Why have we been so ineffectual at fighting COVID? (Part 1)
Some thoughts as cases spike and frustration mounts
Thomas’s Note: Apologies for this being my first post in a while. Work and Parenting 1-year old responsibilities have been kicking my butt in the last few weeks. The good news is that I have multiple half-finished posts in the works, so hopefully, I will be getting those out to you soon!
As always, please give me feedback on what is on here! Please email me directly (thomasirwin13@gmail) or leave a comment at the end of the article. It really helps me develop my thoughts and see what folks are interested in!
A week and a half ago, Brittney and I took Zeke to a local park. Where we live in East LA, there are no parks within a short walk, but there is a small one about 10 minutes away that Zeke enjoys to run around on. It is not a perfect park: it is not cleaned super often, and we often have to grab trash from Zeke’s hands before he sticks it in his mouth. But we will gladly take that trade-off for the opportunity for some fresh air, light exercise, and most importantly, a happier one-year-old.
However, when we arrived at the park, we discovered the playground had been closed as a part of LA and California locking down in response to the surge COVID cases. This was, of course, very frustrating: there is very little evidence that COVID is transmitted by surfaces or by small children Zeke’s age. This playground is almost always empty on Sunday afternoons, and the few times we have, we have had Zeke run around on the grass and kept him off the playground to avoid even taking the risk of transmission from another kid.
As the week has gone on, I have discovered that I am not alone in my frustration: parents around the county have seemingly been pushed to the limit with the closing of playgrounds (which were not even opened here in LA until mid-fall). Meanwhile, LA has continued to let many activities continue: down the street, the outlet mall by our house has been packed with people looking for black Friday deals. But apparently, playgrounds are too dangerous.
I say this as someone who is not wholesale anti-lockdown: LA faces a genuine crisis right now. Over the course of 1 month, LA has gone from a relatively contained spread of COVID to an uncontrolled spread. Daily case counts briefly dipped below 1000 around November 1, while the past week, LA County has seen daily case counts hover around 10,000, higher than at any point in 2020. I have known people who have almost died of the virus here in LA and others who have had their lives significantly disrupted by having it. I have also heard many painful and emotional stories of healthcare workers who have to suffer through an almost indescribable pain, watching patient after patient dies at the hands of this virus. Simultaneously, people continue to go about life as usual, with gatherings and parties to celebrate the holidays.
My concern is that after 9 months of being in this pandemic, we keep making the same mistakes in our response. Why has our society failed so badly in the face of this virus? We have both faced an incredible death toll and an incredible amount of economic/social devastation. Vulnerable people die, many people are still out of the labor force, and many kids are falling behind terribly in school. Both sides of the aisle should agree that there are real problems here. What has gone so wrong?
COVID as an Invisible Externality
To think about this problem, I think it is helpful to zoom out not think about COVID specifically, or even pandemics as a category, and think about the category of problem that COVID poses to society. To me, the place to start is the concept of externalities, a term thrown about quite a bit by economists. The idea is simple: while many decisions we make have costs and benefits that only apply to us as individuals, some decisions affect our neighbors. While choosing to listen to music on my headphones affects only me, my decision to listen to music on a speaker affects my wife trying to work right next to me. She did not choose to listen to my noise, so my decision is posing costs on her she has no control over. Thus, it is an externality.
To use a more serious example, many Americans drink modest amounts of alcohol and do a fairly responsible job of managing the risk rationally regarding the risk to their health. Sipping bourbon while writing an evening pontification is a relatively risk-less activity. But there are many ways to consume alcohol that pose costs on others. Every year in America, 70,000 deaths are linked to alcohol. Much domestic violence can be traced to alcohol, as, of course, can many fatalities involving vehicle accidents. And this is to say nothing of the harder to measure how serious alcohol consumption can strain familial relationships, even if it doesn’t escalate to violence. Thus, we can see how alcohol's cost to society is clearly higher than the cost to the individual consumer of alcohol.
