Life Lessons

The Washington Post and others of that ilk have been sounding the alarm about declining life expectancy in the U.S., driven mostly by rising death rates due to opioid use and obesity among the young and middle-aged.

There’s something terribly wrong says the Post headline. Maybe. But the content of the article suggests that a better headline might have been There’s something to celebrate. Certainly the article offers no argument or evidence for the former interpretation.

There is something to celebrate if rising death rates result from voluntary, informed choices. The thing to celebrate, of course, is not the deaths themselves, but the fact that people have found something worth dying for — just as, when you buy a house, I’ll congratulate not for the expense, but for finding something that made the expense worthwhile.

You’d think (or at least I‘d think) this was entirely obvious, but apparently it’s not obvious to everyone, so maybe it’s worth offering an extreme example. Suppose we’re all tied to beds in hospital rooms, with doctors monitoring every blip in our health and attending to it immediately. As a result, we mostly live long and miserable lives. Now one day, we engineer a mass escape. Life expectancy goes down for reasons that call for a celebration.

Opioids offer escape from miserable lives. They also offer enhancement of non-miserable lives. The decision to be obese comes with a great many perks — you can spend a lot more time eating M&M’s and a lot less on the treadmill. (I myself spend much of my treadmill time wondering whether I’ve made the wrong choice.) Like all good things, these come with costs. In this case, the cost is in the form of increased mortality. Apparently, people think that’s a cost worth paying. We should be glad for them.

Now you can certainly tell a story in which mortality due to opioid use is up because the world has gotten so much worse that there’s more demand for escape. That would be a bad trend (though even then, we’d want to celebrate the ability of opioids to mitigate some of that misery). Or you can tell a story in which mortality due to opioid use is up because people have gotten so much richer they can afford to be opioid addicts, or because opioids have gotten better, or because they’ve become more readily available. That would be a good trend. The Washington Post article alludes to the former possibility, without a shred of a good reason to think it’s the right story, as opposed to one of many possible stories.

You know what else is way up over the past couple of decades? Expenditures on smartphones. That sounds really really bad if you choose to ignore the fact that the people who are spending all that money get to have smarthpones. Likewise, an upward trend in mortality from M&M consumption sounds really really bad if you choose to ignore the fact that the people who are shortening their life expectancies also get to eat a lot of M&Ms. There is more to life than life expectancy.

Print Friendly, PDF & Email
Share

44 Responses to “Life Lessons”


  1. 1 1 Dave

    Does this mean you prefer the Becker view of the rational addict to the Schelling view?

    I think the issue is that people underestimate the power of opioid addiction. Smart phones and M&Ms may have an element of addiction, but not nearly to the degree of opioids. You can argue that people should know this by now, but not everyone has good judgment, and addiction can occur after just one use. That’s steep price to pay for poor judgment, relative to someone who splurges one night on M&Ms or upgrades a cell phone too early.

    Not to mention people who became addicted from prescription opioids weren’t necessarily making much of a choice.

    I’m not saying everything in the post is wrong, but the analogies seem weak.

  2. 2 2 Harold

    There was a big claim that newer opioids were less addictive. This turned out to be false. People were not making an informed choice.

    On a more philosophical note, we identify with our future selves less than completely. I think it is worth reducing my pain now and risk some other person (my future self) being addicted. The result is a sort of negative externality, which we all know can result in decreased social welfare. This is familiar old ground and nothing new, but I think it applies particularly with opioids. This intertemporal bargaining may be more broken in addicts.

    George Ainslie comments about policy makers “Unless they have experienced such bankruptcy themselves, trying to put themselves in the addict’s shoes leads them to conclude that she just needs an extra push, whereas in fact she needs to re-establish a relationship with her prospective future selves.”

    The market requires rational, informed actors in order to produce an efficient outcome. It is true that such a being does not exist at all, and thus the outcomes we expect from the market are less than perfect, but often we have no better way to do things. With opiates, the degree of irrational and ill-informed actors could well be so extreme that there are better ways than leaving it to the market.

    As Dave says above, this analysis assumes the rational addiction theory. There does seem to be supporting evidence with smoking. The picture is not so clear with street drugs, and one recent study found evidence supporting the utility misprediction model, which is basically what I was describing above. They say:

    “This provides empirical support for the utility misprediction model. Further, we find that the decrease in life satisfaction following the consumption of illegal/street drugs persists 6 months to a year after use. In contrast, the consumption of cigarettes is unrelated to life satisfaction in the close past or the near future.”

    https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3088298

    My initial thoughts are that someone in pain and prescribed a drug they are told will not be addictive is even more likely to suffer from utility misprediction.

  3. 3 3 Jonas

    @Dave,

    I know several people who have quit heroin, but not a single person who has quit the smartphone. Of course one can think of many differences between the cases, but if we are to tread addictiveness as a single variable, surely the actual frequency of successful quitting ought to be the most important figure?

    (Especially considering that the reported addictiveness is likely to be distorted by things like people wanting Methadone prescribed, or being unwilling to admit to smartphone addiction because that will be seen as mere weakness of will, &c.)

