My friend Mike asked me about the proposed law in the Canadian province of Quebec that would levy a special tax on people who choose to remain unvaccinated against covid, given their likelihood of putting extra strain on the health care system in this time of crisis. So you can blame him for how long this post is—almost 4000 words. Sorry.
By the time you read this most hospitals in Canada will be overflowing, and a disproportionate number of those cases will be unvaccinated people. It would be desirable if our hospitals were not overflowing.
Would a tax on the unvaccinated help? And would it be justified, either in the current crisis or in the post-crisis environment we all hope to reach sooner rather than later?
This is long in part because I first spend a lot of time laying out some ideas on moral reasoning, which is something I do slowly and in a fairly unusual manner. I wander around a lot. This is vital to my process, and helps me avoid getting captured by the first idea that comes into my head. As I emphasize in this peice, running out of attention is always a risk when thinking about anything, and taking a long, leisurely stroll through the garden of ideas is one way to ensure we take time to consider more than one angle. We can’t get more attention, but we can get more time.
I did not have the opinion I do now when I started this. I might not have the opinion I do now by the time this is pubilshed. But for now, this is what I think about taxing the unvaccinated, and how I reached that point.
Approaches
A lot of moral reasoning is done via thought experiments and hypotheticals.
I'm not a big fan of thought experiments as a means of exploring external reality because they privilege our imagination in ways that are contrary to everything we know about our imagination. Thought experiments are sure to feed our biases and limitations in ways that do far more to illuminate our cognitive shortcomings than yield any particular moral or physical insight, and they often lead us down blind alleys where we end up stuck, sometimes for generations.
This is not to say a thought experiment can't be useful for motivating an argument: Einstein used them, after all. So did Newton. But three paragraphs later Einstein was talking about measurements we could actually do, which has become my rule for evaluating arguments involving thought experiments. Call it the "Einstein Criterion": if someone introduces a thought experiment, and they're still talking about it three paragraphs later, I'm done. They don't have anything very interesting to say about the reality outside of their own head, and while I'm sure the reality inside their head has some fascinating features, they aren't likely to be of general interest in discussing external-world problems.
Consider the Trolley Problem. It's a nice illustration of one of the most basic problems with thought experiments, which is that we are not very good at thinking. In particular, our imaginative capacity gets exhausted very quickly--three to five concrete elements are all we can manage to hold onto at once--so we can't keep all the moving parts of the thought experiment in mind together, which makes us susceptible to whatever otherwise-indefensible idea the purveyor of the thought experiment wants to put over on us.
I'm increasingly convinced that a great deal of rhetoric is aimed at achieving "attentional exhaustion": introduce half a dozen things and then push the idea you want to sell. Our minds are tuned up to grab on to that single "solution" for no better reason than it gives us license to quit doing the work to juggle those half-dozen things, which don't even have to be relevant to the problem at hand. So keep an eye out for that.
The most basic version of the Trolley Problem goes like this: you are standing at a switch point where a train track has a Y-branch, and there are three NAZI war criminals tied to the tracks on one branch of the tracks and and a scientist who is in sole possession of the cure for cancer is tied to the other. A train (called a "trolley" because the "Train Problem" would sound too prosaic) is approaching along the tail of the Y, and for some reason you can't do anything but pull the switch--the scenarios to justify this constraint get extremely silly even by the standards of philosophers--and the supposed problem is: Where will you and the scientist have lunch? At least I think that's it. Maybe I've muddled the details.
One thing that's never made clear is what the problem is supposed to be about. It can't be about moral reasoning, because how people answer the problem has literally nothing to do with how they behave in real life. The problem was first published in 1968 and it took until 2018 for anyone to bother to check that it had anything to do with... anything, really. And when someone did, they found it didn't.
This has resulted not in the abandonment of the Trolley Problem as an instrument of interest to philosophers, but rather engendered a new subfield in moral philosophy that might be called "Special Pleading about Why the Trolley Problem Still Matters".