COVID is in many ways similar: most people can manage their risk in a way that makes sense to them as individuals, and how they weigh their individual risk of getting COVID. The problem is that getting COVID doesn’t just pose a risk to you; it poses a risk to everyone you contact: your family, your friends, workers whose places of employment you patronize. The problem is not irrationality (though this is clearly a problem for some, more on this later) but that most people naturally tend to take into consideration only themselves (and maybe their close family) when they make “rational” decisions. Young and healthy people face minimal risk from COVID, and so there is rationality to participating in some risky behavior Yascha Mounk points this out well in the Atlantic:
The dynamics of contagion in a pandemic do not work like that. If the case numbers in your part of the country are still relatively low—spoiler alert: They’re not—inviting five of your friends over for an indoor dinner party might not pose an unacceptably high risk for any one of you. As long as everyone is relatively young and nobody has serious preexisting conditions, the direct risk to your health is quite small.
But if everyone who isn’t at especially high risk held similar dinner parties, some percentage of these events would lead to additional infections. And because each newly infected person might spread the virus to others, everyone’s decision to hold a one-off dinner party would quickly lead to a significant spike in transmissions.
The dynamic here is reminiscent of classic collective-action problems. If you go to one dinner, you’ll likely be fine. But if everyone goes to one dinner, the virus will spread with such speed that your own chances of contracting COVID-19 will also rise precipitously.
Now small-scale externalities can often be solved by negotiation between the parties. If my wife is bothered by my playing music on a speaker while working, we can agree to both use headphones in our office. Neighbors can agree to a common set of rules for handling noise or parking. But in a large society, the ability to create these kinds of formal or informal agreements becomes unwieldy and difficult. COVID presents an especially tricky externality problem because the damage by spreading the disease is invisible and unseen at the point of transmission. When someone drives recklessly on the highway, other drivers have a chance to minimize risk by avoiding them and giving them negative feedback (living in LA has made me more liberal in the use of my car horn for this purpose, for better or worse). Asymptomatic COVID carriers present little opportunity for avoidance, or timely negative feedback, to change their behavior or mitigate the risk they pose.
Tools for Externalities
It is these dynamics that make it very difficult to manage a pandemic apart from government intervention. But there is a suite of options a government can choose to use in the face of a pandemic:
Command and Control: government authorities can step in and create legal rules that will mitigate the spread by “Locking down” places that could create risky behaviors. Closing restaurants, limiting travel, mandating masks, all of these are actions governments can (and have) taken to minimize risk taken by public members. These measures' strength is their force: they carry the legal system's punishment if violated. Of course, this power is also their dark side; people do not like being “commanded” or “controlled” and can react negatively to the use of this strategy (more on this later).
Create Incentives for Social Action: If command and control is a “stick,” incentives for good behavior is a “carrot:” governments can create rewards for citizens and businesses that mitigate risk and act in a pro-social way. This can include incentivizing people to stay home, businesses to go remote, and creating incentives for the mass productions of technology that can mitigate risk (vaccines, testing, contact tracing, etc.). Incentives are powerful because when well utilized, they can harness economic markets' power to achieve social goals in ways governments acting on their own could never do.
Clearly Communicate Information: Governments also have a key role to play in communicating clearly with the public, especially given how the “new” and “invisible” nature of COVID-19 make it hard for your average person to understand what is going on fully. Governments can provide information about risk and how to mitigate risk, which ideally helps people better protect themselves and not unnecessarily put others at risk.
As we have all seen, the problem is that having these tools at your disposal does not necessarily mean that you will use them well. There have been vigorous debates about the tradeoffs and decisions that governments have made with these tools in the face of lockdowns. I could spend the rest of this article on all the places where I think governments have failed to use these tools effectively, but I will highlight three that have been particularly bad from each toolbox:
1) Locking-Down Schools
The most egregious failure in our decision making about government ordered lockdowns is the placement of schools within the priorities of things being open and/or closed. The impact of virtual learning on students, especially the most vulnerable students, has been absolutely devastating. I could tell my own anecdotes about basketball players I coach, many of whom are deeply struggling in school and are being asked to do school remotely. At the same time, their parents work, are responsible for childcare for younger siblings, and live in loud, overcrowded apartments. But even in affluent areas, like where I grew up in northern Virginia, the data is sobering:
Stories from schools in south-central LA are even worse:
By the first week of October, several classes at the Communication and Technology School had more than 70% of their students failing. Many were not logging in. The dropout rate had noticeably increased from prior years, particularly among newcomer students. Technology problems — from weak wifi signals to broken iPads — plagued more than half the student body.
This is despite heroic efforts by the school counselor, Antonio Roque:
That evening, Roque came armed with a face mask and a plan. When the phone calls or pleas to meet him on campus don’t work, Roque visits the homes of the school’s students most at risk of dropping out, hoping to convince students like Kevin to return.