  4. 4 4 Harold

    This is an interesting read.
    https://www.hannapickard.com/uploads/3/1/5/5/31550141/alternative_models_of_addiction.pdf

    In the paper Addiction: choice or compulsion? Hendon et al say of rational choice theory:
    “Put differently: the choice problem that is claimed to generate addictive behaviors is neither the one individuals actually face, nor the one they believe themselves to be facing, nor indeed the one the average smoker or “junkie” is likely to recognize if it
    was explained to them, nor a choice problem they would be able
    to find the optimal solution to in practice.”

    they say
    “In our view the most interesting work on addiction from economists comes from researchers who try to establish a middle ground between viewing addiction as a disease that completely removes the addict’s ability to control his behavior and viewing it as nothing more than an extremely subtle and sophisticated form of maneuvering undertaken by the addict in the face of complex incentives.”

    This seems a very reasonable statement.

    Part of the problem seems to be that promoting addiction as a brain disease was a tactical decision in order to get more funds for treatment. This was very successful, but the downside is that the treatments offered are not as effective as they could be if a more sensible approach had been taken. I suggest that an equally damaging result follows from viewing it a simply a matter of rational choice.

    The authors above discuss similarities between addiction and compulsion. People suffering from OCD may be said to be making a rational choice to wash their hands for the 15th time, because that is their revealed preference. However, this misses the point that the behavior is destructive, and the person knows it is destructive and wishes not to be doing it. To me it is not very useful to view this as simply a rational choice in the face of irrational compulsive urges.

  5. 5 5 Ted

    I don’t understand your claim that “the content of the article suggests that a better headline might have been ‘There’s something to celebrate.'” As you point out later in the post, you can tell a story in which the increase is opioid deaths is a good thing and you can tell a story where it’s a bad thing. You seem to suggest in your post that the Washington Post article doesn’t give any evidence either way for these two stories, so what article content are you referring to that suggests that the latter story is more likely?

    (Also, there’s recently been a large increase in the suicide rate, particularly among opioid users. Isn’t that circumstantial evidence for the latter interpretation?)

  6. 6 6 nobody.really

    There is more to life than life expectancy.

    Q: What is a Klingon’s life expectancy?

    A: Zero. They have disciplined themselves not expect to live.

  7. 7 7 nobody.really

    If you read further down the Washington Post story, it reveals that life expectancy has been declining due to the number of people trying to get Google Assistant to play the next song and inadvertently crashing into light posts.

    At least, I think it is done inadvertently. But perhaps the statistics merely reveal that wresting with Google Assistant has become so rewarding that people willingly sacrifice their lives for the pleasure of doing so. You know–revealed preference and all that.

  8. 8 8 nobody.really

    More seriously:

    The Peterson-Kaiser Health System Tracker reports the following average life expediencies for 2017:

    Japan: 84.1 years and rising
    France: 82.4 and rising
    Industrialized nations overall: 82.2 years and rising
    Canada: 81.9 and rising
    US: 78.9–and falling

    But wait–I’m lying. The US reached average life expectancy of 78.9 years back in 2014–and has been DECLINING EVER SINCE, now falling to 78.6. American exceptionalism on display.

    Still, striking a blow for national solidarity, we can report that life expectancy is declining for men, women, white, black, Hispanic, and Native American/Alaskan Native populations simultaneously. In Tom Lehrer’s immortal mortal words, we truly all go together when we go.

    Ok, I’m lying again. The rich seem to be getting older; death comes ever more quickly to the poor.

    Who wants to regress data on rates of life expectancy decline on voting patterns? Are voters who swing to Trump disproportionately likely to live in places where life expectancy is declining fastest (e.g., Ohio, Kentucky)?

    Another fun fact: Rural black people have received opiod prescriptions at a lower rate than rural white people, which may have reduced their death toll by 14,000. Can we somehow stuff that fact into a rational choice model?

  9. 9 9 Rob Rawlings

    If space travel became widely commercially available then I think people would sign up for it even if they knew it carried a relatively high risk of death. Space travel would tend to decrease life expectancy even while increasing consumer utility. This would be an innovation worth celebrating

    But if companies offering space travel gave monetary inducements to doctors to recommend space travel to their patients by correctly stating the benefits but lying about the risks then societal utility (and life expectancy) may be reduced even after this innovations in space travel.

    In a free market for healthcare such dodgy doctoring may be eliminated by market forces. But in a government-controlled system can we be so sure ?

  10. 10 10 Steve Landsburg

    Dave (#1): I think the issue is that people underestimate the power of opioid addiction.

    I think you have to be very careful about what this means. Some people have (for genetic and other reasons) a very high probability of becoming addicted (conditional on use) and others a very low probability. I think you’re right that people don’t know their own probabilities — but they probably do have a pretty good estimate of the probability distribution of probabilities — e.g. that, say, 20% of people have a 10% chance of addiction, etc.

    And, assuming nobody else has a better estimate of your personal probabilities than you do, there’s no reason to think anyone can make better choices for you than you can make for yourself. So yes, I think it is reasonable to believe that there are lots of people saying “If only I had known that I had a high propensity to become addicted, I never would have started”. It can still be true that these people made the right ex ante choice, given the information available — and this means that on average, people making those choices are better off than if they hadn’t.