All of which is to say: there are a lot of ways to approach reasoning about moral problems, and introducing imaginary scenarios--beyond their use as a transient tool to motivate arguments that have actual substance--is not one of them.
Covering the Ground
So having established that, what about taxing the voluntarily unvaccinated? This is a phase of thinking I call "covering the ground" where I look at what other people are saying and rough out some preliminary ideas so I've got a sense of landscape I'm wandering around, and especially what to avoid in it. What other people are saying is often a very good guide to what should be avoided, but otherwise it's a bit of a random walk, with a bias toward any fact or idea that makes me uncomfortable.
If one were to set this up as a Trolley Problem scenario, there would be several thousand people a week with postponed surgeries--cancer patients, people with severe mobility limitations, chronic pain, and so on--tied to one branch of the tracks, and a rabble of conspiracy-minded free-dumb loving red necks on the other. Or not: the average anti-vaxxer in Canada is a 42 year old white woman from Ontario who votes Liberal.
And she's not an "anti-vaxxer", although I don’t seem to be able to ween myself off that term: she's vaccine hesitant, which can be caused by any number of things, from genuine and extreme fear of needles to a not-entirely-unjustified distrust of the vaccine-industrial complex, which has sent a variety of mixed messages over the past two years, from "healthy younger people aren't at much risk" to "if you don't get vaccinated today you're killing grandpa".
People are not great at parsing empirical, data-driven arguments (trust me on this) but they do have a pretty good sense when they're being manipulated and lied to, and while I am 100% behind vaccinations for all and pretty severe restrictions on the unvaccinated, there is no way I can say with a straight face that public health authorities have done a good--or even honest--job of communicating in Canada over the course of the pandemic.
In BC our Public Health Officer, Bonnie Henry, is famous for word-salad answers that avoid simple, straight-forward questions, for appearing unmasked at hockey games, and for wearing a cloth mask when the expert consensus has been for a long time that N95 or similar masks are strongly indicated. She's incoherent, inconsistent, and gives every appearance of dishonesty. I am not surprised people distrust someone who insists that when you sit down in a restaurant you are no longer at risk from second hand smoke.
Constructing unsavoury caricatures of your opponents is an ancient and entirely discreditable enterprise, regardless of the target. The claim "you support policy X therefore you must have reasons Y which makes you a person of type Z" is not an argument, it's a smear that attempts to shut down argument by painting anyone who favours X for any reason as being the worst kind of person imaginable, and who would listen to them? This is true even in cases when the policy position has no known good arguments in its favour.
This move to caricature can been seen in attempts to paint anyone in favour of taxing the unvaccinated as being necessarily motivated by purely punitive reasons, which is so obviously false that my policy is simply to stop reading when I encounter it: my time is too valuable to waste engaging with such fundamentally disingenuous or myopic rhetoric.
Taxing the unvaccinated is a question of policy, not motivation. It would, after all, be pretty bad to argue that policies that have resulted in thousands of preventable deaths--as many of Canada's covid policies have--are good policies because the people who implemented had the purest motives. Likewise, if a policy is good, it's good even though some of the people who support it have bad motives.
What is a Good Policy?
Me, I'm in favour of policies that save lives, but not at any cost. I am deeply aware that any policy choice is a trade-off between what is done and what is not done, all constrained by what is practical, politically and economically. Political and economic constraints are elastic, but not infinitely so.
This means that the question is never "Is this policy ideal?" but "Is this policy plausibly enough among the better practical alternatives that government resources would be usefully spent on formulating and implementing IT rather than SOMETHING ELSE?" The "something else" is important here, because in most cases the best argument against a given policy is, "You'd be way better off doing THIS rather than THAT."
So a lot of policy debate is relative, rather than absolute. That said, some absolutes matter.
For example, I have a absolute tendency to favour the innocent and helpless over those who have a choice to behave differently. In Trolley Problem terms, given a single cancer patient tied to the tracks on one branch and a single anti-vaxxer standing voluntarily on the other with their eyes closed and refusing to move, I wouldn't hesitate to send train in the direction of the anti-vaxxer, although I know I'd feel terrible about it.