Kevin says he’s discouraged by his English-spoken classes, where he understands little and the Zoom format makes it harder than usual to follow along.
His grandmother, Dolores Aguilar, had brought Kevin onto campus earlier in the year in search of help, her patience with the boy wearing thin. She was frustrated that he spent most of his days watching TV and playing games on his phone before she took it away from him. He seemed unmotivated and distant. They’d been arguing for weeks.
None of Aguilar’s five adult children graduated from high school. She worries about her grandson when she’s working 12-hour days as a housekeeper. She sees her grandson struggling academically. She wants to help him with his schoolwork, “but I can’t,” she told the counselor.
“He says he doesn’t understand, he doesn’t know, and I don’t know either.”
We let kids (especially kids on the margins) potentially go to waste (to say nothing of the economic impact on parents), even though the best data being collected shows that kids are in little danger, and schools do not become superspreaders. Economist (and data-based parenting guru) Emily Oster has been painstakingly collecting data on COVID spread in schools over the course of the pandemic. Emily Oster pointed out in her piece in the Atlantic:
School-based data from other sources show similarly low rates. Texas reported 1,490 cases among students for the week ending on September 27, with 1,080,317 students estimated at school—a rate of about 0.14 percent. The staff rate was lower, about 0.10 percent.
These numbers are not zero, which for some people means the numbers are not good enough. But zero was never a realistic expectation. We know that children can get COVID-19, even if they do tend to have less serious cases. Even if there were no spread in schools, we’d see some cases, because students and teachers can contract the disease off campus. But the numbers are small—smaller than what many had forecasted.
Predictions about school openings hurting the broader community seem to have been overblown as well. In places such as Florida, preliminary data haven’t shown big community spikes as a result of school openings. Rates in Georgia have continued to decline over the past month. And although absence of evidence is not evidence of absence, I’ve read many stories about outbreaks at universities, and vanishingly few about outbreaks at the K–12 level.
My point is not that ALL places should have left schools open through the pandemic. Shutting schools down in the Spring when we were all clueless about what was going on was entirely defensible. Even over the summer, there was not much data on the spread in schools. And now, in places like LA County, the spread of COVID has gotten to such a point that re-opening schools would be challenging. Instead, my point is that governments should have recognized quicker that schools need to be at the top of the priority list for re-opening, not at the bottom. Whether you are a lockdown “hawk” or “dove” (and I understand the arguments for both), schools should have been higher in the priority lists than restaurants, bars, and the film industry. As much as I love sports, the idea that the Dodgers had had baseball games in LA before LAUSD returned to school is downright perverse.
Now there are reasons why this is the case. In many places, school districts operate as their own “governments,” so the elected and public health officials have decided to lockdown/re-open much of the other of the economy do not hold as much power over schools. And powerful lobbying from teachers and school staff has certainly played a role in keeping schools remote here in LA, and nationwide data shows a pretty convincing link between unions and schools re-opening or closing, which I think is worth scrutinizing (though correlation there does not mean causation). But I think these were all barriers that our leaders could have overcome once they recognized the harm being done.
2) Testing
Our other biggest failure in terms of incentives was the failure to create massive incentives to test far more people for COVID. We are 9 months into the pandemic, and even pre-surge, many people still cannot get a COVID test easily and quickly enough to help them mitigate the spread and align their behavior with risk. This is an incredible failure and one that should have been easy to fix.
Testing is a classic example of a problem where the externalities of COVID create a market failure: while the individual being tested derives benefit, but the significant benefit is also derived from every person they come into contact with each day at work, at the restaurant, etc. But it’s logistically impossible to create a way to transfer money from all those assorted people you interact with to the individual who needs a test, so the number of people who seek out getting tested each day will always be less than is optimal. This has a widespread impact on the economy: research by economist Raj Chetty has shown that a huge swath of the population is opting out of economic activity, not because of lockdowns, but because they are not confident in the safety of the various economic activity. A testing regime would have gone a long way to restoring that confidence and almost certainly pay for itself in getting Americans back to work safely. I will also say that the same argument also holds for contract tracing and supportive isolation, but I think the implementation of those things is a little more complex, which I will talk about in part 2.
There are many simple ways that governments can step to ensure that there is as much testing as is socially optimal, but the simplest one is to:
Heavily subsidize the cost of tests to the general population to incentive private companies to build out capacity and
Reduce regulatory hurdles to getting tests approved and to market.
