Study on Auto Accident Frequency in Legal Cannabis States

This story has been making the rounds. It was actually e-mailed to me by my boss. A study by the Highway Loss Data Institute (HLDI, pronounced "Hildy") supposedly finds that states that recently legalized recreational cannabis have seen an increase in auto accidents. They reach this result using a naive sort of time-series trend analysis. This is highly dubious.  I wanted to get down a few quick reactions.

- Why are they looking just at auto collision? Why not also look at the liability coverages? Presumably liability property damage should show a similar trend. If there are more stoners and these stoners are hitting things with their cars, it should show up here. Liability bodily injury should ideally show the same trend, although admittedly this is a lower frequency coverage, so noise can swamp even a real trend. Still, it's another data source to validate their results on collision.

-In Colorado the overall collision frequency is about 5%. (As in, 5% of motorists will file a claim in a given year.) A 3% increase to this (the effect overall effect size found in the HLDI study) changes this to a whopping 5.15%. We're talking about small potatoes here.

-And yet...the effect is actually too large to be plausible. According to this document (second table, past month cannabis use rates), the population of users in Colorado rose from 7.3% in 2008/09 to 14.7% in 2014/15. So let's say 7% of the population were previously non-users but have become somewhat-regular users of cannabis (this figure would be 4% for Oregon and also 4% for Washington). For 7% of the population to drive a 3% increase in accidents, the accident frequency for those drivers would have to increase by an implausibly large 43%, which I think is higher than what anyone actually believes. Maybe if all of these new "past month users" were high all the time, but even then this is near the high end of what anyone believes is a plausible effect size.

-The calculated increase in collision frequency was 14% in Colorado, 6% in Washington, and 4.5% in Oregon. But when calculated on a "states combined" basis, the effect size was only 3%, which is smaller than any individual state. There's no mystery here; see Simpson's Paradox. The aggregate effect size can be smaller than any group's effect size (generally, whatever you're measuring) for a number of reasons. I just want to note the very wide range of estimates. It would be a mistake to pick any single one of them as the effect size. Considering my argument in the preceding bullet point, the by-state effect sizes are even more implausible than the overall effect size.

-If you look at a time series of collision frequency (industry-wide data) for Colorado and its comparison states (Utah, Nebraska, and Wyoming), nothing really jumps out at you. Colorado and Utah are sort of trending up, Nebraska is sort of flat, and Wyoming is sort of trending down. But you could easily look at the pattern and say that collision frequency is basically flat, but oscillating randomly around the 4-5% range. Any trends picked out by your eye or by regression analysis are likely to be spurious. The study is looking for very small trends (in the -5% to +5% annual change range), and saying: "We should predict that Colorado would be X based on how its neighbors are trending, but instead it's 14% above X." And it's attributing this difference entirely to cannabis legalization. Similar for the other states. This is not even close to identifying cannabis as a causal factor.

Perhaps I'll have more later.

One more thing. From the link up top: "Moore of the Highway Loss Data Institute said they hope the study's findings will be considered by lawmakers and regulators in states where marijuana legalization is under consideration or recently enacted." Matt Moore is a fine gentlemen. I've met him once and I've had several e-mail contacts with him. But I sincerely hope that lawmakers will not use his study. The study is a fine piece of time-series analysis, perhaps a good exercise for a first course in econometrics. But it's pretty lousy social science.

How Drug Laws Degrade Respect for the Rule of Law

Sometimes I see a traffic stop in which several police cars have pulled over a single vehicle. The driver and passengers are standing by the side of the road, and the police are searching the vehicle. The driver and passengers are almost always young, male, and black.

My thoughts are always, "Jeez, they're looking to bust some poor kid for drugs." I'm guessing if they searched a lot of similar vehicles, they could occasionally find a joint or a dime bag and make a few arrests. Once in a while they'd turn up a "distribution" quantity of illegal drugs. I think these kinds of traffic stops are, for lack of a better term, complete bullshit. It is not the proper role of government to harass innocent motorists in search of this kind of so-called contraband.