  11. 11 11 Steve Landsburg

    Rob Rawlings (#9) (and Harold (#2)): I love all parts of your space travel analogy in principle. In practice, I am skeptical that lies about the addictiveness of opioids had much to do with the increase in recreational use, but of course I could be wrong.

  12. 12 12 Steve Landsburg

    Ted (#5): As you point out later in the post, you can tell a story in which the increase is opioid deaths is a good thing and you can tell a story where it’s a bad thing.

    Be careful about shades of meaning here — increased voluntary and informed opioid use (packaged with the concomitant deaths) is unambiguously a good thing. It can be caused by either a good thing (increased wealth, increased leisure time, better opiates, etc) or by a bad thing (feelings of despair, etc). But taking the ambient conditions as given, the fall in life expectancy is a symptom of an unambiguously good thing.

  13. 13 13 joe henry

    You have joined some of the greats.

    Jackie Onassis, “If I had known I was going to die this young I wouldn’t have wasted so much time jogging.”

    Patrick Henry, “Give me pleasure! Or give me death.”

    Descartes, “I drink! Therefore I am.”

    But reading all things Steve does not come without a price. I went to the local cigar shop and, as I went to grab my usual $2 “lawn mower” cigar, I asked myself, “Is this worth dying for?” I put it down and reached for the $15 Monte Christo.

    As I went to order my usual cheap house bourbon I again asked myself, “Is this worth dying for?” I then heard myself asking for Basil Hayden, double.

    Ibo, ergo sum.

  14. 14 14 nobody.really

    @8: [S]triking a blow for national solidarity, we can report that life expectancy is declining for men, women, white, black, Hispanic, and Native American/Alaskan Native populations simultaneously. In Tom Lehrer’s immortal mortal words, we truly all go together when we go.

    Ok, I’m lying again. The rich seem to be getting older; death comes ever more quickly to the poor.

    Paul Krugman elaborates:

    In 1990, today’s red and blue states had almost the same life expectancy. Since then, however, life expectancy in Clinton states has risen more or less in line with other advanced countries, compared with almost no gain in Trump country. At this point, blue-state residents can expect to live more than four years longer than their red-state counterparts.

    Is this all about deaths of despair in the eastern heartland [“the South”]? No. Consider our four most populous states. In 1990, Texas and Florida had higher life expectancy than New York and almost matched California; today, they’re far behind.

    What explains the divergence? Public policy certainly plays some role, especially in recent years, as blue states expanded Medicaid and drastically reduced the number of uninsured, while most red states didn’t. The growing gap in educational levels has also surely played a role: Better-educated people tend to be healthier than the less educated.

    So perhaps greater opiod deaths reflect a laudatory expansion of choice. But if so, it appears to be an expansion in the choice of inferior goods, ‘cuz it doesn’t seem to be a choice especially prized by richer US citizens. Or Canadians of any demographic. Or Europeans. Or Japanese.

    At what point does the statement “The differences in outcomes are explained by exogenous differences in taste and preferences” become a simple deus ex machina excuse?

  15. 15 15 AJ

    Steve writes: “Be careful about shades of meaning here — increased voluntary and informed opioid use (packaged with the concomitant deaths) is unambiguously a good thing.”

    What makes this unambiguously a good thing? Is it always the case that increased voluntary and informed use of X is unambiguously a good thing or can it be possible that eventually one reaches a point where it is no longer an unambiguously good thing and becomes an unambiguously bad thing?

    Your comment doesn’t sound like an economic conclusion but a philosophical one. But I don’t see any reason to accept your philosophical conclusion about increased voluntary and informed use of opioids as an unambiguously good thing.

  16. 16 16 Nathan

    The big problem with opioid deaths is how needless most of them are. Used (or even abused) properly, the death rate from opioid use is very low. Most deaths result from some combination of the following:
    1) Uncertainty of dose. Most addicts get their drugs illegally, which means they have no way of regulating dosage. The heroin they get one week might be cut less than usual, or dosed with something far more powerful like fentanyl.
    2) Polydrug usage. Opioids are far more dangerous when mixed with alcohol, methamphetamine, and other drugs.
    3) Restarting usage after abstinence. Most addicts are unaware that their tolerance levels reset after prolonged abstinence. If you don’t use for a few months and then start again at your regular dosage, you have a high chance of OD’ing. The most common cause of forced abstinence for a few months? Incarceration.

    There’s not much we can do about 2, but 1 and 3 are the direct result of our current legal system, not the inherent property of opioids. It is therefore perfectly appropriate to voice concern about an increase in opioid deaths.

    To draw the analogy, if there’s an increase in deaths from skydiving simply because more people are taking up skydiving as a hobby, we should not be concerned. However, if most of those deaths result from a government policy that forbids skydivers from having a backup parachute, that’s a concern. It’s still a concern even if the government policy is unchanged from before, because the effects of the bad policy are now amplified due to more skydivers.

    Unfortunately, most conversation about opioid deaths revolves around doubling down on our current bad policies, but that’s another topic entirely.

