Indeed, the death rate in surgery is almost unique among regulated professions. One person has died in a commercial aviation accident in the US in the last nine years. Structural engineering related accidents killed at most 251 people in the US in 2016  and only approximately 4% of residential structure failures in the US occur due to deficiencies in design.
It isn’t accidental that Canada and America no longer see many plane crashes or structural collapses. Both professions have been rocked by events that made them realize they needed to improve their safety records.
The licensing of professional engineers and the Iron Ring ceremony in Canada for engineering graduates came after two successivebridge collapses killed 88 workers . The aircraft industry was shaken out of its complacency after the Tenerife disaster, where a miscommunication caused two planes to collide on a run-way, killing 583.
As you can see, subsequent safety improvements were both responsive and deliberate.
These aren’t the only events that caused changes. The D. B. Cooper high-jacking led to the first organised airport security in the US. The Therac-25 radiation overdoses led to the first set of guidelines specifically for software that ran on medical devices. The sinking of the Titanic led to a complete overhaul of requirements for lifeboats and radios for oceangoing vessels. The crash of TAA-538 led to the first mandatory cockpit voice recorders.
All of these disasters combine two things that are rarely seen when surgeries go wrong. First, they involved many people. The more people die at once, the more shocking the event and therefore the more likely it is to become widely known. Because most operations involve one or two patients, it is much rarer for problems in them to make the news .
Second, they highlight a specific flaw in the participants, procedures, or systems that fail. Retrospectives could clearly point to a factor and say: “this did it” . It is much harder to do this sort of retrospective on a person and get such a clear answer. It may be true that “blood loss” definitely caused a surgical death, but it’s much harder to tell if that’s the fault of any particular surgeon, or just a natural consequence of poking new holes in a human body. Both explanations feel plausible, so in most cases neither can be wholly accepted.
(I also think there is a third driver here, which is something like “cheapness of death”. I would predict that safety regulation is more common in places where people expect long lives, because death feels more avoidable there. This explains why planes and structures are safer in North America and western Europe, but doesn’t distinguish surgery from other fields in these countries.)
Not every form of engineering or transportation fulfills both of these criteria. Regulation and training have made flying on a commercial flight many, many times safer than riding in a car, while private flights lag behind and show little safety advantage over other forms of transport. When a private plane crashes, few people die. If they’re important (and many people who fly privately are), you might hear about it, but it will quickly fade from the news. These stories don’t have staying power and rarely generate outrage, so there’s never much pressure for improvement.
The best alternative to this model that I can think of is one that focuses on the “danger differential” in a field and predicts that fields with high danger differentials see more and more regulation until the danger differential is largely gone. The danger differential is the difference between how risky a field currently is vs. how risky it could be with near-optimal safety culture. A high danger differential isn’t necessarily correlated with inherent risk in a field, although riskier fields will by their nature have the possibility of larger ones. Here’s three examples:
Commercial air travel in developed countries currently has a very low danger differential. Before a woman was killed by engine debris earlier this year, commercial aviation in the US had gone 9 years without a single fatality.
BASE jumping is almost suicidally dangerous and probably could be made only incredibly dangerous if it had a better safety culture. Unfortunately, the illegal nature of the sport and the fact that experienced jumpers die so often make this hard to achieve and lead to a fairly large danger differential. That said, even with an optimal safety culture, BASE jumping would still see many fatalities and still probably be illegal.
Surgery is fairly dangerous and according to surgeon Atul Gawande, could be much, much safer. Proper adherence to surgical checklists alone could cut adverse events by almost 50%. This means that surgery has a much higher danger differential than air travel.
I think the danger differential model doesn’t hold much water. First, if it were true, we’d expect to see something being done about surgery. Almost a decade after checklists were found to drive such large improvements, there hasn’t been any concerted government action.
Second, this doesn’t match historical accounts of how airlines were regulated into safety. At the dawn of the aviation age, pilots begged for safety standards (which could have reduced crashes a staggering sixtyfold). Instead of stepping in to regulate things, the government dragged its feet. Some of the lifesaving innovations pioneered in those early days only became standard after later and larger crashes – crashes involving hundreds of members of the public, not just pilots.
While this only deals with external regulation, I strongly suspect that fear for the reputation of a profession (which could be driven by these same two factors) affects internal calls for reform as well. Canadian engineers knew that they had to do something after the Quebec bridge collapse created common knowledge that safety standards weren’t good enough. Pilots were put in a similar position with some of the better publicized mishaps. Perhaps surgeons have faced no successful internal campaign for reform so far because the public is not yet aware of the dangers of surgery to the point where it could put surgeon’s livelihoods at risk or hurt them socially.
I wonder if it’s possible to get a profession running scared about their reputation to the point that they improve their safety, even if there aren’t any of the events that seem to drive regulation. Maybe someone like Atul Gawande, who seems determined to make a very big and very public stink about safety in surgery is the answer here. Perhaps having surgery’s terrible safety record plastered throughout the New Yorker will convince surgeons that they need to start doing better .
If not, they’ll continue to get away with murder.
 From the CDC’s truly excellent Cause of Death search function, using codes V81.7 & V82.7 (derailment with no collision), W13 (falling out of building), W23 (caught or crushed between objects), and W35 (explosion of boiler) at home, other, or unknown. I read through several hundred causes of deaths, some alarmingly unlikely, and these were the only ones that seemed relevant. This estimate seems higher than the one surgeon Atul Gawande gave in The Checklist Manifesto, so I’m confident it isn’t too low. ^
 Furthermore, from 1989 to 2000, none of the observed collapses were due to flaws in the engineers’ designs. Instead, they were largely caused by weather, collisions, poor maintenance, and errors during construction. ^
 Claims that the rings are made from the collapsed bridge are false, but difficult to dispel. They’re actually just boring stainless steel, except in Toronto, where they’re still made from iron (but not iron from the bridge). ^
 There may also be an inherent privateness to surgical deaths that keeps them out of the news. Someone dying in surgery, absent obvious malpractice, doesn’t feel like public information in the way that car crashes, plane crashes, and structural failures do. ^
 It is true that it was never discovered why TAA-538 crashed. But black box technology would have given answers had it been in use. That it wasn’t in use was clearly a systems failure, even though the initial failure is indeterminate. This jives with my model, because regulation addressed the clear failure, not the indeterminate one. ^
 This is the ratio between the average miles flown before crash of the (very safe) post office planes and the (very dangerous) privately owned planes. Many in the airline industry wanted the government to mandate the same safety standards on private planes as they mandated on their airmail planes. ^
 I should mention that I have been very lucky to have been in the hands of a number of very competent and professional surgeons over the years. That said, I’m probably going to ask any future surgeon I’m assigned if they follow safety checklists – and ask for someone else to perform the procedure if they don’t. ^
Last week I explained how poor decisions by central bankers (specifically failing to spur inflation) can make recessions much worse and lead to slower wage growth during recovery.
(Briefly: inflation during recessions reduces the real cost of payroll, cutting business expenses and making firing people unnecessary. During a recovery, it makes hiring new workers cheaper and so leads to more being hired. Because central bankers failed to create inflation during and after the great recession, many businesses are scared of raising salaries. They believe (correctly) that this will increase their payroll expenses to the point where they’ll have to lay many people off if another recession strikes. Until memories of the last recession fade or central bankers clean up their act, we shouldn’t expect wages to rise.)
Now I’d like to expand on an offhand comment I made about the minimum wage last week and explore how it can affect recovery, especially if it’s indexed to inflation.
The minimum wage represents a special case when it comes to pay cuts and layoffs in recessions. While it’s always theoretically possible to convince people to take a pay cut rather than a layoff (although in practice it’s mostly impossible), this option isn’t available for people who make the minimum wage. It’s illegal to pay them anything less. If bad times strike and business is imperiled, people making the minimum wage might have to be laid off.
I say “might”, because when central bankers aren’t proving useless, inflation can rescue people making the minimum wage from being let go. Inflation makes the minimum wage relatively less valuable, which reduces the cost of payroll relative to other inputs and helps to save jobs that pay minimum wage. This should sound familiar, because inflation helps people making the minimum wage in the exact same way it helps everyone else.
Because of increasingly expensive housing and persistently slow wage growth, some jurisdictions are experimenting with indexing the minimum wage to inflation. This means that the minimum wage rises at the same rate as the cost of living. Most notably (to me, at least), this group includes my home province of Ontario.
When the minimum wage is tied to inflation, recessions can become especially dangerous and drawn out.