But perhaps this is completely unfair. Maybe the police know the individuals in the vehicle. Maybe they were looking for one of them specifically in order to question them about a recent crime. Maybe the vehicle, or one closely matching its description, was spotted at the scene of a crime. (A real crime.) Maybe a similar vehicle was recently spotted speeding away from the scene of a murder, and the police are legitimately searching for the murder weapon. 

It could be that the true explanation of the traffic stop is some combination of the legitimate policing functions described in the above paragraph. And yet my mind always jumps to "They're looking to bust some poor black kids for a bullshit drug charge." I don't think that my gut reaction is terribly unfair, either, because the police really do harass a lot of innocent people in the enterprise of drug law enforcement.  

So here I am. I'm a pretty average guy. An actuary. My job is literally to compute averages. Mid 30s. Never been arrested or had any trouble with the authorities. And yet when I see a fairly routine police activity, I'm quietly cursing them and assuming they are up to no good, because so often they are up to no good. I'm not the only one who feels this way, either. I don't instinctively react this way to the sight of law enforcement. There are many law enforcement folks at the martial arts club I am in, and I like all of them personally. I think, "Thank goodness these folks are learning how to physically handle another human being, so they don't one day unnecessarily resort to deadly force. I will coach them the best I can so they have the confidence to handle a bad situation without escalating." When I see a police car patrolling my neighborhood, I'm glad they're there. If I saw one pulling over and ticketing one of the young morons who speeds through my neighborhood, I'd quietly cheer.

A vehicle search is very different from a ticket, though. When I see a search, I have two thoughts at once: "That cop had better know for certain that that kid did something wrong" and "That's rarely the case." Feel free to write me off as being completely unfair, but just understand that this perception exists. It's a pretty common attitude, even among "upstanding taxpayers." My calling attention to it doesn't change anything. I should be able to see police officers doing their job and not have to second-guess their motives. If cops only harassed suspected murderers, thieves, batterers, and rapists, I could rest easy. But my knowledge of the injustices of drug policing forces me to do this kind of second-guessing. I think that for a police force to function properly, it needs to command the respect of the public it supposedly serves. Drug enforcement precludes them from earning that respect. 

Inefficiency of High Taxes and Large Government Sectors

There are some inefficiencies that are due to the sheer size of government and the required revenue to run such a government. A government that consumes 10% of GDP is categorically different from one that consumes 40% of revenue. The latter case isn’t just the former case scaled up by a factor of 4; the latter creates problems that don’t exist at all in the former case.

Take tax avoidance. Rich people and big businesses hire teams of lawyers and tax accountants to lower their tax bill. If you can craft a legal argument that a large expense is tax-deductible, you might be able to save your company, say, $10 million. Your employer will be willing to pay up to $10 million for those savings. From the standpoint of society as a whole, resources used to argue over who gets what are a sheer waste. The loss to society isn’t the $10 million in lost government revenue. The real loss is the alternative uses of those brilliant legal and accounting minds employed to minimize tax bills. With a very low tax rate, the tax code could be very simple. There could be no tax deductions at all, so there’s no game of “thinly slicing the salami” over what is or isn’t a legal deduction. One could do away with capital taxes, which inefficiently double-tax income that has already been earned and taxed (originally as labor income). It could all be replaced with a simple income or sales tax. But the massive revenue required to run a large government sector has led to the taxation of almost all transactions, which inherently means multiple rounds of taxation on any given dollar of earned income.