  17. 17 17 Henri Hein

    Is this all about deaths of despair in the eastern heartland [“the South”]? No. Consider our four most populous states. In 1990, Texas and Florida had higher life expectancy than New York and almost matched California; today, they’re far behind.

    If Krugman is making the error he seems to be making, it’s really disappointing. The demographics in the four states he mentions are completely different. The population in Texas has grown 50% and the immigrant population there has almost doubled in that time-frame. FL is similar, whereas the native-born population in NY is not growing and its immigrant population is growing much slower.
    Maybe his conclusions are correct, but you cannot make raw comparisons like that without adjusting for demographics. The same holds true overall when making international comparisons. Japan has a much more static population than the US and you cannot making direct comparisons in health outcomes and life expectancies between them.

  18. 18 18 F. E. Guerra-Pujol

    I like Nathan’s skydiving illustration, but given the problems of peer pressure, hyberbolic discouting, and ignorance, are there any examples of truly “voluntary, informed choices”?

  19. 19 19 Harold

    “ob Rawlings (#9) (and Harold (#2)): .. In practice, I am skeptical that lies about the addictiveness of opioids had much to do with the increase in recreational use, but of course I could be wrong”

    Your post does not mention “recreational.”

    Do you have evidence that all the deaths are due to people starting using opiates for recreational purposes? Specifically, not for pain relief?
    From my previous post on this:

    “Approximately 91.8 million adults aged 18 or older were past year users of prescription pain relievers in 2015, representing more than one-third (37.8 percent) of the adult population.” That sounds a staggeringly large number to me, although it includes all prescription painkillers, most if not all are opiates I think.

    Report here: https://www.samhsa.gov/data/sites/default/files/report_3210/ShortReport-3210.html

    Whilst many of these people will not suffer from addiction, given that many over-use and the prescription limits themselves are based on bad science some sort of crisis seems almost inevitable.

    The study I quoted supports the fact that most people use their medication correctly, but contradicts the “small sub-set.” In 2015, 12.5% of pain relief users misused. That is 11.5 million people. This is a big sub-set. That was 4 years ago, and I don’t think it has got better since then.

    We have 11.5 million people misusing their prescription painkillers. Few of these stared recreationally. Some of these are going to go on to die because of their misuse. How many I am not sure, but I also not sure that you know either.

    That also ignores the fact that people who decided to use recreationally were unaware of the risks because granny was prescribed this by her doctor and told it was not very addictive. I can imagine many people who would have avoided heroin because they knew it was addictive, trying prescription opiates because they believed it was not so addictive.

  20. 20 20 Harold

    ” The population in Texas has grown 50% and the immigrant population there has almost doubled in that time-frame. FL is similar, whereas the native-born population in NY is not growing and its immigrant population is growing much slower.”
    Are not most immigrants relatively young and healthy? Why shouldn’t immigration increase life expectancy? Making your demographic argument self defeating.

  21. 21 21 Z

    @Harold #19

    Do you have evidence that all the deaths are due to people starting using opiates for recreational purposes? Specifically, not for pain relief?

    Why does that matter?

    Ignoring the more unsavory characters, doctors have limits to the number of opioid prescriptions they can give a patient. Once this threshold is reached, someone who abuses their medication is actively choosing to obtain illicit opioids to relieve their pain, thus choosing to risk a shorter pain-free life over a prolonged painful one. Part of the over-prescribing part of the “Opioid Epidemic” was because over 1/3 of Americans were in chronic pain. Why is this not something to celebrate?

  22. 22 22 Harold

    Why does that matter? Because SL said the lies about addictiveness had little to do with increase in recreational use.

    If a patient is lied to about the sddictiveness they cannot make an informed choice about whether the pain relief is worth beooming addicted.

  23. 23 23 Z

    So you’re still focusing on the “Patients were lied to.” argument. That’s a myth.

    The main causes of increases in opioid related deaths are from heroin and synthetic opioid consumption.

    https://www.cdc.gov/drugoverdose/images/data/opioid_deaths_multicolor.gif

    Heroin is illegal and can only be obtained through the black market. And synthetic opioids, which are recently the cause of most opioid-related deaths, are also primarily obtained from illicit sources and not from doctor’s prescriptions. In fact, there doesn’t seem to be any association between state-by-state fetanyl related deaths and prescription rates.

    https://www.cdc.gov/drugoverdose/data/fentanyl.html

  24. 24 24 Steve Landsburg

    Nathan (#16):

    if there’s an increase in deaths from skydiving simply because more people are taking up skydiving as a hobby, we should not be concerned.

    Yes, great analogy so far.

    However, if most of those deaths result from a government policy that forbids skydivers from having a backup parachute, that’s a concern. It’s still a concern even if the government policy is unchanged from before, because the effects of the bad policy are now amplified due to more skydivers.

    Yes, it’s a concern. But it’s still evidence that the world has gotten better, not worse. People are choosing to skydive even given the bad government.

    Of course it would be better to reform the government, just as it would be better to design a safer parachute. But reforming governments, like designing new parachutes, requires skills and effort and luck that might or might not be forthcoming. Given the current state of parachute design, and given the current state of government, an increase in skydiving deaths is probably a good thing. It’s also a reason to invest more resources in technological and political improvements, but that doesn’t change the fact that it’s a good thing.