With the minimum wage rising in lockstep with inflation, any attempts to decrease payroll costs in real terms (that is to say: inflation adjusted terms) is futile to the extent that payroll expenses are for minimum wage workers. Worse, people who were previously making above the minimum wage and might have had their jobs saved by inflation can be swept up by an increasingly high minimum wage.
This puts central bankers in a bind. As soon as the minimum wage is indexed to inflation, inflation is no longer a boon to all workers. Suddenly, many workers can find themselves in a “damned if you do, damned if you don’t” situation. Without inflation, they may be too expensive to keep. With it, they may be saved… until the minimum wage comes for them too. If a recession goes on long enough, only high-income workers would be sparred.
In addition, minimum wage (or near-minimum wage) workers who are laid off during a period of higher inflation (an in this scenario, there will be many) will suffer comparatively more, as their savings get exhausted even more quickly.
Navigating these competing needs would be an especially tough challenge for certain central banks like the US Federal Reserve – those banks that have dual mandates to maintain stable prices and full employment. If a significant portion of the US ever indexes its minimum wage to inflation, the Fed will have no good options.
It is perhaps darkly humorous that central banks, which bear an unusually large parcel of the blame for our current slow wage growth, stand to face the greatest challenges from the policies we’re devising to make up for their past shortcomings. Unfortunately, I think a punishment of this sort is rather like cutting off our collective nose to spite our collective face.
There are simple policies we could enact to counter the risks here. Suspending any peg to inflation during years that contain recessions (in Ontario at least, the minimum wage increase due to inflation is calculated annually) would be a promising start. Wage growth after a recession could be ensured with a rebound clause, or better yet, the central bank actually doing its job properly.
I am worried about the political chances (and popularity once enacted) of any such pragmatic policy though. Many people respond to recessions with the belief that the government can make things better by passing the right legislation – forcing the economy back on track by sheer force of ink. This is rarely the case, especially because the legislation that people have historically clamoured for when unemployment is high is the sort that increases wages, not lowers them. This is a disaster when unemployment threatens because of too-high wages. FDR is remembered positively for his policy of increasing wages during the great depression, even though this disastrous decision strangled the recovery in its crib. I don’t expect any higher degree of economic literacy from people today.
To put my fears more plainly, I worry that politicians, faced with waning popularity and a nipping recession, would find allowing the minimum wage to be frozen too much of a political risk. I frankly don’t trust most politicians to follow through with a freeze, even if it’s direly needed.
Minimum wages are one example of a tradeoff we make between broad access and minimum standards. Do we try and make sure everyone who wants a job can have one, or do we make sure people who have jobs aren’t paid too little for their labour, even if that hurts the unemployed? As long as there’s scarcity, we’re going to have to struggle with how we ensure that as many people as possible have their material needs met and that involves tradeoffs like this one.
But when we’re making these kind of compassionate decisions, we need to look at the risks of whatever systems we choose. Proponents of indexing the minimum wage to inflation haven’t done a good job of understanding the grave risk it poses to the health of our economy and perhaps most of all, to the very people they seek to help. In places like Ontario, where the minimum wage is already indexed to inflation, we’re going to pay for their lack of foresight next time an economic disaster strikes.
I like to keep track of my life over time. I’m an obsessive journaler (and, as this blog can attest, a fairly regular blogger). At the end of every day, I track my mood, my sleep, my productivity, my social life, and how well I did in spaced repetition exercises. Last May, I decided to track one more thing about myself and start a tradition of publishing my Political Compass results yearly.
I’m a bit late this year (I kept the title because I started the post in May) because there’s actual politics happening; I’ve been volunteering for my local MPP’s re-election campaign. Of explanations for being late with a politics related blog post, that might be the best one I ever give.
Last year, I scored -3.25 on the economic axis and -6.56 on the authority axis.
This year I scored -2.0 on the economic axis and -6.46 on the authority axis, leaving me in the left libertarian camp, although continuing a seemingly inexorable trend towards economic centrism.
While my position on authority has remained virtually unchanged (I’m sure the difference was random fluctuation in how I might answer borderline questions), I do think I have meaningfully (although not largely) different views on economics than I did a year ago and I think there were two key object level updates that drove this change.
The first was the overview of rent-seeking in The Captured Economy (review forthcoming). I was already skeptical of regulation in 2017. The Captured Economy turned this up to 11 when it showed how a lot of regulation actually results in redistribution of wealth from people who are struggling to people who are affluent.
To take just one example, let’s talk about occupational licensing.
Many areas have occupational licensing. You have to complete training and apply to a licensing board in order to practice certain professions. In some cases, this makes sense. You want to know that the person building the bridge you drive across or removing your appendix really does know what they’re doing.
For other professions, the stakes are somewhat lower. There are certainly consequences if your barber doesn’t know what they’re doing. These consequences can even be quite severe if, say, improper sterilization techniques lead you to catch a blood borne disease. It would be reasonable to require all barbers to take a blood borne disease safety course every two years and have them post a proof of completion in their shops.
There is no conceivable world in which interior designers need to be held to higher regulatory standards than EMTs. None!
This isn’t a knock on barbers. I have seen just how much difference a good haircut can make. I could never in a million years do the job that interior decorators do. What I probably could do is pass a certification course on either subject. Aesthetics, the true marker of a master in either field can rarely be taught or properly judged in a classroom. But that is exactly what a lot of occupational licensing boils down to.
Really, it seems that all most occupational licensing does is raise barriers to many relatively well compensated and respected positions for people who are unlucky enough to have no education beyond a high school diploma or GED. I can see why people already in those professions would want this. Lower supply means that they can charge higher prices. They’re padding their margins at the cost of everyone who doesn’t have the free time or energy to take the necessary licensing classes (like people who work exhausting jobs or lack reliable work schedules).
If there was no occupational licensing, or if licensing was restricted to minimal courses on necessary safety for an occupation, many more people would have access to careers like hairdressing or interior decorating, careers that often pay better and afford more respect than minimum wage jobs; careers where you can be your own boss (if you’re into that sort of thing).
I know of at least one person who found out they were good at giving haircuts, but just couldn’t sit through all the schooling that was needed for a license. So now they’re giving illegal haircuts. They advertise with word of mouth, because they’re good illegal haircuts, but the whole situation, the whole idea that we can have illegal haircuts is ridiculous.
Long ago, I used to think that regulation was mostly about stopping manufacturers from dumping industrial waste in every nearby pristine forest. But now my sense is that the majority of regulation is like the rules making it illegal to give haircuts if you don’t first drop more than $3,500 on school.
The second object level update was learning just how important monetary policy was to the economy.
For a long time, I had accepted the liberal pseudo-Keynesian economical orthodoxy: we need to spend lots of money when times are rough in order to stimulate aggregate demand . Over the past year, I’ve read some monetarypolicyblogs and have started to understand that things are rather more complicated than the simple concepts I used to parrot.
I still don’t have as rigorous a grounding in economics as I would like (that’s one of the things I’d like to work on this year), but as I begin to learn more, I do find myself shifting to the economic centre because that seems to be where the truth is to be found.
I remain committed to a society that ensures enough for everyone. But I think over the past year I’ve become more disillusioned with the general level of economic literacy on the left (even in relation to myself!)  and more skeptical of the left’s ability to create the sort of plenty I still think we’re going to need to ensure human flourishing.
Last year, I made six predictions about how my political views would change over the coming year and all of them turned out to be correct . They were:
I will have an economic score > -2.25: 50%
I will have an economic score > -4.25: 80%
My top level economic identity will still be “capitalist”: 80%
I will have an authority score > -7.56: 70%
I will have an authority score < -5.56: 90%
My top level social identity will still be “libertarian”: 90%
I like this as a concept, so I’m going to try it again. My predictions for this year are:
I will still be on the left side of the graph: 80%
I will move further to the right economically: 80%
My position on the social axis will not change by more than 0.5 points: 90%
My top level political identities will not change: 90%
I will actually read an economics textbook before May 2019: 70%
I hope these predictions are more properly calibrated than my last ones!
 The common liberal take on this is different from pure Keynesian economics, because they don’t restrict “times are tough” to recessions and depressions. For a lot of modern people who say they just “support Keynesian economics” (and I was one of them), it’s always tough times for someone. ^
 I basically never see monetary policy even mentioned on the left. My guess is this is because the left largely views this as simple bean counting that is nowhere near as interesting or important as making sure we have lots of spending on social programs. Reality seems to be different, especially when you get it wrong. ^
No, this isn’t a post about very pretty houses or positional goods. It’s about the type of beauty contest described by John Maynard Keynes.
