On January 1st, I wrote that the slow vaccine rollout was a national scandal. California’s overly complex priority system did not seem promising compared to having a more straightforward system. I argued that this was an example of “leveling-down” where trying to make things more equal would make everyone worse off, rather than making everyone better off:
In the case of vaccine distribution, it seems that the ethics of trying to create a complex system for pinpointing who exactly should get the vaccine first (and to make sure no one “jumps the line”) are really just delaying the date at which everyone will eventually get the vaccine. It seems that we would be better off if we accepted a simpler system for determining proximate fairness of the vaccine distribution queue (instead of perfect fairness) and focused a lot more on the speed with which we will get through the queue. It seems that the vast majority of people would end up getting vaccinated at a much earlier date if we had gone with this approach.
Two weeks later, California has only distributed 26% of all its vaccine doses, ranking 45th out of 50 states in the US. This is almost 1/3 as efficient as the state of West Virginia, which has administered 73% of its vaccine doses, including vaccinating every single long-term care facility resident. The Federal Government has indicated that it will start to reward states that administer the vaccine faster by giving more doses to states that have used up their previous allocation. Today the LA Times Board wrote an editorial that criticized California’s process:
Azar said a major barrier to a quick U.S. rollout of the vaccine has been overly prescriptive rules by states that are focusing only on front-line healthcare workers. California has crafted a complicated system for vaccine distribution that puts entire groups of workers ahead of some groups at higher risk of death. On Tuesday, California Health and Human Services Secretary Mark Ghaly acknowledged that the focus on equity in the guidelines has led to delays in the vaccine rollout.
We applaud the goal of equity, but it may be time for the state to throw out its playbook and focus on getting as many people protected as possible.
To be fair to California, the logistics of giving vaccines to healthcare workers at overwhelmed hospitals is not easy. But moving at one-third the pace of leading states is simply not acceptable. Doses sitting for weeks in a freezer somewhere help no one even if they find the right arm eventually. Only today did the state of California acknowledge a problem: the state pivoted and opened up vaccination queues to those 65 and older:
It’s also worth noting that despite California’s efforts to enforce fairness, many news outlets have reported that California’s system still has many easily exploitable loopholes.
Markets and Information
F.A. Hayek, the famous economist and intellectual, is well known for pointing out that the main purpose of markets within economic life is to convey information in the form of prices. In his paper The Use of Knowledge in Society, Hayek argued that by creating a market exchange system for a good or service, you create the mechanism of prices, which helps society make use of information that is too widely dispersed for any one person ever to know. For example, I do not live in Argentina, so it is unlikely that I would know if there is a catastrophe that wipes out half of its cattle herd. But I will notice the price of beef going up when I go to the grocery store, which will likely cause me to eat chicken or vegetables instead. My neighbor, who is on a Paleo-diet and only eats beef, may still choose to pay the higher price. Thus, I will help “solve” the problem of a worldwide beef shortage and indirectly help my neighbor, not because I am a nice guy, but because I am simply responding to prices at the grocery store!
Hayek wrote that this market process leads to the illusion of a well-thought-out order in an economy, when in fact, most people are living simple lives where they don’t concern themselves at all about the economy:
The whole acts as one market, not because any of its members survey the whole field, but because their limited individual fields of vision sufficiently overlap so that through many intermediaries the relevant information is communicated to all.
This gets at something I referenced in my post about experts: part of the reasoning expertise can be counterproductive is because it gives the illusion that complex problems can be solved if you think about them for long enough. When you look at “smart, specialist academicians” as a category, one pattern you notice is the tendency to respond to complexity by engaging in lengthy analytical discourse. I feel fairly confident in saying this because I tend to do this. I was raised in a family that tended to do this (if you are reading this, Mom and Dad, feel free to dispute this characterization in the comments). This tendency even extended to Irwin's family decisions about where to go out to eat: we would discuss at length, get everyone’s input, and see if we could weigh all relevant factors to see if we could come to a decision. One hour later, we would all pile in the car and go out to eat, having endured a very fair but inefficient process.
