Bryan Caplan writes an argument against Universal Basic Income:

To sum up, the majority of the people not in the labor force spend most of their time on screens, and this is bad for their health and their wellness. If more people were to exit the labor force, many of them would presumably behave the same way, therefore UBI is a bad idea.

I am not sure of the validity of the object-level argument (the causal link could be in the other way: maybe people who spend their entire day scrolling their Facebook feed are more likely to become NEET, and not the reverse).

However, on a more abstract level, this hit me as an uncommonly Puritan argument. If people did not have the need to work, many of them would end up living a miserable life, pursuing short-term petty pleasures. This rings true to me, it somewhat resonates with my "high-level generators of disagreement". So my brain is trying to find ways to defend this argument.

Is it reasonable to think that, if relieved from the necessity to work, the majority of people would just procastinate all day? But I guess it is possible to conceive a model in which lotuses tend to trap men, and if you decrease the incentive to getting things done (in the example of this article, if you decrease the reward from green circles), more people will spend more times eating lotuses. There are many social incentives to work, which would not disappear if we remove the economic incentive to work; but it is also true that social rewards are often easy to pursue on social network. 

So we should expect an increase the mean time that the population spends on activities that we might judge as "lotuses". But on what lotuses in particular? Is procastinating on the Internet the most brain-gripping short-term reward of this age, or there are other competing lotuses?


New Answer
Ask Related Question
New Comment

1 Answers

Drugs, alcohol and porn. How many people have a preexisting tendency towards overuse of these substances that is kept in check by the need to get up on time for work and be reasonably productive and presentable at work? This is no limit to the amount of time an addicted person can spend pursuing their addiction.

You could extend this to other potentially addictive activities, like shopping, video games, and social media.

As someone that has had surveillance capabilities at work, let me assure you that a huge chunk of the workforce is drunk, high, and either jerking off or fucking. You only have to be slightly above terrible and causing few issues to remain employed (at least where I am. We have laws that make firing difficult).

The fundamental question is whether society has an interest or onus in policing sin. If people want to kick their addictions that's one thing, but it's quite another to say here's some busywork because we don't want you touching yourself too much.

If people aren't causing problems, and aren't asking for addiction intervention, then why shouldn't their agency and privacy be respected?