Imagine a newspaper that publishes one hundred pictures of strapping young men. It asks everyone to send in the names of the five that they think are most attractive. They offer a prize: if your selection matches the five men most often appearing in everyone else’s selections, you’ll win $500.
You could just do what the newspaper asked and send in the names of those men that you think are especially good looking. But that’s not very likely to give you the win. Everyone’s tastes are different and the people you find attractive might not be very attractive to anyone else. If you’re playing the game a bit smarter, you’ll instead pick the five people that you think have the broadest appeal.
You could go even deeper and realize that many other people will be trying to win and so will also be trying to pick the most broadly appealing people. Therefore, you should pick people that you think most people will view as broadly appealing (which differs from picking broadly appealing people if you know something about what most people find attractive that isn’t widely known). This can go on indefinitely (although Yudkowsky’s Law of Ultrafinite Recursion states that “In practice, infinite recursions are at most three levels deep“, which gives me a convenient excuse to stop before this devolves into “I know you know I know that you know that…” ad infinitum).
This thought experiment was relevant to an economist because many assets work like this. Take gold: its value cannot to be fully explained by its prettiness or industrial usefulness; some of its value comes from the belief that someone else will want it in the future and be willing to pay more for it than they would a similarly useful or pretty metal. For whatever reason, we have a collective delusion that gold is especially valuable. Because this delusion is collective enough, it almost stops being a delusion. The delusion gives gold some of its value.
When it comes to houses, beauty contests are especially relevant in Toronto and Vancouver. Faced with many years of steadily rising house prices, people are willing to pay a lot for a house because they believe that they can unload it on someone else in a few years or decades for even more.
When talking about highly speculative assets (like Bitcoin), it’s easy to point out the limited intrinsic value they hold. Bitcoin is an almost pure Keynesian Beauty Contest asset, with most of its price coming from an expectation that someone else will want it at a comparable or better price in the future. Houses are obviously fairly intrinsically valuable, especially in very desirable cities. But the fact that they hold some intrinsic value cannot by itself prove that none of their value comes from beliefs about how much they can be unloaded for in the future – see again gold, which has value both as an article of commerce and as a beauty contest asset.
There’s obviously an element of self-fulfilling prophecy here, with steadily increasing house prices needed to sustain this myth. Unfortunately, the housing market seems especially vulnerable to this sort of collective mania, because the sunk cost fallacy makes many people unwilling to sell their houses at a price below what they paid for it. Any softening of the market removes sellers, which immediately drives up prices again. Only a massive liquidation event, like we saw in 2007-2009 can push enough supply into the market to make prices truly fall.
But this isn’t just a self-fulfilling prophecy. There’s deliberateness here as well. To some extent, public policy is used to guarantee that house prices continue to rise. NIMBY residents and their allies in city councils deliberately stall projects that might affect property values. Governments provide tax credits or access to tax-advantaged savings accounts for homes. In America, mortgage payments provide a tax credit!
All of these programs ultimately make housing more expensive wherever supply cannot expand to meet the artificially increased demand – which basically describes any dense urban centre. Therefore, these home buying programs fail to accomplish their goal of making house more affordable, but do serve to guarantee that housing prices will continue to go up. Ultimately, they really just represent a transfer of wealth from taxpayers generally to those specific people who own homes.
Unfortunately, programs like this are very sticky. Once people buy into the collective delusion that home prices must always go up, they’re willing to heavily leverage themselves to buy a home. Any dip in the price of homes can wipe out the value of this asset, making it worth less than the money owed on it. Since this tends to make voters very angry (and also lead to many people with no money) governments of all stripes are very motivated to avoid it.
This might imply that the smart thing is to buy into the collective notion that home prices always go up. There are so many people invested in this belief at all levels of society (banks, governments, and citizens) that it can feel like home prices are too important to fall.
Which would be entirely convincing, except, I’m pretty sure people believed that in 2007 and we all know how that ended. Unfortunately, it looks like there’s no safe answer here. Maybe the collective mania will abate and home prices will stop being buoyed ever upwards. Or maybe they won’t and the prices we currently see in Toronto and Vancouver will be reckoned cheap in twenty years.
Better zoning laws can help make houses cheaper. But it really isn’t just zoning. The beauty contest is an important aspect of the current unaffordability.
I don’t understand why people choose to go bankrupt living the most expensive cities, but I’m increasingly viewing this as a market failure and collective action problem to be fixed with intervention, not a failure of individual judgement.
There are many cities, like Brantford, Waterloo, or even Ottawa, where everything works properly. Rent isn’t really more expensive than suburban or rural areas. There’s public transit, which means you don’t necessarily need a car, if you choose where you live with enough care. There are plenty of jobs. Stuff happens.
But cities like Toronto, Vancouver, and San Francisco confuse the hell out of me. The cost of living is through the roof, but wages don’t even come close to following (the difference in salary between Toronto and Waterloo for someone with my qualifications is $5,000, which in no way would cover the yearly difference in living expenses). This is odd when talking about well-off tech workers, but becomes heartbreaking when talking about low-wage workers.
If people were perfectly rational and only cared about money (the mythical homo economicus), fewer people would move to cities, which would bid up wages (to increase the supply of workers) or drive down prices (because fewer people would be competing for the same apartments), which would make cities more affordable. But people do care about things other than money and the network effects of cities are hard to beat (put simply: the bigger the city, the more options for a not-boring life you have). So, people move – in droves – to the most expensive and dynamic cities and wages don’t go up (because the supply of workers never falls) and the cost of living does (because the number of people competing for housing does) and low wage workers get ground up.
It’s not that I don’t understand the network effects. It’s that I don’t understand why people get ground up instead of moving.
But the purpose of good economics is to deal with people as they are, not as they can be most conveniently modeled. And given this, I’ve begun to think about high minimum wages in cities as an intervention that fixes a market failure and collective action problem.
That is to say: people are bad at reading the market signal that they shouldn’t move to cities that they can’t afford. It’s the signal that’s supposed to say here be scarce goods, you might get screwed, but the siren song of cities seems to overpower it. This is a market failure in the technical sense because there exists a distribution of goods that could make people (economically) better off (fewer people living in big cities) without making anyone worse off (e.g. they could move to communities that are experiencing chronic shortages of labour and be basically guaranteed jobs that would pay the bills) that the market cannot seem to fix.
It’s a collective action problem because if everyone could credibly threaten to move, then they wouldn’t have to; the threat would be enough to increase wages. Unfortunately, everyone knows that anyone who leaves the city will be quickly replaced. Everyone would be better off if they could coordinate and make all potential movers promise not to move in until wages increase, but there’s no benefit to being the first person to leave or the first person to avoid moving  and there currently seems to be no good way for everyone to coordinate in making a threat.
When faced with the steady grinding down of young people, low wage workers, and everyone “just waiting for their big break“, we have two choices. We can do tut-tut at their inability to be “rational” (aka leave their friends, family, jobs, and aspirations to move somewhere else ), or we can try to better their situation.
If everyone was acting “rationally”, wages would be bid up. But we can accomplish the same thing by simple fiat. Governments can set a minimum wage or offer wage subsidies, after all.
I do genuinely worry that in some places, large increases in the minimum wage will lead to unemployment (we’ll figure out whether this is true over the next decade or so). I’m certainly worried that a minimum wage pegged to inflation will lead to massive problems the next time we have a recession .
So, I think we should fix zoning, certainly. And I think we need to fix how Ontario’s minimum wage functions in a recession so that it doesn’t destroy our whole economy during the next one. But at the same time, I think we need to explore differential minimum wages for our largest cities and the rest of the province/country. I mean this even in a world where the current minimum $14/hour wage isn’t rolled back. Would even $15/hour cut it in Toronto and Vancouver ?
If we can’t make a minimum wage work without increased unemployment, then maybe we’ll have to turn to wage subsidies. This is actually the method that “conservative” economist Scott Sumner favours .
What’s clear to me is that what we’re currently doing isn’t working.
I do believe in a right to shelter. Like anyone who shares this belief, I understand that “shelter” is a broad word, encompassing everything from a tarp to a mansion. Where a certain housing situation falls on this spectrum is the source of many a debate. Writing this is a repudiation of my earlier view, that living in an especially desirable city was a luxury not dissimilar from a mansion.