In the city where I work, there are several buildings in the downtown area that have been vacant for years. (This has changed very recently, but there was a long period of perfectly usable yet unused building space.) Someone suggested to me that this is just runaway speculation: someone is sitting on a property because it could be worth a fortune some day. But I suspect this explanation doesn’t quite cut it. If the owners are able to write off depreciation, then a vacant building can be a substantial “tax asset” to a wealthy owner. That probably doesn’t make it worthwhile to buy and hold a bunch of vacant buildings, but it can certainly move the margin of “acceptable sale price” for the owner. It means that buildings remain vacant longer than they should, for inefficient tax avoidance reasons. I don’t know how big a deal this is, but surely there are other examples of this phenomenon. We have some people holding onto these crumbling "tax assets", whereas in a saner world these things would be seen as a pure loss. The owners would sell these properties to someone who places a real value on them, and society as a whole would derive some kind of use from them. With very low rates of taxation, this becomes almost a non-issue. With 30-40% rates of taxation, the write-off becomes substantial. We miss out on the apartments, restaurants, offices, and stores that those buildings might have become.

None of this is to say it's easy to simply trim our waistlines and cut government to 10% of GDP. Sure, that would create some losers along with some winners. (It's easier than you probably think, though. The big budget items, Social Security, Medicaid, and Medicare, could be mostly replaced by a forced savings policy. We would no longer have those huge liabilities requiring high tax rates.) In this post I am not arguing for a huge reduction in government. Rather I am pointing out that a large government has very high costs, which scale up much faster than 1-for-1. A public sector that consumes 50% of GDP is much more than five times as costly as a public sector that consumes 10% of GDP. The real cost probably scales up with the square or cube of the size of government, rather than a simple linear scaling. My point here is to articulate a trade-off, not to state where we should sit on the trade-off curve. There are costs here that we should face with our eyes wide open. There is the entrepreneur who fails to sell his stock portfolio to invest in a private start-up, because he doesn't want to incur a capital gain this year. There is the business that is worth starting under a zero (or at least single-digit) tax rate, but which is rendered unprofitable at a 35% tax rate. There is the bloating of home sizes, because a more expensive home leads the a larger tax asset (because of the mortgage interest deduction). There is the implicit favoring of debt financing over equity financing because of the high corporate tax rates (because interest paid on debt is now a tax asset). There are the thousands of other distortions that don't come to mind at the moment but that economists spend their lives researching.Ideally taxes are structured so as to minimize distortions.  You don't want a tax to cause someone to zig rather than zag. (Maybe you do if it's a Pigouvian pollution tax, but if the purpose is to collect revenue you want to minimize distortions.) If having a very large government sector is worth all these collateral costs, someone should be explicitly arguing that "It's worth the cost, because the benefits are so large" rather than blithely ignoring the costs or claiming they don't exist. 

Government Gives Us the Illusion of Control

Sloppy language can betray hidden assumptions. The case for this-or-that government intervention often employs such sloppy language. I want to explore an example here.

The case for government often rests on an assumption that government action gives us control over a problem, whereas non-government solutions don’t. I often hear phrases such as, “You can’t just leave that to the marketplace.” Or  “What if there isn’t enough charity? Won’t some people get left out in the cold?” Or “Should we just let people do whatever drugs they want? Like, all the time?”

Implicitly there is this assumption that government gives us a handle on social problems. We have some kind of social problem: poverty, crime, greed, discrimination, whatever. It is a mistake to insist that leaving the problem to the marketplace is “doing nothing.” There are always private individuals working hard at solving existing problems. Some of these problems are very difficult, even inherently unsolvable. The great mistake is to think that “government” gives us a control lever over this problem, and the only question is how hard to crank the lever. One might ask, can you think of no examples of government intervention backfiring? Don’t welfare programs create very high marginal tax rates for the beneficiaries, discouraging work and leading to greater poverty? Doesn’t drug prohibition create an unpredictable, violent black market and lead to more overdoses? Do state-run services never run out of resources, such that the needy get left out in the cold anyway? Aren’t those private citizens who are supposedly stingy with their charitable givings the very same taxpayers who fund the welfare state? Should we suppose those people will suddenly become more "generous" in one context than the other?