4ForensicOceanography12di agree on this principle: making people work to avoid them falling in vice seems definitely an exceedingly patronizing position for the State. However, I can think to two possible answers to your question: - many addicted people are not really happy with their addiction, and do not ask for help for pride or because they are costantly believing that are quitting (like the stereotypical smoker that decides avery day to quit smoking). So it would be a net utilitarian harm if more people were addicted. - maybe the society could have a legitimate interest in preventing the spreading of addictions. Maybe there are no problems if 10% is high, but there could be trouble on the streets if 80% of the populations was high.
2Stuart Anderson12dMany people are perfectly fine with their addictions. Barring harms to others It is the consenting individual that should decide whether they have a problem or not, and what they want to do about it. The same argument against addiction also applies to other things. I would offer obesity as an example. If the state can control what you put in your body then what greater utility than stopping you from getting fat? Society isn't a machine that is meant to promote optimum human efficiency.
1ForensicOceanography12dWhile I agree that the state can not prevent you from becoming obese or drunk (mush less sure about, say, heroine), I think it is legitimate to apply economic incentives to decrease the expected number of people engaging in a given activity. Many states apply taxes on tobacco and sugar, and there are advertising and sale restrictions on cigarettes and alcohol.
1Stuart Anderson11dAnything you tax you have little incentive to decrease. You want to stop people from doing something? Make it legal and boring/embarrassing.
3gjm11dIf you (= the government) tax something, then you (i.e., the government) have little incentive to decrease it, but the people doing it have a clear financial incentive to decrease it. So a government that puts a big tax on whatever-it-is gives up its own ability (or at least motivation) to decrease the thing directly, while giving the people who actually do it an incentive to do so. It's reasonable, at least in some cases, to hope that those people are better placed to decrease the thing than the government is.
7Stuart Anderson10dExcept we know that's not how addiction works for individuals. We also know that increasing the expense of drugs generally moves the addiction rather than quells it (for example, the oxy to heroin pipeline). As I stated, people still smoke here, and when they can't smoke tobacco for financial reasons the most common alternative is marijuana. The results of actions here are measurable. We know this doesn't work, and we know the reason for that is that people want to get intoxicated so badly that they'll huff solvents. If you know that you cannot stop a behaviour but you might be able to reduce the harm involved then doesn't that tell you the experiment you ought to be trying?
3gjm10dFirst of all, this isn't only about literal drug addictions; other examples mentioned in the subthread are eating too much sugar (which surely has some addiction-like qualities but I wouldn't bet heavily that it works just the same way) and advertising (where the people who might be deterred by taxation are the advertisers, whose relationship to advertising is surely not one of addiction). Even for drug addiction, the relevant taxes might reduce profits for the sellers who are generally not addicted, whether or not they affect the behaviour of the buyers. Second, I dispute that we "know" that increasing incentives not to take some addictive drug consistently does nothing to reduce addictive-drug-taking. Maybe I'm just not aware of some relevant evidence; would you like to give me a pointer to, say, the two things that you think are the strongest evidence for that proposition? (As opposed to weaker ones like "sometimes if you deter people from using one drug they switch to another", which I'm sure is true.) For the avoidance of doubt, I entirely agree that governments should be trying much harder to reduce the harm done by drug addiction and trying much less hard to punish people for getting addicted to drugs. I'm just not at all convinced by one specific argument you made, namely that taxing "lotuses" can't reduce how much they're used because it decreases the government's incentive to do it.
1Stuart Anderson6dIsn't a lotus by definition something that is chosen in preference to utility to society or state? It shouldn't matter whether it is a formal addiction or not, merely whether the individual chooses it and society or state doesn't want them to. The government is offering no incentives, only disincentives. As long as the lotus is preferred culture and state can only punish and threaten, and that only gains them grudging temporary compliance at best. I'm not going to say that interventions have no effect, but I will say that the effect of taxation is poor on deterring drug use, and that taxation represents a conflict of interest for the government in a matter of public health. Think of some lotuses you use, now imagine the government decided to tax one to be point of prohibition in practice. What would you do in that situation? Just say "the state knows best!" and give up your pleasure cold turkey and with no replacements? I seriously doubt that.
3gjm6dYes, "lotus" is broader than "addictive intoxicating drug" (though I'm not sure it's specifically about "utility to society or state"; it might be utility to oneself-thinking-longer-term, for instance). That was exactly my point: you are trying to generalize from something allegedly true of literal drug addiction (that if an addict's drug of preference becomes unavailable or much harder to get, they will switch to another drug rather than stopping being addicted) to all lotuses, even ones that might be quite different from drugs. Disincentives are incentives. If I know that I'll get punched in the face if I do X, that is an incentive to do something other than X instead. If I know that I'll get punched in the face if I don't do X, that is an incentive to do X. You may be right that positive incentives are more effective than negative ones, and it's absolutely reasonable to want to reduce the negative incentives imposed by governments. I've no problem with any of that. But you made stronger claims than those, and those are what I disagreed with. If the government decides that something I enjoy[1] should be taxed into oblivion, then I'll probably do it much less or not at all. I may well do more of something else I enjoy[1] to compensate. If that's something the government is happier for me to do, then they may be pleased with the outcome. I don't think anyone is claiming that carefully chosen taxes will turn people into idealized productivity machines. (How certain is it that if denied one lotus I'll switch to another lotus? Not very, I think. The nearest available substitute for a newly-unvailable lotus may be less lotus-y.) [1] Or "desire", which of course isn't the same thing.