  25. 25 25 Henri Hein

    Are not most immigrants relatively young and healthy? Why shouldn’t immigration increase life expectancy? Making your demographic argument self defeating.

    That’s a little strange. You are saying that there are reasons to believe the life expectancy of the new people is different, so therefore we should go ahead and compare the old and new populations anyway.

    Immigrants tend to be young. I don’t know if they tend to be healthy, but the life expectancy of a young person is lower than that of an older person. They tend to come from countries with lower life expectancies than the destination country. They tend to be from lower income quintiles. There are many reasons to think the new population is different than the old one, making direct comparisons meaningless.

  26. 26 26 Harold

    #23 Z
    It is not a myth, but a fact. Patients and doctors were lied to. I will grant you that I don’t have details of exactly how much this affects the number of deaths, so it would be a myth to blame all deaths on this.

    If most fatalities are from illegal sources that does not undermine my argument if the addiction started from prescription drugs. I know, again anecdotally, that many people start on the illegal route by topping up their prescription because they do not get the 12 hours they were promised. I know from data that in 2015 11.5 million people mis-used their prescription drugs. These numbers are likely to be higher for later years. It is not too much of a stretch that people are dying due to obtaining illegal drugs to feed the habit they acquired from prescription drugs, and they could not make an informed choice because the data was misrepresented.

    However, the CDC finding that deaths do not correlate with prescription rates suggests that the direct cause of most deaths is not from the mechanism I described just above, so thank you for the information. I will look further.

    This is not the only argument I have made here.

  27. 27 27 Z

    @Harold #26

    Pardon…I should’ve prefaced “a myth” with “mostly”. I don’t doubt some were lied to (e.g. Purdue Pharma), but to say or imply that it’s a main driver or a significant cause of the increase in opioid related deaths and/or addiction isn’t correct. (And just so we’re clear, you’re argument would carry more weight if we’re talking, say, 1995 to the mid/late 2000s, but my understanding is that we’re talking about a more recent timeline — i.e. no more than a decade).

    This is not the only argument I have made here.

    I’m aware, but I’ve time to only respond to what I thought is your strongest.

  28. 28 28 Will

    Is the assumption that the obese and drug addicted are making rational choices about calories or narcotics necessary to accept the argument here? As others have said, that seems like a pretty dubious assertion in its own right.

    Setting that aside, the claim seems to be that increased consumption of a product is necessarily a good thing, even if the product is dangerous or unhealthy.

    The most extreme example of that might be a product that is intended to kill the user. We could imagine a roller coaster that is designed to kill the rider in an enjoyable way. If people are only ever making rational decisions, then yes, we should celebrate use of the roller coaster. If we consider that there is a serious possibility that the riders are not acting rationally — or are even capable of making irrational decisions — it quickly becomes a lot harder to justify celebration.

  29. 29 29 Harold

    Z #27. I agree with you in part.
    I looked at this source
    https://www.drugabuse.gov/drugs-abuse/opioids/opioid-overdose-crisis
    This is reference by many newspapers.
    Some claims from it.
    1) Roughly 21 to 29 percent of patients prescribed opioids for chronic pain misuse them.6
    2) Between 8 and 12 percent develop an opioid use disorder.6
    3) An estimated 4 to 6 percent who misuse prescription opioids transition to heroin.7–9
    4)About 80 percent of people who use heroin first misused prescription opioids.7

    I have some significant issues with the way this is put.
    1) seems fine. The source reliably claims that 25% of people misuse their prescribed painkillers and this is roughly consistent with CDC studies I have read. Misuse covers a wide spectrum, including taking too many. Misuse is different from abuse, so it would be very wrong to conclude that 25% of people abuse their prescribed opiates.

    2) The abstract of the source reports that “Rates of addiction averaged between 8% and 12%”. The study was looking at chronic pain treatment, so this is medical use and seems an accurate summary.

    3 and 4) This is where it falls apart. The key here is that it looks at people who “misuse” prescription drugs, not people who misuse the drugs they are prescribed. Some of the studies specifically look only at non medical use. It would be more appropriate to say “abuse” rather than “misuse” here. The conflation of the terms is misleading and disingenuous. Reference 7 does not use the term misuse at all, but “abuse” occurs 41 times. The impression given is that the mis-users from points 1 and 2 are the same mis-users in points 3 and 4, but this is very much not the case. 1 and 2 look at mis-use, 3 and 4 look at abuse.

    So yes, they are selling a misleading story.

    However, points 1 and 2 are compelling in their own right. I think we can conclude that the “over prescription” of opioids caused very significant problem of addiction among initially legitimate users. 10% of 100 million+ of people is a lot of addicts and these are likely to suffer quite a bit. This is a serious issue in its own right, but we cannot conclude (from this data) that the majority of deaths originated with this group. It is plausible that the majority of deaths occur in people who started using prescription drugs they were not prescribed.

    You say that this argument of mine seemed the strongest. I don’t agree, but it is the easiest. If I could substantiate that the people who were dying were the same people who were effectively mis-sold the drugs, it would be an easy argument. It seems that I cannot do that. It may be the case, but we don’t know from the data I have seen. We do know that at least some of the deaths were from this origin.