Markets are an imperfect way of doing things. Government is another imperfect way of doing things. Both solutions are unpredictable. Both operate under enormous uncertainty. Market institutions and government institutions both often fail to achieve their stated goals. Let’s not implicitly favor one over the other. If anyone actually believes, “This government program will solve this problem, and with relative certainty!” let them say so plainly. How much private effort goes into solving that particular problem? How well does it work? What has been the track record of other government programs (similar to the one being proposed)? What can be learned from history? From other countries trying similar approaches? Are there countries where the problem is much smaller or doesn’t exist, but where the government is “doing nothing”? What’s needed here is a comparison of institutions. What one often actually gets is this glib presumption that the government intervention will work as intended.

Here are some examples of what I'm talking about. See this video with Tibor Machan on William F. Buckley's show Firing Line, around the 14 minute mark. Ernest Van den Haag apparently thinks he has a terrific "gotcha": either government intervenes or orphans starve. Van den Haag is far too glib. Bryan Caplan pointed out in an excellent post from a few years ago that every system eventually has to say "tough luck". Sure, sometimes private charity is inadequate and somebody gets thrown to the wolves. But this happens under a robust welfare state, too. (The very familiar phrase "falling through the cracks" expresses this phenomenon.)  Van den Haag offers a vaguely worded historical example of orphans starving because of inadequate private charity, apparently not realizing that the government also failed these same orphans. There was no magical "government" lever to crank. The same society that provided insufficient charity also provided insufficient government. Everyone's gotta say "tough luck" now and then. (Machan, who passed away recently, appears to  have had Caplan's "tough luck" insight on the fly, in spite of being ambushed by this "gotcha" question.)

For another example, a guest from a recent Econtalk:

Moreover, I think I would say that I don't think it should be the case that if you are a child born to poor parents you should have to rely on charity to get an education.

Again, the guest is assuming that "government education" is a control lever, which you can simply crank up (perhaps all the way to 11) if "free-market education" is inadequate.

And a guest from a not-so-recent Econtalk:

And frankly, I realize that your libertarian views aren't necessarily the same as my views about what should be done, but you libertarians have a fundamental problem that you don't seem to get, I think, which is that you are altruistic; you care about people; and when people care about other people we have the free rider problem. Who is going to take care of other people? You say charity. But if I care about that person who has got a broken arm laying in the street, and I know you do, I let you go out and try and help them; and then you let me go out and try and help them. And the guy stays there in the street with a broken arm. That's what we have--it's a public good taking care of that person.

Kotlikoff pulls out the classic "free-rider problem" (or "externality" or "tragedy of the commons" or "prisoner's dilemma") from econ 101. He apparently doesn't realize that the very same free-rider problem is what prevents us from getting good government. I could paraphrase him like this:

and when people care about good government we have the free rider problem. Who is going to read social science literature and policy whitepapers and become an informed voter? You say "civic virtue". You say "democracy". But if I care about informed voting, and I know you do, I let you go out and research government policy; and then you let me go out research government policy. And the citizens sit there in the street with broken institutions and bad policy. That's what we have--it's a public good becoming an informed voter.

Kotlikoff, like others who make this argument, is being far too glib. He's assuming there is this exogenous solution called "government" that fixes social problems, such as those free-rider type problems he describes. But "government" is endogenous. It is the same individuals living in that society infested with social problems who are supposed to give us those government solutions. "Government" isn't a generic steering wheel, or control lever, or series of adjustable dials. It's those same imperfect people who are causing society's problems, except now we are empowering them with the legitimate use of force over their fellow creatures. What could possibly go right?

Money Changes Hands, Therefore Your Rights Disappear

Apparently our freedoms disappear the moment money changes hands. I don’t understand this.

For example, some people think we have a basic right to privacy. The government has no business snooping in our private lives, peering into our bedrooms, or scoping out our associates. We have the right to be left alone. Except…suddenly this all changed the moment money changes hands. For sure, I can have all the secret dalliances I want and go to meetings of clandestine political organizations without the government tracking me. But if my interaction with another human being is “Here’s a sandwich, now pay me $6” that is suddenly blown wide open for government scrutiny. The government can demand to know the sum total of these dollar transactions, and it’s a federal crime to miscalculate this sum. I have to open my kitchen to nosy government inspectors, even though I would not have to do so if I were simply making sandwiches for my own children. (An empirical question: I wonder who is actually more likely to subject someone to food poisoning. A parent using ingredients from the fridge, or a sandwich shop? Not sure if data on this exists, but my guess would be the parent.) If “the right to privacy” is actually a principle, then that principle keeps applying after money changes hands. Money does not fundamentally taint human interactions in a way that negates our rights. If anything, market transactions, where we risk losing customers if we make mistakes, cause us to be more honest than we are in our personal lives.