1Stuart Anderson4dDisincentives are not incentives from the perspective of the utility of operant conditioning. If we want to change people's behaviour with operant conditioning that distinction matters.
3gjm4dIncentives are not only about operant conditioning, and what matters is whether a given measure is successful in changing people's behaviour, not whether it's successful specifically by means of operant conditioning. I don't think "incentive" is a standard term in behaviourist psychology at all, but unless I'm very confused -- which is always possible -- operant conditioning absolutely does include responses to the presence or absence of unpleasant stimuli, and also to the absence of a pleasant one that would otherwise have been present. If you're claiming that disincentives can't be effective in changing behaviour, I think you need to provide some actual evidence for that.
1Stuart Anderson3dIf you want active measures to alter behaviour then positive operant conditioning is the gold standard. Passive measures are typically about environmental design. Negative stimulus can work but it doesn't work as well as positive stimulus. It also opens the door to bad behaviour because the attention of the punishment literally becomes a maladaptive positive stimulus. You don't need to put up an academic fence around the subject of animal training. I think incentive and disincentive are reasonable words to use in the context of altering animal behaviour, and I think you know what I mean when I use them.
4gjm3dI wasn't attempting to put up an "academic fence", I was responded to what seemed like your attempt at doing exactly that. I've no objection to the terms "incentive" and "disincentive". Perhaps I misunderstood your intention. Anyway, it seems to me that the claims you're making have got weaker and weaker, to the point at which they don't offer any real argument against (e.g.) taxes on things the government wants people to do less of. You started off by claiming that doing this would be counterproductive because the government's incentives would be wrong. "Anything you tax you have little incentive to decrease." I pointed out that that could be counterbalanced by the fact that taxing things give the people actually doing them an incentive to do them less. You then shifted to claiming that specifically for addictive drugs making one drug less available just makes people switch to another. I pointed out that we aren't talking only about addictive drugs, and questioned whether there's good evidence that the substitution effect you describe is strong and consistent enough to justify the claim that there's no point trying to dissuade people from taking a drug because they'll necessarily switch to another that's just as bad. You ignored my request for evidence, and switched to arguing that taxes give disincentives which are completely different from incentives. I pointed out that disincentives are one variety of incentives. You then switched to saying that they're completely different from the perspective of operant conditioning. I pointed out that human motivation isn't just operant conditioning and that operant conditioning works with disincentives as well as incentives. You switched to claiming that operant conditioning is "the gold standard", whatever exactly that means, and that positive stimuli work better than negative stimuli in operant conditioning. By this point, what you're saying has very little to do with the original question. We shouldn't tax lotus-l
1ForensicOceanography10dI am not really into the studies, but I know that in 1950/1960 virtually everyone smoked (also, if you read books from that period, is it taken as given that everyone smokes), while now it is quite uncommon for a young person in Italy to smoke. I think that also in the USA tobacco consumption rate is plummeting, so why are you saying that it does not work? It may be misleading to conflate all "addictions" together. I can see how this can not work with heroine, but addiction to candies is a different thing.
1[comment deleted]10d
2Viliam11dThis, and even more monstrous when you put it into context that maybe part of the population would spend their time touching themselves, but another part would spend their time doing something amazing... but we will sacrifice the possibility of the latter, just to make 100% sure no one touches themselves too much.
1ForensicOceanography12dAs a slightly unrelated question, I would be very interested to hear if you think that the quality of the work you watch over is somehow affected by the workers being addicted, and how much.
1Stuart Anderson12dIf you can work a day when you're sick then you can probably work a day intoxicated too.
3ForensicOceanography12dI definitely can not work when I am sick. Can I ask what kind of job are you overseeing?
1Stuart Anderson11dI was running the backend of a call centre. I've worked drunk before (not my choice), I've worked ill, sleep deprived, etc. Most jobs aren't hard, and even within that subset most of any given role isn't hard either. This is Pareto territory: 80% of your job doesn't matter that much, so if you can sustain a burst of 1/5th of your capacity you'll be okay for the day. I can guarantee that you can work a day when you're sick. I'm not telling you that you'll enjoy it, that it will be your best work, or that it won't have consequences, but you most certainly can do more than you think you can. If the first thing to go in a given domain of labour is your brain saying I can't then you can. When it is your body giving you pain that is at the point where you think you must stop that means you have 30-50% more left in the tank.
2ForensicOceanography11dMaybe I am making confusion between two claims: A) If it happens that you are sick one day, you can still (with pain) carry out an acceptable amount of work for that day. B) You can work in a decent way, in the long run, while being sick most of the time. Are you saying that (B) is true, or just (A)? I fully concede (A) - I also did it. But (A) does not imply (B). I work as a PhD student (which in Europe is a job: you do not have to attend lessons, but you have to do research), and I am sure that (B) is false for me. Maybe there are jobs for which (A) implies (B), but my intuition is that they are not the majority.