Unhelpful Things
We've already established that putting things in Trolley Problem terms is unhelpful: I mention it above simply to illustrate to the hard of thinking that imputing a single, punitive, motive to everyone who favours saving cancer patients over anti-vaxxers is wrong.
Other things that are unhelpful: pretending that a policy that is somewhat inconsistent with other policies is somehow shocking or novel. People who say, "OMG we can't do that because it would be inconsistent with this other thing we already do!!!" are merely identifying themselves as not knowing anything much about the welter of policy contradictions and inconsistencies we already comfortably live with.
Likewise, slippery slope arguments are rarely worth much: arguing that marriage equality would lead to people wedding their pet llama is about as silly as any other slippery slope argument. Arguments should be for or against the policy in question, not the (imagined... always imagined) “ultimate consequences” of such a policy.
A slight variant on the slippery slope is "Where do we draw the line????" as if we don't do a rough and ready job of drawing pretty arbitrary but still reasonably functional lines every single day. Pretending that no one is capable of one of the most ordinary everyday executive activities is not a compelling argument.
And finally on the list of things to avoid: pretending that we've never done anything like this before is almost always an indicator that the person making the argument has never done any thinking about this before, and giving prominence to such voices is yet one more distraction where we can least afford it.
As you'll see from the above, heuristics to avoid or dismiss distractions and irrelevancies, often stemming from our imagination, loom large in my reasoning about questions of ethics and public policy. Again: we only have so much attention. Letting people grab it with empty, imagination-based rhetoric, is not helpful.
Identifying Principles
So what is right?
As I said--and as you can now see--I reason very slowly, and rarely get to the answer I eventually settle on with my first try. As I write these words right here, I have no idea where the next few paragraphs will lead me. Let's find out! [Edit: I've since edited some of the previous sections after writing what is below, since I now know where I'm headed.]
I find it useful to identify the principles I think are involved, but again: this may not be so useful to others who approach these questions differently. I'm not equipped with much in the way of intuition or whatever, so this is the way I do it.
There are at least three principles involved, and two timescales that I can see. The principles are taxation simplicity, social democratic universality, and emergency justifications. The timescales are short--weeks to months--and long: years to decades.
The principle of taxation simplicity is that tax policy should for the most part be used as a revenue generator, not an instrument of social policy. Taxes should be broad, as simple as possible consistent with ability to pay, and aimed at nothing more than generating sufficient revenue to fund the government.
This is a conservative principle that is well-backed by data: complex, boutique taxes of the kind favoured by Conservatives (as opposed to conservatives) are easily gamed by the rich and lead to unnatural economic distortions. The possibility of such distortions should always be considered when tax changes of any kind are proposed because--this will come as a great shock to many--individual human beings are not passive recipients of government policy, but active, quite clever, and often rather perverse creatures who will respond to policies in ways that policy makers cannot possibly imagine. The latter point is important: like the universe itself, human reactions are not just stranger than we imagine, they are stranger than we can imagine.
That said, in some instances major shifts in tax structure can be justified by sufficiently dire social need, and when taxation is an efficient and effective tool for addressing those needs. Carbon taxes fall into this category when properly implemented: they are broad, simple, and revenue neutral. They also tax something we want less of, and "when you tax something you get less of it" is usually pretty accurate.
With regard to social democratic universality, boutique taxes on specific unhealthful behaviours also have a long and somewhat chequered history. The most obvious examples are tobacco and alcohol, but they come with object lessons as well. Cigarettes were heavily taxed in Ontario in the late '80s and early '90s, which resulted in the unintended consequence of vastly increasing the level of smuggling by indigenous people whose traditional lands spanned the St Lawrence River.