I remember when the Panama Papers story broke, and everyone was seething with outrage that so many wealthy individuals had offshore accounts. I am quite certain that many of those “outraged” news junkies would, in a slightly different context, assert that we have a strong right to privacy. If asked to articulate why we need such privacy, they might give the hypothetical (or real) example of a government that prosecutes homosexuals or persecutes opposition political parties. We require a principled right to privacy for this to work. Otherwise any clever person (perhaps a government lawyer) can simply argue that the principle doesn’t apply in some particular case. And once we start playing that game, it’s not a principle at all.

We have many other rights that are supposedly negated the moment money is transferred from one hand to another. It is important that we enjoy the freedom to associate with other human beings, under whatever terms we find mutually agreeable. Again, many people will agree with this principle as it applies to marriage or political associations. But the moment I pay someone to perform a task for me, the range of possible arrangements is severely restricted. If freedom of association is even a thing, then it protects my right to sign any labor contract I want with anyone who is willing to accept those terms. Put aside for a moment the economic arguments against minimum wages and other labor market restrictions. There is a moral case for freedom of contract. If freedom of association applies to marriage contracts, there is no good reason why it wouldn’t also apply to labor contracts.

Free speech is another example. Commercial speech is severely restricted. The rationale often given for restricting commercial speech is overstated; the “problems” created by free commercial speech are completely overblown. Under no conceivable regime would businesses be able to completely misrepresent their products to their customers. Fraud would still be a crime even under a radically deregulated regime, even under full-blown anarchocapitalism. Companies that skirt the boundary between actual fraud and acceptable exaggeration (present in all advertisements, but usually obvious enough to not be misleading) run the risk of lawsuits.  The regulation of commercial speech in practice is absurd. (We have government bureaucrats telling a brewery to change the labels on their beer bottles because Santa’s eyes are “too googly” on the Christmas-themed beer.) At any rate, it’s fine to make some sort of argument that the “benefit outweighs the harm” for restrictions on commercial speech. But then you no longer believe in freedom of speech as a principle, and you should say so.

Market interactions are no different from any other kind of human intercourse. Couples, families, clubs, churches, political parties, businesses, and corporations are all just different ways of getting things done together. These are just different arrangements, specialized for achieving certain goals. Economic rights are not second-class rights. They are not even a separate category from other civil rights. The principle that allows me to arrange my social and personal affairs without government interference should logically apply to my economic affairs.

I could respect someone for saying, "Meh. I don't really believe in principle. I'm really a min-maxer. I would support whatever policy happens to be optimal in some particular case." But many people use the language of "rights" to justify things like gay marriage and political organization. And once you've articulated a right or a principle, you don't get to decide when it does or doesn't apply. 

Freedom Expands At the State Level First

Should the left favor strong, central government? They often seem to do so as a matter of principal. I think this is a mistake. Many important fights get won at the local level. They establish a foothold in a few localities and then spread to the rest of the country. The ultra-stagnant, conservative federal government simply does not move fast enough.

Gay marriage, for example, appeared to get a foothold when the mayor of San Francisco, Gavin Newsom, directed his city clerk to issue marriage licenses to gay couples. This eventually was adjudicated in California's state supreme court, which legalized gay marriage in the state. I don't know if the 2015 supreme court decision legalizing gay marriage countrywide could have happened without this initial kick at the local level.