    This leaves us with the more complex arguments.
    1) Rational choice theory is flawed.
    2) It is particularly flawed in this case because the wide prescription of “safe” opiates causes people to underestimate the addictive power of the drugs.

    This is supported by the information in points 3 and 4 above. Many people moved from prescription drugs to heroin, not the other way around. They may not have started at all if heroin was the only choice, because they knew heroin was addictive, but they did not appreciate that prescription drugs were equally addictive.

    To summarise, I reject Landsurg’s conclusion that people using addictive drugs such as heroin is a good thing because they have chosen to do so. I think this applies even more to abuse of prescription drugs, because at least some of the abusers were misled.

  30. 30 30 Harold

    Will #28. the closest thing to your rollercoaster that i can think of is Dignitas.

    It does not claim to end life in an enjoyable way, but in a dignified and suffering-free way.

    We may celebrate an increase in Dignitas users, because we see it as reducing suffering, or simply as people exercising their choice to die.

    We may not celebrate a hypothetical Dignitas that could simply cause pain free death in anyone who wished it. We know that many attempted suicides go on to have reasonable lives, and are glad that they failed. Some say it is a cry for help. Say we had an effective suicide mechanism that only required someone to wish for death and it would be delivered. Landsburg (perhaps) would claim that this is a great boon, and represents at last a mechanism for effecting peoples’ choices. I see a problem with this, as it does not allow for future change of mind and does not acknowledge that people can be ill when they make such choices.

    The more I think about it, the more it seems a good test of choice theory. Steve, would you claim that such a mechanism would be a great boon?

  31. 31 31 iceman

    AJ 15 – I believe the reason for saying choices that are voluntary and informed (both being debated here of course) are unambiguously good is the presumption that no one else can know your preferences, goals, utility function, risk aversion etc. better than you. (So some economics there.) The alternative would be “saving people from themselves” paternalism. (Maybe a little philosophy there.) To me the point of interest of the post is whether / how this position is negated or modified even by “avoidable” deaths and hindsight regrets.

    Henri Hein 17&25 – well said. This seems like a general problem with international comparisons of healthcare outcomes e.g. infant mortality as well. The US is both more diverse and probably more regionalized in terms of policy than other developed nations. I’ve long wondered why no one ever compares (as far as I know) say northern states to northern countries etc.

  32. 32 32 Bob Murphy

    Steve, very provocative as always. But can you help me to understand exactly what your claim is? Suppose all I told you was that the number of people who hanged themselves in 2019 was up 400% from 2018. Would that be enough for you to celebrate?

    If not, what if you learned that the price of rope had fallen in half. Would that make you confident in celebrating?

  33. 33 33 Adam Wilhelm

    This theory of viewing increased mortality rates as a positive thing neglects the lack of knowledge available to opioid users. Oxycontin supplier Purdue Pharma faced litigations for falsifying data that minimized the addiction potential and negative side-effects associated with the drug. The fact is, many who became addicted did not realize the addiction potential. Even doctors who were prescribing did not realize a patient could become addicted so quickly. I suggest you analyze the situation from a historical perspective. Relying on rational choice theory really misses the mark here.

  34. 34 34 GregS

    The government actually collects statistics on prescription opioid abuse and addiction, and these data don’t show any obvious increase in recreational opioid use *or* opioid addiction. See this SAMHSA report:
    https://www.samhsa.gov/data/sites/default/files/NSDUH-FRR1-2014/NSDUH-FRR1-2014.pdf

    Page 11 shows past month nonmedical use of pain relievers. It’s flat from 2002 to 2014. Page 30 shows pain reliever use disorders within the past year, which is flat over the same period. This is a period during which prescriptions were increasing dramatically, and deaths attributed to prescription opioids were rising. This all happened *without* creating a new class of opioid addicts, apparently. (The number of people misusing prescription opioids in 2014 was 4.3 million, not the 11.5 million people Harold suggests upthread. The number has trended down since 2014, btw.)
    The Monitoring the Future report, which uses a different data set, shows similar trends; see page 30 of this report:
    https://files.eric.ed.gov/fulltext/ED594190.pdf

    The attempts to blame opioid manufacturers for the crisis are overblown and inappropriate.

    Studies that try to measure the “risk of addiction” for medical opioid users typically arrive at a very low number, in the range of less than a percent to a few percent. See the paper What Percentage of Chronic Nonmalignant Pain Patients Exposed to Chronic Opioid Analgesic Therapy Develop Abuse/Addiction and/or Aberrant Drug-Related Behaviors? A Structured Evidence-Based Review by Fishbain et. al. We should keep in mind that addiction rates and misuse/abuse rates are very low, and the SAMHSA data bears this out. Some of the comments upthread suggest that large fractions of the population were sleep-walking into addiction by taking a typical course of prescription opioids or taking them for chronic pain. Apparently this isn’t the case.
    Drug overdose death rates (that’s dividing by population, not the raw number of deaths) have been rising exponentially since the late 1970s. See this paper: https://science.sciencemag.org/content/361/6408/eaau1184