Of course, when my wife first came into the family, she thought this was crazy! In her family growing up, one of her parents would say, “Let’s go to Cracker Barrel,” and the family would immediately get in the car and go to Cracker Barrel (Brittney, if you are reading this, feel free to dispute in the comments). To them, the fairness and thoroughness you would gain by analyzing and creating complexity was just not worth the tradeoff in the speed of decision-making. And frankly, I have come to think her family’s intuition when it comes to restaurants works better for picking a restaurant. That does not mean that it is necessarily the best approach for every family decision (like what school a child will attend), but it works well for going out to eat.
Of course, my experience with managers in the business world is that when it comes to analysis, most of them are much closer to the “Hojo” side of the spectrum than the “Irwin” side. Speed is critical in business. One of the core principles of finance is the Time Value of Money, a framework for thinking through how much value there is in whether an investment takes a short or a long time to pay off. In the face of complexity, the best business strategy is often just to try something, knowing that you will fail, but you will learn new information and quickly muddle your way to a better equilibrium. For this reason, many Hojos have gone on to successful careers in getting stuff done, while Irwins are busy writing pontifications.
Demand for Vaccines
I fear that the vaccine distribution has suffered under a cognitive error that smart academicians (and the Irwin family) tend to make: sacrificing speed in favor of a more complex response. The complex response, confronted with market realities, fell apart. The academic reasoning that it would be more equitable for frontline workers (especially frontline workers of color) to be vaccinated first failed to consider that many frontline workers of color do not want to take the vaccine first.
For some frontline workers, this is for historical reasons: at various points in American history, medical researchers have used people of color in unethical medical experiments. Likewise, historically medical trial participants have skewed heavily towards white men, which meant their findings were not representative of other groups. Others have already been infected and do not want to take vaccines from others who need them more. Many do eventually want the vaccine; they just do not want to be first to get it.
Given that the vaccine trials showed robust efficacy in all groups, some of this could be solved by more concerted efforts at education and building trust But considering that many older residents desperately want the vaccine, including many older adults who live in high-risk multi-generational families, the case for waiting on an education campaign to change minds was also dubious.
Now, this is not to say that a free-market for vaccines would have been the best solution. While auctioning off vaccines would have clearly demonstrated who wanted it the most (and created more of an incentive for fast delivery), it would have been ethically and politically disastrous for the wealthiest to get a vaccine when the poor are suffering more. However, I think it was unwise for the vaccine distribution plan not to anticipate the problems they were going to face by working entirely outside the market system. The system would work better if there were some mechanism to create an information feedback loop like the one which would normally be created by the market supply and demand.
It seems clear now that 1) opening up vaccinations to a wider pool of people (like California is now doing) and 2) having them sign up for a wait list would have been very much preferable to our current regime. Once you have a wait list, you can create a sorting function to prioritize those who you think are most in need, and then start calling people for an appointment. If someone does not respond, they fall to the back of the line (but can remain on the list), and you go through your list as fast as possible. We have all been to restaurants on Mother’s Day; creating this system is not rocket science!
Likewise, we could use economist Richard Thaler’s idea: hold a very limited charity auction for a certain amount of vaccines a month. After all, we can all presume that the rich and powerful are trying their darndest to cut in line. Rather than have them get it for free, why not raise several billion dollars by auctioning off one thousand doses for a million dollars each? If we had held the auction on December 11th when the vaccine was approved, we could have had plenty of time to use the money to build a distribution infrastructure and workforce while the federal government fought over the stimulus bill. We could have also asked the wealthy and celebrities to take the shot on a televised program to build trust in the vaccine’s safety.
Now you might find these proposals half-baked. That is fine; they are just ideas. But it is still essential to consider the problem of information in the absence of a traditional market. California failed to keep this in mind when rolling out the vaccine.