1Stuart Anderson10dAll of this is contingent on what you are being asked to do, and to what standard. How much error can a given task endure before you not working is the optimal strategy? Business (and life) favours completion over perfection. You might have a feel for whether you are underperforming at work but the question is whether others can see that (and especially whether they can quantify it). The vast majority of work is not on the critical path or a showstopper. If your work isn't urgent it can be deferred. People are sick all the time. A third of the population is on antidepressants or other psych meds, and script drug addiction is massive. Work still gets done. In regards to you and B: If you haven't worked at breaking point then you don't know what you're capable of. B isn't long term sustainable, and it will hurt you in some manner (because that's what overwork does), but you can do it.
1ForensicOceanography10dSo you are saying that you can still pretend to do a good work if many people do a work just a bat as yours. This is different from saying that your work is decent. In the town I grew up in, it is common for people to do not work at all (not because they are sick, but because they do not care). They "can" do it in the sense that they do it and they face no consequences - but we all pay the price, for our public services are terrible to nonexistent. Do you think that the performance of a workforce on antidepressants would be the same as the performance of a drunk workforce?
1Stuart Anderson6dI'm saying that most labour simply doesn't matter that much. Most of just about everything is filler. Just because people are employed doesn't mean they care about working hard, or that they are particularly good at their jobs. As I've said elsewhere: a huge portion of much of employment is simply turning up and doing the little that is expected of you. If you want people to work then you have to pay them enough to justify the imposition. One way to do that is to make sure they starve without a wage, but if they aren't starving then you have to compete with whatever welfare they're on. I'm willing to bet your town's average wages are terrible in comparison to what you get doing nothing. Antidepressants are a masking agent in the context of employment. Your staff are taking something to suppress their psychiatric issues. It's no different to someone doing coke to fight off sleep. I wouldn't fire someone for being on antidepressants, but I also wouldn't fire someone with any other kind of drug use. If it doesn't cause problems at work then I don't care and it's none of my business.
1ForensicOceanography6dNo, it is not that the wages are low, it is that they can not be fired (both for legal and for cultural reasons). So they do not risk to lose their wage by not working. To clarify your position, are you saying that if more people were sick/intoxicated then the quality of their work would deteriorate, but this does not really matter because there is sufficient slack in the system and nothing really bad would happen?
1Stuart Anderson4dIf employment is defacto welfare then it will have the same effects welfare does. My position is that everyone is already sick and intoxicated at work and we don't notice or care most of the time. Consider this from the perspective of you having to cover someone's role whilst they're off work. You can probably do it well enough, at least until a better solution is found. That's all you need, even in the worst situations - to get a temporary solution that will get you to a better solution, and then the next better solution, and so on. People get sick, get pregnant, quit, and they even die. You lose people from the workplace all the time. You also gain people and have to spend time getting them up and running. The system has to have slack in it otherwise the first resignation or accident would be the end of the business (and this is something that does happen most commonly in small business. If a principle is incapacitated or killed that will typically kill the business too). Anywhere drug use (or illness) really matters there will be policy to account for it. I can guarantee that if this sort of thing matters to your employer it will be a clause in your employment contract and the organisation's officially documented procedures will spell it all out explicitly.
2ForensicOceanography4dI do not think this is true. I think that it is important that we clarify this point before continuing.
0Stuart Anderson3dI had surveillance capabilities at work, those probably exist where you are too. Somebody already has all the answers to the questions you're asking. You already know where the smokers go, where the machine with the sweets is, and how busy the kitchen with its caffeine supply is. You can find a drug user and I can guarantee they'll know everyone else that's using too. Know the people that go clubbing? They're taking ecstasy at the very least. Half the time you can figure out which young people are using and what they're using just by looking at their musical preferences. Injectables leave puncture marks. Pay attention to anyone that is always wearing sleeves in weather that doesn't warrant it. Lots of drugs have side effects, either from use or withdrawal. Abnormal pupil dilation or response, shaking or sweating, twitches, ticks, paresthesia that makes them claw at themselves, speech speed, respiration depth and rate. Physical reactions generally aren't enormously obvious, but they're noticeable if you're looking. Assume that 1/4 of the staff are using psychiatric medication. More if the gender balance skews female. Drinking is endemic. Plenty of people will tell you all about their physical maladies and what meds they're on if you just sit there and listen. Speak to people outside of your work about their work environment. Again, people will happily tell you all about it if you just sit there and listen. For both of the above you'll be more likely to get info about others in the form of gossip. All you need to do is go to the target of the gossip and give them the opportunity to gossip back. All gossipers routinely betray each other and you'll rapidly know not only who's on drugs but every other bit of scandal in the office too. Finally, take a test swab of the boss's desk. There'll be cocaine on it (whether or not it's his doesn't matter. The boss's desk is like a magic cocaine altar that somebody has to snort off. Every time, without fail). If the peopl