This is the playbook that we took to vaccines, and there is no reason we could not have taken the same approach to COVID testing. But we did not take this approach. We now know that Moderna created an effective vaccine by January 13th, only two days after the genetic sequence was made public. The Pfizer Vaccine was created shortly thereafter. On the flip side, the FDA and CDC did not have an effective COVID test until the end of February, and regulations kept private labs (who had bootstrapped an effective test) from being allowed to perform any tests in the meantime. This but the US was in a terrible position coming into the March surge to really understand where the hot spots were, which ultimately was one reason we lockdown so strictly even in places where no surge was happening.
Of course, we have had plenty of time to make up ground after our initial screw up. In April, an inter-disciplinary, bi-partisan task-force spearheaded by Danielle Allen of Harvard recommend we scale up to doing 20 million COVID tests per day over the summer as one of the pre-condition to maximizing safe economic activity over the summer. But their efforts to work with congress largely stalled out partially because of partisan disagreement on COVID relief, and partially because President Trump indicated he did not want more testing to be a priority. Just because something is politically hard does not mean it is not politically possible, and the failure to overcome that barrier has continued to hamper us until today.
3) Manipulative Messaging
Part of being a good teacher (or coach, or parent) is teaching your students to make good decisions for themselves. Different teachers will have different strategies to do this well (some lean towards freedom, others towards safety), but all good coaches recognize that teaching people to make decisions is crucial. Too often during the pandemic, it seems our officials have not learned this lesson. Instead of explaining that COVID is very risky and teaches the principles that can lead to risk-reduction, officials (and media members who take the lead of officials) have tried to manipulate people’s behavior in a way that has had tremendous negative impacts.
Three specific unhelpful behaviors that have often been used:
Relying on overly simplistic messages that are unrealistic, both in a “stay closed” direction and in an “open up” direction. Simple messages are good, but when they are too simple, they invite backlash by insulting the intelligence of the person listening. (Really, do will I really be able to go to an Easter service safely? Am I really going to stay in my house until a vaccine?)
We have relied extensively on moral grandstanding in response to the behavior we disapprove of, a move that simultaneously degrades the person whose behavior we disprove while uplifting our own moral righteousness. Unsurprisingly, usually leads to the backlash, as people want to do the opposite of what you shame them for:
In some cases, public officials have even told outright lies that are not true but will try to influence behavior in the direction you want, whether in a “stay home” direction or an “open up” direction.
The issue of masking illustrates all three of these problems. At the beginning of the pandemic, many officials lied or played down the masks' effectiveness (using the logic that made little real-world sense). They then pivoted to simple messages on masking (always mask!) in a way that could have benefitted from more nuance (do I really need a mask when alone running?) Then the mask culture wars started, with pro-mask and anti-mask people shaming each other publically, usually along partisan lines. Ultimately, one of our best, least intrusive tools in pandemic fighting has been severely hampered by messaging failures. As Matt Yglesias writes:
We need to get to a point where a larger share of the public has confidence in the experts and the public health authorities. And the fact that the authorities seem to have been deliberately misleading people about masks — and doing so because they lacked the creativity to think of the cloth masks as an alternate solution — is not great for building anyone’s confidence
Thankfully, there are increasingly media members and some public officials who have recognized how badly this messaging arm is going and are pivoting to one centered on simple but effective rules that can reduce harm. Oregon, a state which has done relatively well in combatting COVID, actually did this back in late Spring. They tried to message around 4 simple COVID rules:
Smaller Gatherings are safer than bigger.
Farther Social Distancing is safer than closer.
Outside gatherings are safer than inside.
Masks are safer than no masks.
This messaging treats people like adults who will ultimately make their own choices, which I firmly believe works better in the long run. This is not to say that harm-reduction is the only way governments should message; I think there are times when drawing black and white moral lines is appropriate. But moral messaging is a tool, and if used too much, people will not listen.
The Question at the Heart
The Meta Problem
Of course, I think it is worth taking a step back and asking the more important question: why have we failed on so many different fronts? Many micro-lessons are fighting viral pandemics that we can learn from these mistakes, but chances are, the next existential threat we will face will not be a viral pandemic. How can we generalize these lessons in a way that actually leaves us more resilient as a society? That is the question I will try to tackle in part 2.