So while taxes were high enough to reduce the number of smokers and the amount of smoking--thereby saving lives and freeing up hospital beds for people who needed them for harms that were not self-inflicted--they also resulted in tax avoidance behaviours that made parts of the St Lawrence dangerous to night-time navigation because of the number of high-speed smuggling craft that were crossing without lights. Notices to Mariners identified this hazard and recommended not traversing the river at night on a stretch of water near Brockville due to safety concerns.
As an example of why imaginary arguments fail, who would have imagined that increasing taxes on cigarettes would make part of the St Lawrence un-navigable at night? No one. Slippery slope and other "dire consequence" arguments, when not backed by external-world data, reveal nothing but the deepest fears of the people who make them, not what is likely to actually happen.
For example, slippery slope-ists imagine that eliminating drug laws will result in universal addiction and the destruction of society as we know it. Two decades of external-world data from Portugal tells us things will be better in a lot of ways than they are in drug-prohibitive states.
So what we can say about taxing the unvaccinated is that while it may induce some of them to get vaccinated, I would want to see external-world data that shows a tax sufficiently high to motivate enough laggards to make a difference wouldn't be sufficiently high to motivate potentially deleterious evasive behaviour that I can't imagine. Neither can you.
If Quebec does go ahead with its plan, I would recommend it be a five- or ten-year pilot program designed to gather these data. I'll leave it up to more clever people than me to design such a program, but I'd suggest increasing the level of tax each year for five years to see how people respond to it. The trial period has to be long enough that folks won't just wait it out.
My expectation is that taxing the unvaccianted won't work very well: it'll motivate some tens of percent of the vaccine hesitant to get their shots, but it will also almost certainly harden the resistance of many. We have any amount of external-world data that coercion almost always has the primary result of resistance, not acquiescence. It's why so many modern governments, like those in Canada, use lies, misinformation, and marketing to induce complacency and compliance, rather than overtly coercive policies in most cases.
So I would rank the plausibility of the idea "taxing the unvaccinated will substantially reduce the problem unvaccinated people pose for our social democratic health care system" as low.
And at the same time, it will expand the scope of government limitations on what you can or cannot do and still be covered by the socialized health care system. We already do that with alcohol and tobacco, as I mentioned, so there is nothing especially new about this, and one more thing probably won't put us on a slippery slope to death panels, but just as taxation should not in general be used as an instrument of social policy--because it doesn't work very well and has a variety of negative side effects--I'm doubtful about the utility of using access to or payment for the health care system as a social policy tool, particularly because in this case it would be "solving" a problem that is an artifact of bad policy in the first place.
The Big Picture
Any moral argument, especially when dealing with public policy, must be considered in context. The context of any argument about taxing the voluntarily unvaccinated is the current stat of Canada’s health care system.
Health care is a social good. Decades of experience in the developed world shows that the cheapest, most effective way to produce good health outcomes is to have a layer of universal basic coverage available to all citizens. Canada's just happens to not be all that great.
Canada's health care system is expensive and has mediocre outcomes compared to most of the developed world, but rational discussions of how we can do better are inevitably stymied by a combination of entrenched interests--from government bureaucrats to doctors to public sector unions--and clouded by smoke from the dumpster fire next door. We should look further afield for inspiration and ideas as to how w can improve things, and ignore anyone who suggests that the US health care system is in any way relevant to the debate. That includes anyone who accuses those of us in favour of reform along European lines of wanting “US style health care”, which is a smear frequently deployed to stifle debate on health care policy in this country.
To be very clear: universal access to basic care improves the quality of life for everyone. The data from everywhere support this. That Canada does a poor job of it is not a reason to abandon it, but to improve it.
And I don't think a boutique tax on a specific hot-button behaviour is an improvement.
Alternative Policies
Which brings me to the third principle: emergency justifications. And also timescales. These bring in the question of how taxing the unvaccinated compares to our other policy responses in recent months, and if we have shown a willingness to do enough that scraping the bottom of the policy barrel--which I think is what I've shown "taxing the unvaccinated" to be--is justified. Are we there yet?