How many people have a preexisting tendency towards overuse of these substances that is kept in check by the need to get up on time for work and be reasonably productive and presentable at work?

This implies that rich people who don't (or don't need to) work for their living will spend much more time on drugs, alcohol and porn, because they can afford to. Is that the case?

1ForensicOceanography12dProbably some of them do (at least in the popular imagination); I do not know if this can be checked. Maybe it is possible estimate drug consumption in a geographical area by enviromental data, for example the amount of cocaine retrieved in the water, and attempt to infer a correlation with income. But I do not know if there is sufficient data available. Surely not everyone would be like Ogodei Khan.

Also, some people would devote themselves to caretaking activities: lots of kids incl. foster kids, lots of dogs or cats. I’m not saying this is exactly bad, in some cases it’s good, but at extremes it can become hoarding, when the impulse to collect kids/pets overwhelms the motivation to adequately care for them.

1ForensicOceanography12dI understand how this can be very rewarding, but it is also an activity which requires mental effort (you do not just look after a kid the way you can drink a bottle or the way you can scroll your Facebook feed). It does not feel to me like the sort of activity in which you can just fall into, while you are planned to do other things. Are there documented cases of pathological dependence by caretaking?
5 comments, sorted by Highlighting new comments since Today at 8:04 AM

UBI is the modern Rorschach test -- everyone sees a different result when they look at the (few and usually flawed) experiments with UBI. I guess I am not different. Anyway, here is an obvious problem I see with the article:

[prime-age men who are not in labor force] report much less paid work than their peers—an average of just 12 minutes per day, nearly six hours a day less than employed men, and almost five hours a day less than employed women, but also close to an hour a day less than unemployed men. Perhaps more surprisingly, their time freed from work is not repurposed into helping out around the home, such as doing housework, cooking, and other tasks of home maintenance. In fact, they devote significantly less time to such home chores than unemployed men—less, too, than women with jobs. NILF men also spend much less time helping to care for other household members than working women—less time, as well, than unemployed men. Apart from work, by far the biggest difference between the daily schedules of NILF men and everyone else comes in what the ATUS calls “socializing, relaxing, and leisure,” a category that encompasses a range of activities, from listening to music to visiting a museum to attending a party. [...] The overwhelming majority of this “leisure” is screen time: television, internet, DVDs, and all the rest. [...] almost half of NILF men reported taking some form of pain medication every day.