A couple of things changed my mind. First, I paid more attention to the experiences of my friends who might be priced out of the cities they grew up in and have grown to love. Second, I read the Ecomodernist Manifesto, with its calls for densification as the solution to environmental degradation and climate change. Densification cannot happen if many people are priced out of cities, which means figuring this out is actually existentially important.
The final piece of the puzzle was the mental shift whereby I started to view wages in cities – especially for low-wage earners – as a collective action problem and a market failure. As anyone on the centre-left can tell you, it’s the government’s job to fix those – ideally in a redistributive way.
 This is inductive up to the point where you have a critical mass; there’s no benefit until you’re the nth + 1 person, where n is the number of people necessary to create a scarcity of workers sufficient to begin bidding up wages. And all of the people who moved will see little benefit for their hassle, unless they’re willing to move back. ^
 For us nomadic North Americans, this can be confusing: “The gospel of ‘just pick up and leave’ is extremely foreign to your typical European — be they Serbian, French or Irish. Ditto with a Sudanese, Afghan or Japanese national. In Israel, it’s the kind of suggestion that ruins dinner parties… We non-indigenous love to move. We don’t just see it as just good economic policy, but as a virtue. We glorify the immigrant, we hug them at the airport when they arrive and we inherently mistrust anyone who dares to pine for what they left behind”. ^
 I think we may have to subsidize some new construction or portion of monthly rent so that all increased wages don’t get ploughed into to increased rents. If you have more money chasing the same number of rental units and everything else remains constant, you’ll see all gains in wages erased by increases in rents. Rent control is a very imperfect solution, because it changes new construction into units that can be bought outright, at market rates. This helps people who have saved up a lot of money outside of the city and what to move there, but is very bad for the people living there, grappling with rent so high that they can’t afford to save up a down payment. ^
 No seriously, this is what passes for conservative among economists these days; while we all stopped looking, they all became utilitarians who want to help impoverished people as much as possible. ^
In simple economic theory, wages are supposed to act as signals. When wages increase in a sector, it should signal people that there’s lots of work to do there, incentivizing training that will be useful for that field, or causing people to change careers. On the flip side, when wages decrease, we should see a movement out of that sector.
This is all well and good. It explains why the United States has seen (over the past 45 years) little movement in the number of linguistics degrees, a precipitous falloff in library sciences degrees, some decrease in English degrees, and a large increase in engineering and business degrees .
This might be the engineer in me, but I find things that are working properly boring. What I’m really interested in is when wage signals break down and are replaced by a job lottery.
Job lotteries exist whenever there are two tiers to a career. On one hand, you’ll have people making poverty wages and enduring horrendous conditions. On the other, you’ll see people with cushy wages, good job security, and (comparatively) reasonable hours. Job lotteries exist in the “junior doctor” system of the United Kingdom, in the academic system of most western countries, and teaching in Ontario (up until very recently). There’s probably a much less extreme version of this going on even in STEM jobs (in that many people go in thinking they’ll work for Google or the next big unicorn and end up building websites for the local chamber of commerce or writing internal tools for the company billing department ). A slightly different type of job lottery exists in industries where fame plays a big role: writing, acting, music, video games, and other creative endeavours.
Job lotteries are bad for two reasons. Compassionately, it’s really hard to see idealistic, bright, talented people endure terribly conditions all in the hope of something better, something that might never materialize. Economically, it’s bad when people spend a lot of time unemployed or underemployed because they’re hopeful they might someday get their dream job. Both of these reasons argue for us to do everything we can to dismantle job lotteries.
I do want to make a distinction between the first type of job lottery (doctors in the UK, professor, teachers), which is a property of how institutions have happened to evolve, and the second, which seems much more inherent to human nature. “I’ll just go with what I enjoy” is a very common media strategy that will tend to split artists (of all sorts) into a handful of mega-stars, a small group of people making a modest living, and a vast mass of hopefuls searching for their break. To fix this would require careful consideration and the building of many new institutions – projects I think we lack the political will and the know-how for.
The problems in the job market for professors, doctors, or teachers feel different. These professions don’t rely on tastemakers and network effects. There’s also no stark difference in skills that would imply discontinuous compensation. This doesn’t imply that skills are flat – just that they exist on a steady spectrum, which should imply that pay could reasonably follow a similar smooth distribution. In short, in all of these fields, we see problems that could be solved by tweaks to existing institutions.
I think institutional change is probably necessary because these job lotteries present a perfect storm of misdirection to our primate brains. That is to say (1) People are really bad at probability and (2) the price level for the highest earners suggests that lots of people should be entering the industry. Combined, this means that people will be fixated on the highest earners, without really understanding how unlikely that is to be them.
Two heuristics drive our inability to reason about probabilities: the representativeness heuristic (ignoring base rates and information about reliability in favour of what feels “representative”) and the availability heuristic (events that are easier to imagine or recall feel more likely). The combination of these heuristics means that people are uniquely sensitive to accounts of the luckiest members of a profession (especially if this is the social image the profession projects) and unable to correctly predict their own chances of reaching that desired outcome (because they can imagine how they will successfully persevere and make everything come out well).
Right now, you’re probably laughing to yourself, convinced that you would never make a mistake like this. Well let’s try an example.
Imagine a scenario is which only ten percent of current Ph. D students will get tenure (basically true). Now Ph. D students are quite bright and are incredibly aware of their long odds. Let’s say that if a student three years into a program makes a guess as to whether or not they’ll get a tenure track job offer, they’re correct 80% of the time. If a student tells you they think they’ll get a tenure track job offer, how likely do you think it is that they will? Stop reading right now and make a guess.
Seriously, make a guess.
This won’t work if you don’t try.
Okay, you can keep reading.
It is not 80%. It’s not even 50%. It’s 31%. This is probably best illustrated visually.
There are four things that can happen here (I’m going to conflate tenure track job offers with tenure out of a desire to stop typing “tenure track job offers”).
A student can correctly believe they will get tenure
A student can incorrectly believe they will get tenure
A student can correctly believe they won’t get tenure
Ten students will get tenure. Of these ten, eight (0.8 x 10) will correctly believe they will get it (1/green) and two (10 – 0.8 x 10) will incorrectly believe they won’t (2/yellow). Ninety students won’t get tenure. Of these 90, 18 (90 – 0.8 x 90) will incorrectly believe they will get tenure (3/orange) and 72 (0.8 x 90) will correctly believe they won’t get tenure (4/red). Twenty-six students, those coloured green (1) and orange (3) believe they’ll get tenure. But we know that only eight of them really will – which works out to just below the 31% I gave above.
Almost no one can do this kind of reasoning, especially if they aren’t primed for a trick. The stories we build in our head about the future feel so solid that we ignore the base rate. We think that we’ll know if we’re going to make it. And even worse, we think that a feeling of “knowing” if we’ll make it provides good information. We think that relatively accurate predictors provide useful information against a small chance. They clearly don’t. When the base rate is small (here 10%), the base rate is the single greatest predictor of your chances.
But this situation doesn’t even require small chances for us to make mistakes. Imagine you had two choices: a career that leaves you feeling fulfilled 100% of the time, but is so competitive that you only have an 80% chance of getting into it (assume in the other 20%, you either starve or work a soul-crushing fast food job with negative fulfillment) or a career where you are 100% likely to get a job, but will only find it fulfilling 80% of the time.
Unless that last 20% of fulfillment is strongly super-linear , or you don’t have any value at all on eating/avoiding McDrugery, it is better to take the guaranteed career. But many people looking at this probably rounded 80% to 100% – another known flaw in human reasoning. You can very easily have a job lottery even when the majority of people in a career are in the “better” tier of the job, because many entrants to the field will view “majority” as all and stick with it when they end up shafted.
Now, you might believe that these problems aren’t very serious, or that surely people making a decision as big as a college major or career would correct for them. But these fallacies date to the 70s! Many people still haven’t heard of them. And the studies that first identified them found them to be pretty much universal. Look, the CIA couldn’t even get people to do probability right. You think the average job seeker can? You think you can? Make a bunch of predictions for the next year and then talk with me when you know how calibrated (or uncalibrated) you are.
If we could believe that people would become better at probabilities, we could assume that job lotteries would take care of automatically. But I think it is clear that we cannot rely on that, so we must try and dismantle them directly. Unfortunately, there’s a reason many are this way; many of them have come about because current workers have stacked the deck in their own favour. This is really great for them, but really bad for the next group of people entering the workforce. I can’t help but believe that some of the instability faced by millennials is a consequence of past generations entrenching their benefits at our expense . Others have come about because of poorly planned policies, bad enrolment caps, etc.