    Note the almost perfect exponential curve (r-squared = 0.99). So the fact that prescription opioids were more available in the late 90s’ and 00’s probably has something to do with the fact that opioids were being used by drug addicts during this period, but overall drug-related deaths had already tripled between 1979 and 1996 (96 being the year that Purdue introduced Oxycontin). The rising trend in drug overdoses pre-dates the aggressive marketing by opioid manufacturers, and it continued post-2010, when opioid prescriptions leveled off and abuse-resistant Oxycontin was put on the market (the new formulation couldn’t be crushed and snorted or injected). I think this tells a story similar to the one Steve is telling. A society that is growing richer (economic growth is also an exponential process) gives people more resources to indulge their vices. Sadly, some people use this expanded opportunity set to effectively kill themselves. I certainly don’t think that all of these deaths were inevitable. The heroin and fentanyl deaths are definitely a black market effect. There would be far fewer of these deaths if drug users knew what they were buying, which would be the case in a legal market for these drugs.

    Upthread, Nathan points out that most overdoses are actually polydrug use. It’s rare for people to actually overdose on prescription opioids alone. It is the combination of opioids and benzodiazepines (or opioids and alcohol or antidepressants) that causes the typical drug poisoning. This is important in assessing the “risk of addiction”. People don’t adequately appreciate the difference between physical dependence and addiction. You can go through a normal course of opioids and end up with some degree of physical dependence, and ceasing the will induce withdrawal. It’s reasonable to talk about the “risk of physical dependence.” But “addiction” is a different animal. It’s a series of deliberate behaviors that aren’t obvious consequences of the drug’s pharmacology. If an opioid prescription causes physical dependence, it might be reasonable to claim that the pharmacology of the drug *caused* someone to hoard pills or engage in “drug-seeking” behavior from their doctor. It’s a harder sell to blame opioid dependence on benzo or alcohol use. Nothing about the pharmacology of opioids *causes* the addict to use these other drugs. Any talk about the “risk of addiction” is misleading, because it confuses something that’s random (“Will my physiology *cause* me to develop a physical dependence to this drug?”) with something that’s much more deliberate.

  35. 35 35 Joe Elwell

    Steve, you seem to have become untethered to the realities of these behaviors. Speaking of suicide, I almost attempted it myself about 9 years ago in a moment of desperation. My plan was interrupted because I’d misplaced some keys I needed to execute it.

    Would my suicide have eliminated my suffering, so others should be glad for that? Or are they (and I) happier that in the subsequent months and years I got the help that’s allowed me to be leading the happiest life I’ve ever lived?

    As Buddhists will tell you – happiness is not the realization of your desires (or the elimination of your pain). Your article seems to differ from that assessment.

  36. 36 36 Samuel

    Steven – Responding to your response to Dave:

    “Dave (#1): I think the issue is that people underestimate the power of opioid addiction.”

    re: “And, assuming nobody else has a better estimate of your personal probabilities than you do, there’s no reason to think anyone can make better choices for you than you can make for yourself.”

    You might be right that you have the most readily available information to estimate your own personal probabilities, but it’s not always the case that an individual is conscious or aware of those probabilities – most opioid addicts probably aren’t in a mind-state to evaluate every action or inaction for probability of positive outcome. Your argument assumes that other beings aren’t able to help one another, specifically as it relates to bringing awareness and consciousness to one’s own capabilities, probabilities, attributes, and shortcomings. This is an isolationist line of thinking that breeds cultural inequity in an extrapolated sociological context. Being sensitive to each others existence is among man’s greatest tools for growth and sustainability.

    re: “and this means that on average, people making those choices are better off than if they hadn’t.”

    How do you leap to this conclusion? Just because it can be true that an individual can make the right ex ante choice doesn’t mean that on average people making this choice are better off. An analogy here would be a circumstance where I use my own judgement about my own intelligence to inform whether I’ll be able to remember my locker code tomorrow when I return. Just because I decided that my memory is sufficient, and ended up being right to pursue that decision (independent of others’ judgements), doesn’t mean that on average: relying on one’s own judgements will always result in the most optimal outcome. It’s particularly okay for an opioid addict to acknowledge this and seek outlets where individuals like therapists can provide guidance towards optimal outcomes. We should be emboldening mindful decision making with or without external help, not just championing individual choice over all else.

  37. 37 37 vmsmith

    First of all, I don’t have a particular concern about Americans who are dying from opioids. To the extent they get involved in public discourse, e.g., vote, their values appear to be at such extreme odds with with mine that if they want to kill themselves, that’s their business.

    Having said that, to me this article represents why utility theory is completely bankrupt and should be tossed.

    By way of background, I was raised in an inner city and in my youth got involved with pretty much every drug known to God and man. Indeed, I got myself mildly addicted to morphine at several points. I had about a dozen acquaintances die: from overdoses, from violence in drug deals gone bad, in one case from shooting up rat poison that was sold as heroin, and later in life, from Hepatitis C that was passed with dirty needles. I knew of one or two families destroyed because some junkie family member stole money, jewelry, electronics, and anything else of value to buy smack. There were no informed, rational choices being made in that environment. It’s hard to even describe how it happens, and what it becomes. But I can assure you that the term “rational” doesn’t even begin to apply to a mind that is fixated on opioids.

