 I think that it's pretty clear that there won't be enough vaccine for everybody initially. It just takes a long time to get 300 million doses of vaccine made. And if we're looking at November and as a potential date for a new vaccine, then people start thinking about, well, what are we going to do when there's not enough? And that's where this proposal and others might fit in. There's four phases to their distribution suggestions. First, they would suggest giving high-risk workers in health care facilities and first responders access. And then people of all ages who have multiple comorbid conditions at higher risk for illness as well as older adult who live in crowded settings like nursing homes. The next thing they suggest is helping critical care workers, that is workers who are in jobs that are both really important for society and at high risk of exposure, teachers and school staff, people of all ages who have these conditions that put them at higher risk as well as older adults who didn't already get the vaccine. And people in homeless shelters and prisons and other detention centers. And then in phase three, they would give the vaccine to young adults and children as well as people who are at industries essential to society that haven't already received it. And finally, everyone in the U.S. would get a chance to have the vaccine if they haven't already gotten it. Then within each group, they will deploy a social deprivation index, which comes from the Center for Disease Control and looks at a variety of different kinds of deprivation and says, if you have a higher deprivation index or higher vulnerability index, then within that group we'll get it first. If we're talking about distributing a vaccine in the U.S., I do have concerns about the National Academy's proposal. They say that their first priority is to maximize social benefit by reducing death and disability and transmission rates. So at first they'll help try to prevent death and disability and then focus on constraining transmission. But they don't explain exactly what it means to reduce death and disability. You kind of have to explain how to measure that. And instead of really trying to do that, they shift to talking about different risk groups. So they define people in different categories like children and adults by different kinds of risks. So they'll talk about the risk of acquiring infection, the risk of getting really sick if you get it or dying, the risk of negative social impact if individuals have a high priority job, the risk of transmitting disease to others. And then they kind of rank everybody on these different categories and tell us who should get the vaccine first. They do provide some kind of value judgments about their decision, but they don't kind of explain how those connect to the risk evaluation they do in a way that would kind of justify what they're doing. I think what matters more than rationing, though, is changing pharmaceutical incentives and rules for innovation. So about 80% of the manufacturing is done on generics, which means there's a huge manufacturing capacity in the generic side of things, whereas only 20% is on patented medicines. If we let companies keep patents on those products, then they can charge what they want for those medicines, or maybe they have advanced purchase agreements with the government. But we can't use a lot of that manufacturing capacity, and so there'll be a lot less supply of the vaccine. I think we can tie incentives to what I call global health impact. That is the encourage companies to focus on the biggest problems that cause the most death and disability well beyond the pandemic. Using things like prize funds, so rewarding companies for innovations based on the health impact of those innovations, or tying things like the orphan drug tax credit to health impact. Companies will say that profit is such great profit. They're like the richest Fortune 500 companies, right? Such great profits are necessary for getting research and development done, but they don't make their costs transparent. Critics say they're overinflated and that they include a lot of unethical marketing costs in the data that they provide. And a minimum country, companies should substantiate the claim that their profits are justified for a software, then the kind of patent protections that contribute to inequitable access. I guess if I had to decide how to distribute a vaccine, I would try to save the most lives and not the most life years, which I think might be driving the National Academy's proposal, because they don't prioritize the elderly. They kind of give them less priority and give children more priority. I think that the arguments are focusing on life years aren't that great, that for each person, the fact that their life matters so much to them, however long they might have left to live, is sufficient reason to treat that person equally, at least unless there's a better argument against that than I've seen. I think a lot of things matter besides saving lives, so economic consequences of the pandemic, for instance, are really important. And we'd have to know how to trade off lives against economic consequences to try to deal with that within a vaccine allocation system. I guess I'm not sure that's the right place to deal with it. I think there may be a tension between helping people go back to work, protecting the elderly, getting kids to school. But we can also use economic policies to address some of the economic problems. So I think a truly ethical proposal would treat every person equally and help countries get vaccines to people when they don't have the capacity to do that on their own.