These cover the two ways we can deal with a job lottery, we can limit the supply indirectly (by making the job, or the perception of the job once you’ve “made it” worse), or limit the supply directly (by changing the credentials necessary of the job, or implementing other training caps) . In many of the examples of job lotteries I’ve found, limiting the supply directly might be a very effective way to deal with the problem.
Why? Because having people who’ve completed four years of university do an extra year or two of schooling only to wait around and hope for a job is a real drag. They could be doing something productive with that time! The advantage of increasing gatekeeping around a job lottery and increasing it as early as possible is that you force people to go find something productive to do. It is much better for an economy to have hopeful proto-teachers who would in fact be professional resume submitters go into insurance, or real estate, or tutoring, or anything at all productive and commensurate with their education and skills.
There’s a cost here, of course. When you’re gatekeeping (for e.g. teacher’s college or medical school), you’re going to be working with lossy proxies for the thing you actually care about, which is performance in the eventual job. The lossier the proxy, the more you are needlessly depressing the quality of people who are allowed to do the job – which is a serious concern when you’re dealing with heart surgery – or the people providing foundational education to your next generation.
You can also find some cases where increasing selectiveness in an early stage doesn’t successfully force failed applicants to stop wasting their time and get on with their life. I was very briefly enrolled in a Ph. D program for biomedical engineering a few years back. Several professors I interviewed with while considering graduate school wanted to make sure I had no aspirations on medical school – because they were tired of their graduate students abandoning research as soon as their Ph. D was complete. For these students who didn’t make it into medical school after undergrad, a Ph. D was a ticket to another shot at getting in . Anecdotally, I’ve seen people who fail to get into medical school or optometry get a master’s degree, then try again.
Banning extra education before medical school cuts against the idea that people should be able to better themselves, or persevere to get to their dreams. It would be institutionally difficult. But I think that it would, in this case, probably be a net good.
There are other fields where limiting supply is rather harmful. Graduate students are very necessary for science. If we punitively limited their number, we might find a lot of valuable scientific progress falling to a stand-still. We could try and replace graduate students with a class of professional scientific assistants, but as long as the lottery for professorship is so appealing (for those who are successful), I bet we’d see a strong preference for Ph. D programs over professional assistantships.
These costs sometimes make it worth it to go right to the source of the job lottery, the salaries and benefits of people already employed . Of course, this has its own downsides. In the case of doctors, high salaries and benefits are useful for making really clever applicants choose to go into medicine rather than engineering and law. For other jobs, there’s the problems of practicality and fairness.
First, it is very hard to get people to agree to wage or benefit cuts and it almost always results in lower morale – even if you have “sound macro-economic reasons” for it. In addition, many jobs with lotteries have them because of union action, not government action. There is no czar here to change everything. Second, people who got into those careers made those decisions based on the information they had at the time. It feels weird to say “we want people to behave more rationally in the job market, so by fiat we will change the salaries and benefits of people already there.” The economy sometimes accomplishes that on its own, but I do think that one of the roles of political economics is to decrease the capriciousness of the world, not increase it.
We can of course change the salaries and benefits only for new employees. But this somewhat confuses the signalling (for a long time, people will still have principle examples of the profession come from the earlier cohort). It also rarely alleviates a job lottery, because in practice people set this up for new employees to have reduced salaries and benefits for a time. Once they get seniority, they’ll expect to enjoy all the perks of seniority.
Adjunct professorships feel like a failed attempt to remove the job lottery for full professorships. Unfortunately, they’ve only worsened it, by giving people a toe-hold that makes them feel like they might someday claw their way up to full professorship. I feel that when it comes to professors, the only tenable thing to do is greatly reduce salaries (making them closer to the salary progression of mechanical engineers, rather than doctors), hire far more professors, cap graduate students wherever there is high under- and un- employment, and have more professional assistants who do short 2-year college courses. Of course, this is easy to say and much harder to do.
If these problems feel intractable and all the solutions feel like they have significant downsides, welcome to the pernicious world of job lotteries. When I thought of writing about them, coming up with solutions felt like by far the hardest part. There’s a complicated trade-off between proportionality, fairness, and freedom here.
Old fashioned economic theory held that the freer people were, the better off they would be. I think modern economists increasingly believe this is false. Is a world in which people are free to get very expensive training – despite very long odds for a job and cognitive biases that make understanding just how punishing the odds are – expensive training, in short, that they’d in expectation be better off without, a better one than a world where they can’t?
I increasingly believe that it isn’t. And I increasingly believe that having rough encounters with reality early on and having smooth salary gradients is important to prevent this world. Of course, this is easy for me to say. I’ve been very deliberate taking my skin out of job lotteries. I dropped out of graduate school. I write often and would like to someday make money off of writing, but I viscerally understand the odds of that happening, so I’ve been very careful to have a day job that I’m happy with .
If you’re someone who has made the opposite trade, I’m very interested in hearing from you. What experiences do you have that I’m missing that allowed you to make that leap of faith?
 I should mention that there’s a difference between economic value, normative/moral value, and social value and I am only talking about economic value here. I wouldn’t be writing a blog post if I didn’t think writing was important. I wouldn’t be learning French if I didn’t think learning other languages is a worthwhile endeavour. And I love libraries.
And yes, I know there are many career opportunities for people holding those degrees and no I don’t think they’re useless. I simply think a long-term shift in labour market trends have made them relatively less attractive to people who view a degree as a path to prosperity. ^
 That’s not to knock these jobs. I found my time building internal tools for an insurance company to be actually quite enjoyable. But it isn’t the fame and fortune that some bright-eyed kids go into computer science seeking. ^
 That is to say, that you enjoy each additional percentage of fulfillment at a multiple (greater than one) of the previous one. ^
 This almost certainly isn’t true, given that the marginal happiness curve for basically everything is logarithmic (it’s certainly true for money and I would be very surprised if it wasn’t true for everything else); people may enjoy a 20% fulfilling career twice as much as a 10% fulfilling career, but they’ll probably enjoy a 90% fulfilling career very slightly more than an 80% fulfilling career. ^
 I really hope that this doesn’t catch on. If an increasing number of applicants to medical school already have graduate degrees, it will be increasingly hard for those with “merely” an undergraduate degree to get in to medical school. Suddenly we’ll be requiring students to do 11 years of potentially useless training, just so that they can start the multi-year training to be a doctor. This sort of arms race is the epitome of wasted time.
In many European countries, you can enter medical school right out of high school and this seems like the obviously correct thing to do vis a vis minimizing wasted time. ^
The taxi medallion system that Uber has largely supplanted prevented this. It moved the job lottery one step further back, with getting the medallion becoming the primary hurdle, forcing those who couldn’t get one to go work elsewhere, but allowing taxi drivers to largely avoid dead times.
Uber could restrict supply, but it doesn’t want to and its customers certainly don’t want it to. Uber’s chronic driver oversupply (relative to a counterfactual where drivers waited around very little) is what allows it to react quickly during peak hours and ensure there’s always an Uber relatively close to where anyone would want to be picked up. ^
 I do think that I would currently be a much better writer if I’d instead tried to transition immediately to writing, rather than finding a career and writing on the side. Having a substantial safety net removes almost all of the urgency that I’d imagine I’d have if I was trying to live on (my non-existent) writing income.
There’s a flip side here too. I’ve spent all of zero minutes trying to monetize this blog or worrying about SEO, because I’m not interested in that and I have no need to. I also spend zero time fretting over popularizing anything I write (again, I don’t enjoy this). Having a security net makes this something I do largely for myself, which makes it entirely fun. ^
Since the minimum wage increase took effect on January 1st, Tim Hortons has been in the news. Many local franchisees have been clawing back benefits, removing paid breaks, or otherwise taking measures to reduce the costs associated with an increased minimum wage.
TVO just put out a piece about this ongoing saga by the Christian socialist Michael Coren. It loudly declares that “Tim Hortons doesn’t deserve your sympathy“. Unfortunately, Mr. Coren is incorrect. Everyone involved here (Tim Hortons the corporation, Tim Hortons franchisees, and Tim Hortons workers) is caught between a rock and a hard place. They all deserve your sympathy.
It is a truism that a minimum wage increase must result in either declining profits, cuts to other costs, or rising prices. While supporters of the minimum wage increase would love to see it all come out of profits, that isn’t reasonable.