The first impression is quite damning: The lazy voluntarily unemployed men don't do any productive work, don't even help at home, spend most of their time looking at screens... and for some weird reason take a lot of pain pills.

Ignoring the last part, it is exactly what the model "job is the source of all virtues" would make you predict. UBI would remove the need to work, then humanity would lose all its virtues, and we would all wirehead. Thank God for bullshit jobs that saved us from the threat of too much free time because of automation!

What about those pain pills, though?

So, here is another explanation that seems to fit the same data, and provides a different picture. Suppose you have a fraction of population that suffers from incurable chronic pain. It probably makes sense if they work less than healthy people. It could even explain why they help less at home. If you think about what such person could do -- lay in their bed, watch TV or listen to radio, talk to someone, take a walk -- depending on what categories are available in the questionnaire, it could fit under "socializing, relaxing, and leisure". It would definitely explain the taking of pain medication every day. And if these people realize they are unable to keep a job, so they stop actively trying to find one... then they get classified as "people who don't have a job and are not even trying to find one", duh.

Hey, I am not trying to say here that everyone who avoids job is actually a disabled person. Healthy lazy bums definitely exist, too. And I have no idea what is the actual proportion of these two groups in population. I just see an obvious alternative explanation that the article completely missed, because it automatically assumed the "job = source of all virtues" model. Makes me wonder what else they missed.

Yet another alternative explanation: Assume there are by nature two kinds of people: lazy and non-lazy. In a society where everyone must have a job or face serious consequences, the non-lazy people will have a job, and the lazy will be jobless. If you mistake correlation for causation, it is easy to conclude that the job makes people non-lazy (rather than the non-laziness making people employed). Essentially, you take a selection of people who are too lazy to get a job even in situation where not having a job seriously reduces the quality of your life, include the observation that they are also lazy in non-job aspects of their lives... and conclude that everyone is like that, only the jobs magically transform us into something better.

EDIT: To make my objection more simple -- It is statistically shown that in USA people without jobs are more likely to be black than people who have jobs. Conclusion: UBI will make you black. Discuss.

The "a UBI would make people spend time in ways that made them feel miserable" argument has always felt a little odd to me. It's essentially claiming that

  • People who could get jobs where they feel good
  • will either quit their jobs or just not get jobs in the first place
  • and feel miserable as a result
  • but will nevertheless choose to stay unemployed

But if that's the case, why wouldn't those people just... recognize that they are feeling bad, so get jobs and feel better?

I think there's an implication of something like "they will be so badly addicted to lotuses that they can't get a job despite knowing that the lotuses just make them feel worse", but if their mental health is in that bad of a shape, how likely is it that they could or would get a fulfilling job anyway?

I certainly believe that there exists some percentage of the population for whom this combination of factors holds, but it seems hard to believe that they would be such a significant fraction that their loss of well-being would outweigh the increased well-being that others would get from the UBI.

(Now if the argument was something like "people on an UBI would stop working and be happier as a result, and it's morally wrong for non-working people to be happier when they are just living off the who people do work", that would be a different matter, but that's not the argument being made here.)

Is there evidence that giving people a UBI would actually result in significantly more lotus-consumption activity? My understanding is that giving most people in the US an extra $1000/month, for example, would mostly go toward covering expenses, buying higher quality goods, or working slightly less to spend more time with family. 

At least the most typical result of the UBI trials that have been conducted so far seems to be that they neither increase nor decrease the employment rate (though they sometimes get cancelled because people think they make people less likely to work).

I think that Caplan is referring to a scenario in which the UBI is high enough to cause a significant reduction of the employment rate.

1000 $ per month would not achieve this effect.

By the way, here in Italy the state has recently enacted a law to give 780 € per month to unemployed people. The party which proposed this law has been mostly voted by southern Italy, whose ruling classes correctly predicted that it would have had the effect of increasing undeclared work.