Talking about Equity is not a Quick Fix
In recent years, the idea of making “equity” the north star of left-of-center policymaking over and above “equality” has gained a significant amount of momentum. My shorthand explanation is that in these debates, equality often means “equal treatment” or the classical liberal idea of equality of opportunity. In contrast “equity” often means “equality of outcomes,” where we attempt to rectify historic disadvantages to the point where outcomes would be equal. This concept has become so popular that two days before the 2020 election, Kamala Harris was tweeting about it:
As this concept has gained momentum, I have had two concerns about its prevalence:
1) It seems politically unwise to frame issues like education in the much more politically divisive terms of “equity” when one could just as easily argue that we are nowhere close to achieving the less politically divisive goal (equality of opportunity) in most areas of America (a case that Raj Chetty has been making convincingly for years). The “swing voters” or “swing policymakers” in American politics is going to have a lot more sympathy for the idea of “equality of opportunity” than for “equity,” as the reaction to the Kamala Harris video demonstrates:
2) But I also worry that this is a rhetorical move designed to avoid having to confront the hard trade-offs that come with doing policy analysis. While I am personally one hundred percent convinced that we as a society should try to eliminate racial disparities, I do not necessarily think the goal of equal outcomes should come at the expense of all other considerations (for instance, eliminating all free economic markets). To illustrate this, I will point back to a series of blog posts I wrote, trying to outline my own intellectual grappling with systemic racism. I briefly touched on Ibrim X Kendi’s ideas (if you are not familiar, Kendi has become quite an influential figure in race conversations). His book held out as his primary policy initiative that the federal government would create a department that would regulate every institution in American life to ensure equal outcomes (“equity”) along racial lines. This is what I wrote about Kendi’s idea:
I do not have space to engage this proposal fully, but I will say: I have a great deal of skepticism that this will work on a society-wide level. Redesigning systems with only outcomes in mind all have huge trade-offs. Liberal democracies have long held the belief that fairness in a process is vitally important. The US constitution guarantees individual rights and freedoms even though they often lead to bad outcomes in the short run because they believe that these freedoms create processes vital to the long-running health of society. There is a long history of top-down designed social systems being built in the name of better outcomes, only to produce worse results because they ended up destroying locally organized, well-functioning processes. Experimentation has shown that even things as simple as a food distribution at food banks work better when using a decentralized approach, rather than a top-down one. Recall that policy as simple as banning the box created unintended consequences. We can thus expect that Kendi's vision will have vast amounts of unpredictable harmful effects, often falling on people of color.
Mandating equal outcomes requires overturning many existing social processes. That level of disruption creates a large risk that you will end up with a situation where everyone is worse off than before, including the most marginalized in society. I worried that the semantic push for the importance of equity, while it may prove illuminating in some policy areas, would obscure the truth in others.
This does not mean that all efforts to eliminate disparate outcomes will suffer from the same problems of leveling down. For instance, I generally support a policy of equity for education, where more resources go into educating low-income kids. Investing in these young students has the potential to make everyone better off, a finding which has been advocated by both right-wing economists (like James Heckman’s work on the benefit of good pre-schools) to more left-wing economists (like Raj Cherry’s work on “Lost Einsteins”). Ensuring that every member of society can work at full productive capacity pays dividends for all of society for years to come in inclusive economic growth. You can likewise make a case for more equity in greenspace funding, given how many communities (like mine in East LA) are at a severe park deficit (not fun when you have a 1-year-old in a pandemic!). So you can infer that “Equal” funding going forward will not really make a difference for many years, thus the desire for equity.
But I think this actually demonstrates the importance of examining each policy on its own merits and trade-offs, rather than seeing equity as a kind of “quick fix” to all policy problems. In the case of the vaccine rollout, the centralized and complex planning required to ensure equal outcomes really does seem to have run roughshod over established social processes that otherwise could have been leveraged for an effective rollout. I don’t think it’s an accident that small states, with a more decentralized rollout, have achieved better outcomes by ensuring that local pharmacies could work with local nursing homes. Decentralization actually appears to be a strength of small states, as it allows the distribution to resemble a market environment more closely.
In December, I found this neat website that tried to use data to plot how fast the US could get to Herd immunity and how much more we would have to endure to get there. At the time, they projected that ~500,000 Americans would die of COVID by the end of 2021. One month later, they have revised their projections upward to 600,000 people. We have become numb to how terrible this pandemic is, but it is truly staggering that one bad month likely means that another 100,000 families will lose a loved one before this is all over. And many of those families will be my neighbors here in East LA, despite the good intentions of California’s policy.
I closed that post on systemic racism with the following line, which is probably as close to a thesis as this blog has:
I am left with the rather dull idea that the best solution to systemic racism in America is pursuing the slow work of piecemeal social reform. I do believe that by analyzing places in our society where there are unequal outcomes, vigorously examining and debating the processes that are core to those systems, and implementing incremental reform to those processes is the best way to combat racial disparities
I very much believe that if we are going to get optimal outcomes on these social problems, we will have to really embrace the hard work of incremental analysis, debate, and reform. I don’t think there is an alternative that will work.