Basic economics tell us that as we approach a perfect market, profits should fall to zero. The key assumptions underpinning this are global perfect information (so no one can have any innovations that allow them to do better than anyone else) and zero start-up costs (so anyone can enter any market at any time). Obviously, these assumptions aren’t true in reality, but when it comes to fast food, they’re fairly close to true.
It is relatively cheap to start a fast-food restaurant (compared to say opening a factory). The start-up costs for a McDonalds, KFC, or Wendy’s are $1,000,000 to $2.3 million, while a Subway costs about $100,000 to $250,000 to start. This means that whenever someone sees fast-food restaurants making large profits in an area, they can open their own and take a fraction of the business, driving everyone’s profits down.
They’re probably driven down much lower than you think. If you had to guess, what would you say the profit margins for a fast-food restaurant are? If you’re anything like people in this study, you probably think something like 35%. The actual answer is 6%.
In addition to telling me that the average fast food restaurant has a 6% profit margin, that link helpfully told me that 29% of operating expenses in a fast-food restaurant come from labour costs. Raising those labour costs by 20% by increasing wages 20% increases total costs by 6% . The minimum wage isn’t making fast-food restaurant owners make do with a little less in the way of profits. It’s entirely wiping out profits.
Now maybe your response to that is “well my heart doesn’t really bleed for that big multinational losing its profits”. But that’s not how Tim Hortons works. Tim Hortons, like almost all fast-food restaurants is a franchise. Tim Hortons the corporation makes money by collecting fees and providing services to Tim Hortons the restaurants, which are owned by the mythical small business owners™ that everyone (even the proponents of the minimum wage increase) claim to care so much about.
Most of these owners aren’t scions of wealthy families, but are instead ordinary members of their communities who saw opening a Tim Hortons as an investment, a vocation, or as a way to give back. They need to eat as much as their workers.
Faced with rising labour costs and no real profit buffer to absorb them, these owners can only cut costs or raise prices.
Except they can’t raise prices.
That’s the rub of a franchise system. The corporate office wants everything to be the exact same at every store. They set prices and every store must follow them. But there’s divergent incentives here. Tim Hortons the corporation makes a profit by selling supplies to its franchises; critically, they make a profit on supplies whether those franchisees turn a profit or not. They really don’t want to raise prices, because raising prices will hurt their bottom line.
It’s well known that (in general) the more expensive something is, the less people want it. Raising prices will hurt the sales volume of Tim Hortons franchises, which will decrease the profits at corporate Tim Hortons. The minimum wage hike affects Tim Hortons the corporation very little. They might see slightly increased shipping costs, but their costs are far less dependent on Canadian minimum wage labour. Honestly, the minimum wage increase probably is a net good for Tim Hortons the corporation. More money in people’s pockets means more money spent on fast-food.
Tim Hortons the corporation probably won’t say it, because they don’t want to antagonize their franchisees, but this minimum wage hike is great for them.
So, Tim Hortons franchisees have to cut costs or run charities. Given that they are running restaurants and not charities, we can probably assume that they’re going to cut costs. Why does it have to be labour costs that get cut? Can’t they just get their supplies for cheaper?
Here the franchise system bites them again. If they were independent restaurateurs, they might be able to source cheaper ingredients, reduce the ply of the toilet paper in their bathrooms, etc. and get their profits back this way.
But they’re franchisees. Tim Hortons the corporation has a big list of everything you need to run a Tim Hortons and you are only allowed to buy it from them. They get to set the prices however they want. And what they want is to keep them steady.
The only cost that Tim Hortons the corporation doesn’t control is labour costs. So, this is what franchisees have to cut.
There are two ways to decrease your labour costs. You can “increase productivity”, or you can cut wages and benefits. “Increase productivity” is the clinical and uninformative way of saying “fire 20% of your workers and verbally abuse the others until they work faster” or “fire 20% of your workers and replace them with machines”. While increased productivity is generally desirable from an economics point of view, it is often more ambiguous from a moral point of view.
Given that the minimum wage was just raised and it is illegal to pay any less than it, Tim Hortons franchisees cannot cut wages. So, if they’re against firing their employees and want to keep making literally any money, they have to cut benefits.
This might make it seem like corporate Tim Hortons is the bad guy here. They aren’t. The executives at Tim Hortons labour under what is called a fiduciary duty. They have a legal obligation to protect shareholder interests from harm and to act for the good of the corporation, not their own private good or for their private moral beliefs. They are responding to the minimum wage hike the way the government has told them to respond .
Minimum wage jobs suck. For all that economists claim there is no moral judgement implied in a wage, that it merely shows the intersection of the amount of supply of a certain type of labour and the demand for that labour, it can be hard to believe that there is no moral dimension to this when people making one wage struggle to make ends meet, while those earning another can buy fancy cars they don’t even need.
It is popular to blame business owners and capitalists for the wages their workers make and to say that it shows how little they value their workers. I don’t think that’s merited here. Corporate Tim Hortons has crunched the numbers and decided that if they raise prices, fewer people will buy coffee, their profits will decrease, and they might be personally liable for breach of fiduciary duty. In the face of rising prices, franchisees try and do whatever they can to stay afloat. We can say that caring about profits more than the wages their workers make shows immense selfishness on the part of these franchisees, but it’s little different than the banal selfishness anyone shows when they care more about making money for themselves than making money and giving it away – or the selfishness we show when we want our coffee to be cheaper than it can be when made by someone earning a wage that can comfortably support a family.
 As long as there are other available investments approximately as risky as opening a fast-food restaurant that return at least 6%, profits shouldn’t drop any lower than that. In this way, inefficiencies in other sectors could stop fast food restaurants from behaving like they were in a perfectly free market even if they were. ^
 This calculation is flawed, in that there are probably other costs making up total labour costs (like benefits) beyond simple wage income. On the other hand, it isn’t just wages that are going up. Other increased costs probably balance out any inaccuracies, making the conclusions essentially correct. This is to say nothing for corporate taxes, which further reduce profits. ^
Previously I described regulation as a regressive tax. It may not kill jobs per se, but it certainly shifts them towards people with university degrees, largely at the expense of those without. I’m beginning to rethink that position; I’m increasingly worried that many types of regulation are actually leading to a net loss of jobs. There remains a paucity of empirical evidence on this subject. Today I’m going to present a (I believe convincing) model of how regulations could kill jobs, but I’d like to remind everyone that models are less important than evidence and should only be the focus of discussion in situations like this, where the evidence is genuinely sparse.
Let’s assume that regulation has no first order effect on jobs. All jobs lost through regulation (and make no mistake, there will be lost jobs) are offset by different jobs in regulatory compliance or the jobs created when the compliance people spend the money they make, etc., on to infinity. So far, this is all fine and dandy.
Talking to members of the local start-up community, I reckon that many small sized hardware start-ups spend the equivalent of an engineer’s salary on regulatory compliance yearly. Instead of a hypothetical engineer (or marketer, or salesperson, etc.), they’re providing a salary to a lawyer, or a technician at the FCC, or some other mid-level bureaucrat.
No matter how well this person does their job, they aren’t creating anything of value. There’s no chance that they’ll come up with or contribute to a revolutionary new product that drives a lot of economic growth and ends up creating dozens, hundreds, or (in very rare cases) thousands of jobs. An engineer could.
There’s obviously many ways that even successful start-ups with all the engineers they need can fail to create jobs on net. They could disrupt an established industry in a way that causes layoffs at the existing participants (although it’s probably fallacious to believe that this will cause net job losses either, given the lump of labour fallacy). Also, something like 60% of start-ups fail. In the case of failure, money from wealthy investors is transferred to other people and I doubt most people care if the beneficiaries are engineers or in compliance.
But discounting all that, I think what this boils down to is: when you’re paying an engineer, there’s a chance that the engineer will invent something that increases productivity and drives productivity growth (leading to cheaper prices and maybe even new industries previously thought impossible). When you pay someone in sales or marketing, you get a chance to get your product in front of customers and see it really take off. When you’re paying for regulatory compliance, you get an often-useless stamp of approval, or have to make expensive changes because some rent-seeking corporation got spurious requirements written into the regulation.
Or the regulatory agency catches a fatal flaw and averts a catastrophe. I’m not saying that never happens. Just that I think it’s much rarer than many people might believe. Seeing the grinding wheels of regulation firsthand has cured me of all my youthful idealistic approval for it. Sometimes consumers need to be protected from out of control profit-seeking, sure. But once you’ve been forced to actually do some regulatory compliance, you start to understand just how much regulation exists to prevent established companies from having to compete against new entrants. This makes everything more expensive and everyone but a few well-connected shareholders worse off.