    That was about 50 years ago. Cut to…

    …the second decade of the 21st century. I was teaching probability and decision analysis to college sophomore and junior business majors. Out of any given class of 30 or so, perhaps 7 – 8 could grasp the basic notion of expected value. Fewer still could understand concepts like Hurwicz criterion, minimax, or any of the more formal methods of making decisions. These were college kids (and I was a pretty good teacher).

    Again, I couldn’t care less about those who are killing themselves with opioids. But to hear someone who is utterly clueless about what that world is like try to package it up into some neat academic theory the rudiments of which even most college kids — in my experience — have difficulty with…well, again, this is why utility theory is completely bankrupt and should be tossed from any real discussion on public policy.

  38. 38 38 Josh H

    To be quite honest, I’m a bit sympathetic to Steve’s philosophy here, but I will just say this philosophy (as many do)gets messy when applied to the real world of today.

    Here’s a case where I think Steve might be a bit off. Suppose we live in an imaginary world where half the people who use opioids get addicted but everyone who uses them does so only because they’re in excruciating pain and would rather be dead than be in that pain. So nobody just picks up an opioid and says “oh this sounds fun.” It’s all pain based. In that world, if horrible pain starts to increase across the population, then opioid addiction will increase and deaths from opioid addiction. Now, it’s true that opioids may still be a good thing! But the fact that life expectancy is falling would still be bad! More pain is bad, whether you have opioids to help with it or not.

  39. 39 39 chuck martel

    According to the Pew Research Center, in March of 2006 66% of Americans owned a mobile phone. Just a month later 73% of the population owned one. The number continued to rise after that. It seems very strange that the popularity of cell phones and their contribution to the “Great Recession” has never been fully explored. It’s my contention that of the three serious monthly payments consumers needed to make: mortgage or rent, car payments and cell phone contract payments, it was more likely that people would fail to make a mortgage payment than a car payment or cell phone payment when in a financial bind. No one wanted to give up their phone and they could sleep in their car. The house was expendable.

  40. 40 40 Harold

    #34 GregS
    Your post is informative. I have altered my views somewhat during research for this thread, not least because the conflation of abuse and misuse in some of the sources creates a misleading impression.

    These claims in particular from my earlier comment
    ) Roughly 21 to 29 percent of patients prescribed opioids for chronic pain misuse them.6

    3) An estimated 4 to 6 percent who misuse prescription opioids transition to heroin.7–9

    The first is talking about medical misuse. The second is talking about abuse. They are completely different populations and the source does not make that clear, in fact it seems to go out of its way to conflate the two, using “misuse” when the original source for point 3 used “abuse.”

    The misuse figure for prescription drugs looks reasonably robust, so I am not sure if your 4.3 million and my 11.5 million are referring to the same thing. Mine is people who are prescribed drugs but don’t use them as directed.

    I have not changed my opinion of the idea that the increase in deaths should be celebrated.

  41. 41 41 Harold

    One other point, the medical issues are a scandal in themselves, even if they are not the main cause of the dramatic increase in opioid related deaths.

  42. 42 42 joe henry

    As usual Steve brings an element of fun to a complex issue. Some people just love life. Others need a little help.

    Some are just plain wired different. Others are pretty much helpless victims of a hopeless environment.

    In either case therapy and prescription medications can help. Lacking this some simply “self-medicate”.

    (A problem with self-medicating is dealing with the black market. While you can never be sure of what you’re actually getting you can be assured they will try to up-sell you into something even more addicting.)

    I’ve long thought that if something is fun then it is governments’ role to either outlaw it or tax the crap out of it. If there ever was a magic bubbly water that re-franchised the disenchanted and the disengaged you would never see it in NYS.

  43. 43 43 Bob

    The percentage of addicts is likely constant (and it’s the same group of people), so the increase in deaths probably represents a shift to opioids. This supports the idea that addicts have found a better “drug of choice.” It also probably indicates, however, that addicts as a group are not staying alive as well with the current opioids they are using. It could be that the experienced addicts are doing about the same, but that there is a spate of new ones (that switched over from say, alcohol) that are dying more due to lack of experience. There could also be something about the drugs (e.g., they aren’t as consistent in terms of strength which leads to more ODs) or users’ knowledge (e.g., people aren’t aware that mixing benzodiazepines with opiates increases the risk of OD) that has changed. Hopefully, things will settle down as information gets out, as it’s probably a small percentage of people who died from opiate ODs that intended to.

  44. 44 44 Bob

    Oh, and I can imagine an alternative reality without prohibition where drugs and drug use would be safer, and many of those people would still be alive. So, it’s not good news; it’s another example of the negative consequences of government policy.

  1. 1 Nasty, Brutish, and Long, by David Henderson -
  2. 2 Nasty, Brutish, and Long, by David Henderson – Viral News Connection
  3. 3 Nasty, Brutish, and Long, by David Henderson – CNB Reports
  4. 4 Nasty, Brutish, and Long, by David Henderson – Finance Market House
  5. 5 Nasty, Brutish, and Long, by David Henderson | Share Market Pro

Leave a Reply