How to lose trust

I read a good column by Kelsey Piper about trust. It's about COVID and the way people in the media made major mistakes, never acknowledged that, and probably never will. The conclusion is pretty depressing but a lot of it resonated with me.

I was about halfway through a Master of Public Health when COVID first reached my area. Being in a school of public health at the onset of a pandemic was a wild experience, mainly a terrible one, but with some really important lessons. These lessons were all about trust.

1. No worse than the flu

Kelsey's column opens in February 2020, when the U.S. media was calling for people to stop panicking, saying COVID was no worse than the flu. This happened in school, too, and I have a vivid memory of public health professors I trusted laughing about how the public was obviously overreacting. I believed them.

These were not journalists covering a topic they don't have expertise in, but epidemiologists with doctoral degrees and successful research programs! Some of them were getting major public platforms for the first time! And to be clear, many epidemiologists were not downplaying the severity of COVID in January and February and were already working incredibly hard to understand it better. I just remember being blown away by how people that had real expertise, who worked at a great public health school, could just be so wrong about a public health issue the minute their judgment was tested in a real world scenario.¹

When the people you thought were experts fail so badly, it is hard not to lose trust in them. In fact, it's hard not to lose trust in everyone with those kinds of credentials, and try to pay more attention to their actual track records.

2. Changing your mind

Kelsey's article notes that the guidance on masks included, at various points, "masks don't work that well," "masks should be reserved for first responders," and "masks are crucial." This triggered a memory of something I couldn't quite figure out as an MPH student. When a decision changes in response to new data, how are you supposed to tell the public?

At the time, I struggled with the conflict between needing to suggest some action, even if it may turn out to be wrong, and needing to be consistent. I still think there are some very difficult tradeoffs.

One thing that is clear now, but I didn't fully understand then, is that some people will be right from the start. Any process that helps find these people is valuable—at all levels, as an info consumer, in an organization, all the way up to the broader media environment. In an ideal world, the experts most committed to the wrong idea would lose their platform if consensus is reached that they were wrong, and our institutions would help us hear more from the people who are right. Actively bringing in new voices based on their track record seems like a strong way for an institution to regain some trust. But there seem to be, uh, some small problems with the way our institutions currently select experts to spotlight.

Another thing to emphasize is how bad the initial strategy was. In the early stages of a public health emergency, with so much unknown, it is just silly to claim "masks don't work." That would be a reason to skip masks forever, not just right now. Why not say "prevalence is not currently high enough to warrant masks" or "right now, your priority is something else" instead of jumping to such a strong conclusion?

3. Switching roles

Before the pandemic, I saw very little relationship between the School of Public Health and the university. That changed very quickly after some financial projections came out showing a large revenue shortfall.

I remember when the dean of the school, Sandro Galea, was describing the hybrid model that BU had chosen for the Fall 2020 semester. Some background: Dean Galea is a talented public speaker, strategic thinker, and writer. He is extremely charismatic, and has this way of inspiring students that I've never seen from anyone else. Whenever he spoke or wrote it seemed clear that he'd wrestled with the ideas himself, that he'd put a ton of work into understanding them and making them his own. So it was a weird experience when we had the meeting about the Fall plan, in which he became The Voice of the University.

Everything he said (or rather, The Voice) had the same characteristic eloquence and clarity but now I started interpreting it as "we have decided it is safe to go back to school, and the reason we've decided that is because it's kind of our whole business model, but don't worry! It really will be safe, haha, and so far we've been great at predicting this stuff." (Later, my predictions about this ended up being wrong²).

How does trust actually work?

Trust is fragile. People misplace it often and retract it swiftly. As Kelsey writes, "it’s much easier to lose trust than to gain it."

There are some ways to lose trust actively or even on purpose.³ The problem is that there are also a million unintentional ways to lose trust. For people whose job is to understand and communicate data, the most common ones all have to do with being wrong. And people who do research and data analysis understand that we are wrong a lot.

So being trusted is not the natural state of things, it is extremely fragile, and it is a huge privilege. It is not possible to earn trust solely through credentials, or speaking in the right language, or "educating the public,"⁴ or demonstrating that you care about a problem, although these things can certainly get you started. To continue to build trust, you must work patiently, nurturing it over time, and react to any threats quickly but also very carefully. Success depends on repeatedly taking on some risks or responsibilities, and then repeatedly being right.⁵

Honorable mention: making a profit during a pandemic

In June 2020, Boston University reported an expected $243m revenue shortfall in FY2021 due to disruptions from the pandemic. They announced temporary cost-saving measures like layoffs, salary freezes and a pause on university contributions to retirement accounts, which would at least save $168m. Sure, when you're expecting a shortfall it makes sense to cut costs. The problem is they went on to earn a $144m operating gain that year, not to mention a record-breaking year of donations. Net assets were over $5b.

When I read about this, I reacted pretty strongly. At the time, my model of university finances was 1) they charge exponentially increasing tuition rates to students; 2) they do not pay professors⁶; 3) professors bring them money in the form of indirect costs on research grants; and 4) they get lots of donations.

"I'm sorry, do you have some sort of furnace where you take all this money and set it all on fire?"

Of course, that is pretty far off. Grant money only covers the minority of total salary costs and some university money is spent on research-related things. Universities sure seem like inefficient organizations, but I have no knowledge to contribute to that conversation, and the fact is that a lot of real people work for them and that costs a lot of money. Also, the cost of college is actually decreasing, something I only learned about a couple months ago.

When things are going well, everyone will still complain that college is too expensive but the university leadership is pretty safe to just do its job in a competent fashion, and it will be fine. When things are not going well, they get scrutinized not in their capacity as financial stewards, but as leaders of an academic community. Here, the resounding message was "yes I know your life is tough right now, and I know we have a large pile of cash, but the pile can't get smaller, only bigger, so we need to take some of your compensation away too."

Even worse, the headline is about 243 million dollars! Look how much that is! We must do something! while the reality is that it's a few percent of net assets. The optics are just terrible. Then the financial reports came out, and your forecast is so far off, in a way that perfectly fit your needs at the time you made it, and after all of that you come up with an exciting two percent raise. These leaders did their job well, from the perspective of the finances, but I think they failed to change their job when it mattered most.