there’s no justifiable reason to market them as if they were some kind of consumer product. The reason to market a product is to raise awareness of it among the customers who are most likely to want to purchase it. But it’s not as if doctors aren’t aware that opiates exist, or that there’s any need to “get the word out” about these drugs through a marketing campaign!

If, like a lot of people, you would have been uncomfortable with the idea of marketing opiates at all, then you wouldn’t be on McKinsey’s team for that client in the first place. Conversely, the people who were on that project didn’t think there was a problem with trying to use consumer marketing techniques to increase the sales highly addictive chemicals. So it stands to reason that those people would be more likely to push these specific ethical boundaries, and they’d be the least likely to raise ethical concerns to McKinsey’s leadership as the project slides deeper into unethical territory.

Without ethical concerns being raised by the team members, one might hope that there would be some top-down oversight that could have sent up a red flag about this work. But as I’ve mentioned, McKinsey’s enormous scale prevents this. No reasonable number of people can possibly provide such a level of ethical oversight. This is presumably why McKinsey depends on the teams themselves to push any concerns up to the leadership.

One might also hope that even if a particular team did not raise any concerns about a project, other people outside the team might. Whistle-blowers in other organizations are often bystanders who are disturbed by what they see. But obviously, this is extremely unlikely to happen at a place like McKinsey, where confidentiality is taken so seriously. People generally have no idea what’s happening on other teams, or even who the other clients are. They certainly are not allowed to learn about specific details of other engagements. So the probability that someone would just happen to learn about unethical behavior is tiny. And for someone outside the team to raise ethical concerns about another team’s work is tantamount to admitting having breached client confidentiality.

No wonder that ethical lapses can fester and go unnoticed for a long time at McKinsey. The people who are staffed on a client engagement are the ones who are least likely to raise any issues. In the absence of bottom-up direction, there’s precious little top-down oversight that could catch ethical problems when they occur. Thanks to the firm’s culture of secrecy, people outside the team aren’t in a position to raise any red flags. In short, there is no plausible scenario where ethical lapses would be detected and stopped in time.

Speculations and suggestions

Let’s consider McKinsey’s stated values regarding professional standards:

  • put client interests ahead of the firm’s
  • maintain high standards and conditions for client service
  • observe high ethical standards
  • preserve client confidences
  • maintain an independent perspective
  • manage client and firm resources cost-effectively

Other than the blanket catch-all “observe high ethical standards” (which is so nebulous as to be meaningless), what strikes me about the firm’s values is that they are all about how to work with clients. But the glaring omission from McKinsey’s values is anything about which clients they will work with. Their ethical standards kick in as soon as they have taken on a client, but there is nothing about refusing to work with clients who are doing anything unethical.

My speculation is that this omission may originate from the legal profession. In the law, everyone is entitled to a competent legal defense, regardless of whether you’ve done something illegal or unethical. Quite properly, the right to a legal defense needs to be absolutely universal. The single-minded focus on clients is also reminiscent of how a lawyer might think about her work. It is not the job of a defense lawyer, for example, to take into consideration whether society would be better off with their client locked up. It is the lawyer’s role to look out for the client, period.

These values seem to have crept in to McKinsey, but they aren’t appropriate for a management consulting firm. Not everyone has the right to a management consultant, and it is entirely proper for a consultant to refuse to work for a client who is engaged in unethical practices. And for a firm as large and powerful as McKinsey, it is inappropriate to focus entirely on the client. Given the vast number of powerful clients (including government and industry), McKinsey’s stakeholders comprise huge numbers of people other than their clients. To return to their work marketing opiates, among their stakeholders should be included all the people who are users or potential users of opiate medications, the insurance companies who pay for their medical services, and the communities those people are part of. But if success if measured only on the basis of client impact, those stakeholders are left out, to disastrous effect.

Despite my misgivings about the firm’s behavior, I loved my time at McKinsey & Company. Given their resources, expertise, and brilliant, hard-working people, they could do enormous good. But clearly, the firm needs to change in order to live up to its potential.

I’m no expert in such issues, but from what I saw, I think there are a few changes that strike me as obvious:

  • Establish top-down review of 100% of new client engagements. There needs to be a “go/no-go” decision based on the ethics of the client and the specific work. This needs to be done by industry experts and generalists who have no professional or financial incentive to give a pass to dubious client engagements.
  • Expand the set of stakeholders to include communities that are affected by their work. Putting the clients’ needs ahead of the firm is a good value to have. But putting the clients’ needs ahead of the communities they impact is not. Whether a client engagement is to be judged as a success should depend on its measurable impact on a much broader range of stakeholders.
  • Eliminate the right to opt out of client engagements for ethical reasons. In my opinion, the fact that people are pre-selected for their acquiescence to a client’s ethical standards (or lack of such standards) is the core problem. If McKinsey leadership were forced to justify client engagements to people who might be skeptical, then ethical standards might improve.

Machine learning technical lead, former philosophy professor, drinker of too much coffee.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store