Regulations has real trade-offs; there are definite goods, but also definite downsides. And now I think the downsides are even worse than I first predicted.
Part of this is materials and labour. The city will probably go for something a bit more permeant than wood – probably concrete or metal – and will probably have higher labour costs (the mechanic hired a random guy off the street to help out, which is probably against city procurement policy). But a decent part (perhaps even the majority) of the increased costs will be driven by regulation.
First there’s the obvious compliance activities: site assessment, community consultation, engineering approval, insurance approval. Each of these will take the highly expensive time of highly skilled professionals. There’s also the less obvious (but still expensive and onerous) hoops to jump through. If the city doesn’t have a public works crew who can install the stairs, they’ll have to find a contractor. The search for a contractor would probably be governed by a host of conflict of interest and due diligence regulations; these are the sorts of thing a well-paid city worker would need to sink a significant amount of time into managing. Based on the salary information I could find, half a week of a city bureaucrat’s time already puts us over the $550 price tag.
And when the person in charge of compliance is highly skilled, the loss is worse than simple monetary terms might imply. Not only are we paying someone to waste her time, we are also paying the opportunity cost of her wasting her time. Whenever some bright young lawyer or planner is stuck reading regulatory tomes instead of creating something, we are deprived of the benefits of what they could have created.
When it comes to the stairs, regulations don’t stop with our hypothetical city worker. The construction firm they hire is also governed by regulations. They have to track how much everyone works, make sure the appropriate taxes go to the appropriate parties, ensure compliance with workplace health and safety standards and probably take care of a dozen minor annoyances that I don’t know about. When you aren’t the person doing these things, they just blend into the background and you forget that someone has to spend a decent part of their time filling out incredibly boring government forms – forms that demand accuracy under pain of perjury.
Hell, the very act of soliciting bids can inflate the cost, because each bid will require a bunch of supporting paperwork (you can’t submit these things on a sticky note). As is becoming the common refrain, this takes time, which costs money. You better bet that whichever firm eventually gets hired will roll the cost of all its failed past bids (either directly or indirectly) into the cost the city ends up paying.
It’s not just government regulations that drive up the price of stairs either. If the city has liability insurance, it will have to comply with a bunch of rules given to it by its insurer (or face higher premiums). If it chooses to self-insure, the city actuaries will come up with all sorts of internal policies designed to lessen the city’s chance of liability – or at least lessen the necessary payout when the city is inevitably sued by some drunk asshole who forgets how to do stairs and breaks a bone.
With all of this regulation (none of which seems unreasonable when taken in isolation!) you can see how the city was expecting to shell out $65,000 (at a minimum) for a simple set of stairs. That they managed to get the cost down to $10,000 in this case (to avoid the negative media attention of over-estimating the cost of stairs more than one hundred times over?) is probably indicative of city workers doing unpaid overtime, or other clever cost hiding measures .
The point here is that regulation is expensive. It’s expensive everywhere it exists. The United States has over 1,000,000 pages of federal regulation. Canada makes its federal regulation available as a compressed XML dump(!) with a current uncompressed size of 559MB. Considering XML overhead, the sum total of Canadian federal regulation is probably approximately equivalent to that of the United States.
This isn’t it for either country; after federal regulation, there’s provincial/state and local regulations. Then there are the interactions between all three, with things becoming even worse when you want to do anything between different jurisdictions within a country or (and it’s a miracle this can even happen at all) between countries.
People who can hold a significant subset of these regulations in their head and successfully navigate them (without going mad from boredom) are a limited resource. Worse, they’re a limited resource who can be useful in a variety of fields (i.e. there has to be some overlap between the people who’d make good programmers, doctors, or administrators and the people who can parse and memorize reams of regulation). Limited supply and consistent (or increasing) demand drives the excessive cost of buying their time that I mentioned earlier.
This is the part where I’m supposed to talk about how regulation destroys jobs and how we should repeal it all if we care about the economic health of our society. But I’m not going to do that. The idea that regulation kills jobs is based on economic fallacies  and not borne out by evidence (although it is surprisingly poorly studied and new evidence could change my mind here).
As best we can currently tell, regulation doesn’t destroy jobs; it shifts them. In a minimally regulated environment, there will be fewer jobs requiring highly educated compliance wizards and more jobs for everyone else. As the amount of regulation increases, we should see more and more labour shift from productive tasks to compliance tasks. Really regulation is one of the best ways that elites can guarantee jobs for other elites.
Viewed through this lens, regulation is similar to a very regressive tax. It might be buying us social goods that we really want, but it does so in a way that transfers wealth from already disadvantaged workers to already advantaged workers. I think (absent offloading regulatory compliance onto specialized AI expert systems) that this might be an inherent feature of regulation.
When I see progressives talking about regulation, the tone is often that companies should whine about it less. I think it’s totally true that many companies push back against regulation that is (on the face of it at least) in the public good – and that companies aren’t pushing back primarily out of concern for their workers. However, rejecting the libertarian position doesn’t mean we should automatically support all regulation. After reading this, I hope you look at regulation as a problematically regressive tax that can have certain other benefits.
Corporations have no social duty beyond giving returns to their shareholders. It’s only through regulation that we can channel them away from anti-social behaviour . Individuals are a bit better, motivated as they are by several things beyond money, but regulation is still sometimes needed to help us avoid the tragedy of the commons.
Regulation isn’t just the purview of the government. If all government regulation disappeared overnight, private regulation – overseen primarily by insurance companies – would take its place. The ubiquity of liability insurance in this litigious age has already turned many insurers into surrogate regulators .
Insurance companies really hate paying out money. They can only make money if they make more in premiums than they pay out for losses. The loss prevention divisions of major insurers work with their clients, making sure they toe the line of the insurer’s policies and raising their premiums when they don’t.
This task has become especially important for the insurers who provide liability insurance to police departments. Many local governments lack the political will to rein in their police force when they engage in misconduct, but insurance companies have no such compunctions. Insurers have written use of force policies, provided expensive training, furnished use of force simulators, and ordered the firing of chiefs and ordinary officers alike.
When insurers make these demands, they expect to be obeyed. Cross an insurer and they’ll withdraw insurance or make the premiums prohibitively high. It isn’t unheard of for police departments to be disbanded if insurers refuse to cover them. Absent liability insurance, a single lawsuit can literally bankrupt a small municipality, a risk most councillors won’t take.
As the Colombia Law School article linked above suggested, it may be possible to significantly affect the behaviour of insurance purchasers with regulation that is targeted at insurers. I also suspect that you can abstract things even further and affect the behaviour of insurers (and therefore their clients) by making arcane changes to how liability works. This has the dubious advantage of making it possible to achieve political goals without obviously working towards them. It seems likely that it’s harder to get together a big protest when the aim you’re protesting against is hidden behind several layers of abstraction .
Regulation isn’t inherently good or bad. It should be able to stand on its own merits and survive a cost-benefit analysis. This will inevitably become a tricky political question, because different people weight costs and benefits differently, but it isn’t an intractable political problem.
(I know that’s what I always say. But it’s a testament to the current political climate that saying “policy should be based on cost-benefit analyses, not ideology” can feel radical .)
I would suggest that if you’re the type of person whose knee-jerk response to regulation is to support it, you should look at how it will displace labour from blue-collar to white-collar industries or raise prices and ponder if this is worth its benefits. If instead you oppose regulation by default, I’d suggest looking at its goals and remembering that the cost of reaching them isn’t infinite. You might be surprised at what a true cost benefit analysis returns.
Also, it probably seems true that some things are a touch over-regulated if $65,000 (or even $10,000) is an unsurprising estimate for a set of stairs.
 Of course, even unpaid overtime has a cost. After a lot of it, you might feel justified to a rather longer paid vacation than you might otherwise take. Not to mention that long hours with inadequate breaks can harm productivity in the long run. ^
 It seems to rest on the belief that regulation makes things more expensive, therefore fewer people buy them, therefore fewer people are needed to produce them. What this simple analysis misses (and what’s pointed out in the Pro Publica article I linked) is that regulatory compliance is a job. Jobs lost directly producing things are more or less offset by jobs dealing with regulations, such that increased regulation has an imperceptible effect on employment. This seems related to the lump of labour fallacy, although I’ve yet to figure out how to clearly articulate the connection. ^
 In Filthy Lucre, Professor Joseph Heath talks about the failures of state-run companies to create “socially inclusive growth”. Basically, managers in companies care far more about their power within the company than the company being successful (the iron law of institutions). If you give them a single goal, you can align their incentives with yours and get good results. Give them two goals and they’ll focus on building up their own little fief within the company and explaining away any failures (from your perspective) as the necessary results of balancing their dual tasks (“yes, I posted no profits, I was trying to be very socially inclusive this quarter”).