That is, are we in a sufficient emergency, and have we already done enough to address that emergency, to undertake a speculative policy that has a good chance of not working, and a reasonable chance of inciting a significant and unhelpful backlash--when what is most plausibly needed is a longer-term reform of our health care system along European lines, where practically every nation has better outcomes for lower or comparable costs?
I would say not, as we have apparently decided that ten to twenty thousand dead over the next six weeks is a suitable price to pay for avoiding the discomfort and inconvenience of enacting policies that would actually be effective in limiting the spread of the virus:
I called for those policies in mid-December. Many other expert voices were calling for the same: universal N95 masking in indoor public spaces, especially in hospitals and schools. Accelerated boosters. Improve air filtration and ventilation in schools and hospitals.
Our voices were ignored.
It is difficult to take seriously the idea that we care about cancer patients being displaced by anti-vaxxers when our governments did practically nothing to prevent a massive wave of hospitalizations (now) and deaths (soon), and people were pretty much OK with that. Or were there protests that I missed?
The data we had in early December was enough to predict a massive wave of new hospitalizations and consequent deaths in Canada starting in January and peaking in February. We could see from Europe that the risk of hospitalization was about a third to a half of what it was with delta. It certainly wasn't an order of magnitude less, which is all we needed to know when it comes to hospitals being overwhelmed. "Milder" is not a number, and everyone who talked reassuringly about "milder" disease in the run-up to the Christmas holiday is in the same ballpark of evil as the average vaccine-hesitant individual, from where I stand.
One of the very few thing some provinces did right was to cancel "non-urgent" surgeries starting in January. Those cancellations were not predicated on anti-vaxxers, but on total case numbers, which would be overwhelming even without the disproportionate number of the unvaccinated among them.
So now I know what I think: taxing the unvaccinated is not good policy.
Conclusion
Wiser minds than mine will be able to say if it would pass a Charter challenge, but I'm comfortable at this point saying it is a mis-allocation of resources that distracts our precious attention from what we should be focused on right now, which is systemic, incremental, experimental, reform of the Canadian health care system along roughly European lines so that we have better outcomes at lower or comparable cost for all Canadians, in this generation and for the future.
Those reforms were direly needed before the pandemic. They are even more badly needed now. Spending more is not required: we already spend more than most systems that get better outcomes. In the short term we have demonstrated a societal unwillingness to inconvenience ourselves even a little bit to reduce the scope and scale of the omicron wave. We are not denying care to or taxing people who travelled over Christmas and caught covid.
So from where I stand, as an incredibly frustrated scientist whose voice has been ignored along with every other expert's on these questions, singling out anti-vaxxers for special treatment in this regard is not remotely justified.
That's more-or-less where I think I'm going to land on this: our weak policy response has resulted in more deaths than anti-vaxxers have, and that would be true no matter how small the anti-vaxx population gets. In fact, our poor policy response has been a key factor in amplifying the effect of anti-vaxxers, who have not been a threat to the health care system in nations like Taiwan and New Zealand, where covid policy has been far more robust. Whereas in Canada a covid-zero policy gets a shrug and an eye-roll because “it would be hard”… compared to what? Setting up field hospitals? Letting ten thousand people die in the next six weeks?
This is an "upstream" argument: it says that dealing with the downstream effect of people falling into the river is generally a bad way of responding to drownings. We need to go up river and find out why they are falling in. In the present case, it's because Canada's heath care system isn't very good compared to European ones, and our covid policy is not very good compared to Taiwan or New Zealand. We should focus our precious, limited, attention on those deficits, not on taxing anti-vaxxers, which is arguably just a distraction from our larger and more lethal policy failures.
But maybe, for the people who are putting these ideas about, that is the actual point of this policy proposal: the effect on reducing the number of un-vaccinated will be minor. The effect on reducing and diverting discussion around needed policy reforms in Canada’s health care sector will be huge.
I appreciate the time and thought you put into writing this piece. With minimal thought put into the issue, I felt myself leaning toward taxing the unvaccinated. Your writing changed that. I now see that the focus needs to be on the problems upstream. Thanks TJ.