Marijuana legalization also got its start at the state level. California legalized medical marijuana in 1996, with extremely lax standards on who can get a prescription. This was de facto legalization for the nation's largest state. I doubt if the states currently legalizing cannabis would have been so emboldened without California's example. And I doubt if public opinion would have changed so quickly without this kind of policy "experiment" at the state level. There are now several states that have adopted full legalization (well, regulated legalization, at any rate). We can point to them and say, "Look, nothing terrible happens after legalization." There isn't an explosion in vehicle accident rates, or even in cannabis use rates. You don't see really see any change in the trend line of social indicators. (Well, there's the obvious one: fewer arrests for nonsensical nonviolent "offenses".)

See no-fault divorce as another example of something starting at the state level. Once again, this is California leading the pack (under governor guess-who). I don't know the history of decriminalization of homosexuality (remember, there used to be police resources dedicated to prosecuting gays!), but from what little I know this started at the state level. The Supreme Court finally forced a recalcitrant Texas and 13 other states to fall in line in 2003, when it struck down state anti-sodomy laws as unconstitutional. But we probably never would have gotten to "Federal government puts a few hold-out states in line" if we hadn't first had "A few extremely progressive states relax their sodomy laws, against the grain of national popular opinion."

Sure, there are cases where a state or local government is "misbehaving" and needs to be told to straighten up. There were certainly some southern states that held onto racist laws for way too long, and it's probably a good thing that the federal government made them change their racist laws. But once you've so empowered the federal government, you've created the danger that it will start doing the wrong thing in every state. I don't see any particular reason to think the federal government gets policy right more often than it gets it wrong. If the federal government is weak (and can't "fix" bad local policies), then, sure, some states will have bad policy. But people can move away from those states. This weak-but-still-important mechanism of error-correction is missing at the federal level. Migrating to another country is far more costly (traumatic even) than leaving a messed-up state. I don't think it's appropriate to trot out the example of southern states in the era between the Civil War and the civil rights movement as a "proof" of the need for strong federal intervention. It is very hard to argue that this example generalizes. How about the Soviet Union? Did it "fix" the "mistaken" policy of its subject states?  Is it a good thing that China is controlled by a central government? How about India?

Don't mistake me as arguing that local sovereignty is inherently more conducive to freedom than a strong central government. The worst states can be just as stifling of freedom as the central government. Even at the very local level of cities or counties, the police force can become a government unto itself and often operates as an extralegal policy-maker. But with empowered local governments there is more experimentation, and more chances to avoid the worst abuses of government by moving away from the truly bad ones.

I get annoyed when I see various slogans and memes on Facebook, decrying that a politician at the federal level has ceded some power back to the states. This is always done far too glibly, as if there weren't a trade-off between having stronger states versus a stronger central government. As if the central government were error-free and the state governments were always at fault. My cynical interpretation is that people really aren't very principled, but just want this particular culture war conflict adjudicated in their favor (whatever happens to be the outrage of the week). Sometimes that means embracing states rights, and sometimes that means embracing a dominant federal government. Which one a person embraces often depends very much on who is "winning" that particular battle in the culture war. If, say, gay marriage doesn't have a foot-hold yet, its supporters tend to embrace states rights. If gay marriage is well-established in many but not all states, those same supporters tend to embrace a strong federal government "forcing" the last few hold-outs to step in line. To me, it all looks like cynical posturing and strategic outrage, calculated to win policy concessions. I wish that the people making these kinds of statements would articulate a principle and stick by it even when it's not convenient. That's the thing about principles: they still apply even when we wish they wouldn't.

Edit: I begin this post by addressing the left, but I should note that conservatives can also be seething hypocrites when it comes to "states' rights." It seems that they (some of them anyway) want a federal drug war to squash local experiments in cannabis legalization. And it seems they (again, some of them) want federal intervention to prevent cities from being sanctuaries for immigrants.

I avoided the term "federalism." Many people think it means "the supremacy of the federal government" while it actually means something more like "states' rights." People who argue that a policy question should be left to the states are arguing for federalism. Of course "state's rights" carries some historical baggage, but "federalism" is potentially confusing.