Regulation, if set up so that it seriously affects profits (or if set up so that it has high personal consequences for managers) forces the manager to avoid acting in a ruinously anti-social way without leaving them with the sort of divided loyalties that can cause companies to become semi-feudal. ^
 The end game would quite possibly involve supermarkets setting up legally separate (with significant board overlap) charitable organizations that would handle the distribution, and compelling these shells (who would carry almost no cash so as to be judgement proof) to sign contracts indemnifying the source supermarket against all lawsuits. This would require lots and lots of lawyer time and money, which means consumers would see higher food prices. ^
 Actually, higher food prices are pretty much inevitable, because there’s still a bunch of new logistics that have to be worked out as a result of this law. If the logistics turn out to be more expensive then the fines, supermarkets will continue to throw out food (while passing the costs of the fines on to the consumer). If the fines are more expensive, then food will be donated (but price of donating it will still inevitably be passed on to consumers). Any government program that makes food more expensive is incredibly regressive – it’s this realization that underlies the tax-free status of unprepared food in Canada.
Supplemental nutrition programs (AKA “food stamps”) have the benefit of subsidizing food for those who need it from the general tax pool, which can be based on progressive taxation and mainly paid for by the wealthy.
It’s really easy to see a bunch of food sitting around and realize it could be better used. It’s really hard (and expensive) to actually handle the transport and preparation of that food. ^
 Meaning that a government that really wanted to reduce regulation would have to make it rather hard to sue anyone. This seems like an unlikely use of political capital and also probably in conflict with many notions of fundamental justice.
Anyway, you should look at changes to liability the same way you look at regulation. Ultimately, they may amount to the same thing. ^
 This is dubious because it’s inherently anti-democratic (the government is taking actions designed to be opaque to the governed) and also incredibly baroque. I’m not talking about simple changes to liability that will be intuitively understood. I’m talking about provisos written in solid legalese that tweak liability in ways that I wouldn’t expect anyone without a law degree and expertise in liability law to understand. If a government was currently doing this, I would expect that I wouldn’t know it and wouldn’t understand it even if it was pointed out to me. ^
 Note, crucially, that it feels radical, but isn’t. Most people who read my blog already agree with me here, so I’m not actually risking any consequences by being all liberal/centrist/neo-liberal/whatever we’re calling people who don’t toe the party line this week. ^
Where you come down on either of these – or any similar cases where there’s a clear trade-off between maximum access and minimum standards – is probably heavily dependent on your situation. If you’re an American millennial without an employer-provided or parental health care plan, you’re probably quite incensed about the lack of catastrophic health care insurance. For healthy young adults, those plans were an excellent deal.
I like to point out that regulation is a trade-off. Unfortunately (or perhaps fortunately), it’s a trade-off made at the middle. People in the long probability tails – those who are far from the median when it comes to income or risk-tolerance often feel left out by any of the trade-offs made by the majority. This is an almost inevitable side-effect of trade-offs that I rarely see mentioned.
If you have health problems for which Obamacare didn’t mandate coverage, then you might find yourself wishing that the coverage requirements were even more expansive. If you find yourself really hating the illegal AirBnB you’re living in with twelve other programmers, you might wish that the city’s rental enforcement unit was a bit more on their game.
Most articles about people on the extremes leave out the context and leave out the satisfied middle. They don’t say “this is the best trade-off we could get, but it’s still imperfect and it still hurts people”. They say breathlessly “look at this one person hurt by a policy, the policy hurts people and is bad; the people who advocate for it are evil.”
It’s understandable to leave out the middle in the search of a better story. The problem arises when you leave out the middle and then claim all advocates are evil for failing to care about the fringes. Because most of the time, no one is being evil.
The young people skipping out on coverage because it’s not worth it for them aren’t shirking a duty. They’re making the best of their limited finances, ravaged by a tough entry-level job market and expensive university education. The NIMBYs who fight against any change to local building codes that might make housing more affordable are over-leveraged on their houses and might end up underwater if prices fall at all.
Even appeals to principles don’t do much good in situations like this. You can say “no one should live in squalor”, but that might run right up against “everyone should be able to afford a place to live”. It can be that there simply isn’t enough housing supply in desirable cities to comfortably accommodate everyone who wants to live there – and the only way to change that involves higher direct or indirect taxes (here an indirect tax might be something like requiring 15% of new rental stock to be “affordable”, which raises the price of other rental stock to compensate), taxes that will exclude yet another group of people.
When it comes to healthcare in America, you can say “young people shouldn’t be priced out of the market”, but this really does compete with “old people shouldn’t be priced out of the market” or “pre-existing conditions shouldn’t be grounds for coverage to be denied”.
The non-American way of doing healthcare comes with its own country specific trade-offs. In Germany, if you switch from the public plan to a private plan it is very hard to get back on the public plan. This prevents people from gaming the system – holding cheaper private insurance while they’re young, healthy and earning money, then trying to switch back during their retirement, but it also can leave people out in the cold with no insurance.
In Canada, each province has a single, government-run insurance provider that charges non-actuarial premiums (premiums based on how much you make, not how likely you are to use healthcare services). This guarantees universal coverage, but also results in some services (especially those without empirical backing, or where the cost-benefit is too low) remaining uncovered. Canada also prohibits mixing of public and private funds, making private healthcare much more expensive.
Canadians aren’t spared hard choices, we just have to make different trade-offs than Americans. Here we must pick (and did pick) between “the government shouldn’t decide who lives and who dies” and “care should be universal”. This choice was no less wrenching then any of those faced by Obamacare’s drafters.
Municipalities face similar challenges around housing policy. San Francisco is trying to retain the character of the city and protect existing residents with rent control and strict zoning regulations. The Region of Waterloo, where I live, has gone the other way. Despite a much lower population and much less density, it has almost as much construction as San Francisco (16 cranes for Waterloo vs. 22 for SF).
This comes at a cost. Waterloo mandated that houses converted into rental properties cannot hold more than three unrelated tenants per unit, thereby producing guaranteed renters for all the new construction (and alleviating concerns about students living in squalid conditions). The region hopes that affordability will come through densification, but this cuts down on the options student renters have (and can make it more expensive for them to rent).
Toronto is going all out building (it has about as many cranes on its skyline as Seattle, Los Angeles, New York, San Francisco, Boston, Chicago, and Phoenix combined), at the cost of displacing residents in rooming houses. There’s the hope that eventually supply will bring down Toronto’s soaring house costs, but it might be that more formal monthly arrangements are out of the reach of current rooming house residents (especially given that rent control rules have resulted in a 35-year drought on new purpose-built rental units).
In all of these cases, it’s possible to carve out a sacred principle and defend it. But you’re going to run into two problems with your advocacy. First, there’s going to be resistance from the middle of society, who have probably settled on the current trade-off because it’s the least offensive to them. Second, you’re going to find people on the other underserved extreme, convinced all the problems they have with the trade-off can be alleviated by the exact opposite of what you’re advocating.
Obamacare looked like it would be impossible to defend without Democrats controlling at least one lever of government. Republicans voted more than 50 times to repeal Obamacare. Now that they control everything, there is serious doubt that they’ll be able to change it at all. Republicans got drunk on the complaints of people on the long tails, the people worst served by Obamacare. They didn’t realize it really was the best compromise that could be obtained under the circumstances, or just how unpopular any attempt to change that compromise would be.
This is going to be another one of those posts where I don’t have a clear prescription for fixing anything (except perhaps axing rent control aka “the best way to destroy a city’s rental stock short of bombing it”). I don’t actually want to convince people – especially people left out of major compromises – not to advocate for something different. It’s only through broad input that we get workable compromises at all. Pluralistic society is built on many legitimate competing interests. People are motivated by different terminal values and different moral foundations.
Somehow, despite it all, we manage to mostly not kill each other. Maybe my prescription is simply that we should keep trying to find workable compromises and keep trying not to kill each other. Perhaps we could stand to put more effort into understanding why people ask for what they do. And we could try and be kind to each other. I feel comfortable recommending that.