DISCLAIMER: This is an exploration of a theoretical economics problem. This is not advice. I have not made up my mind. Please do not cite this post as support for your plans to indulge in mayhem or selfishness.
Suppose you want to lead a life that, compared to the average resident of an industrialized country, is relatively altruistic. Say you value your own comfort and pleasure at roughly 104 × the comfort and pleasure of a stranger who currently lives across the world from you. You are likewise relatively patient, risk-neutral, and scope-sensitive compared to your peers; if you believe that there is a 10-3 chance that 109 people will die in a specific kind of catastrophe that would occur in the year 2031, then you would factor that into your plans at a discount of only two orders of magnitude. In other words, you would be indifferent between (a) the risk of the future catastrophe and (b) the certainty of 104 people dying today. Moreover, you are reasonably well-calibrated; if you say something will happen with 95% certainty, it actually does happen about 90 out of 100 times. Finally, suppose you have an IQ of roughly 130, and no crippling physical or mental disabilities. My numbers are meant to sketch a crude portrait of one plausible kind of amateur rationalist; if you don’t like the numbers, please mentally substitute your own and move on, because the exact numbers aren’t my point.
My point is this: does it make sense to follow social norms like tipping, waiting in line, making small talk with strangers, and paying taxes when you’re reasonably sure that nobody who’s important to you is watching? Although such norms are cheap to follow in terms of individual situations, they recur frequently—if you make a habit of skewing your time toward bars and clubs that take cash under the table, you’ll put a serious dent in your entertainment budget. Repeat this across several different areas of your life, and you’re looking at significant resources. As an amateur rationalist, with the time and money you save by cheating on ‘responsible’ habits, you should in theory be able to get slightly more work done and/or donate slightly more money to efficient charities. You might even be able to invest the time and money into bootstrapping your own economic productivity, allowing you to accelerate your donation timetable by a few hundred $/yr2 , and thereby, over the course of your life, save another hundred lives or so. Assuming you never get caught, isn’t that worth the mild inconvenience you cause to several thousand strangers? Assuming you do occasionally get caught, does that really tip the balance back toward following the rules? Remember that you’re well-calibrated, so while you do make mistakes, you should in theory be able to identify situations where you are extremely unlikely to get caught, cheat in all such situations, and then only get caught on extremely rare occasions.
It certainly *feels* good to tell stories about how social niceties pay off. Following these kinds of rules and making arguments that we should follow them both signal a certain kind of pro-social, well-adjusted, trustworthy, successful attitude. Plus, it would be nice if everybody got what they deserved; i.e., if not tipping actually led to something like bad karma. Even without mystical influence, not tipping might lead to individual punishments if, e.g., it’s impossible not to feel guilty about it, or it subtly alters your personality for the worse, or if people inevitably catch you when you least expect it and then impose costs on you that are worth more than the cumulative money you saved on that occasion plus the money you saved on all the other occasions when you weren’t caught.
Those kind of mechanisms seem unlikely to me, though. Because of wishful thinking, though, I would imagine that we tend to overestimate the chance that something like karmic balance actually obtains. What do you think?
EDIT: Please try to cope with The Least Convenient Possible World when imagining examples, rather than simply picking average examples that suggest, e.g., that you will certainly be caught, as this is contrary to the spirit of the exercise. E.g., for tipping, imagine dining alone at a chain restaurant with high turnover while traveling through a town you rarely visit. For avoiding small talk, imagine you are on a light rail in the exurbs with your laptop, typing up some notes, a full twelve minutes from the nearest likely source of enough extra passengers to disturb your personal space, alone in a car with a nice old man who wants to talk to you, apparently ad infinitum, about the 1954 Brooklyn Dodgers’ infield. If you don’t want to eat at restaurants, that’s fine—pick some other relevant pleasure or luxury that an amateur rationalist might indulge in. If you own a car, should you ever refuse to let someone in who’s waiting to merge, assuming that maintaining your current speed in your current lane is unlikely to increase your risk of an accident? If you rent or own a house in the western United States, should you ever use scarce water to irrigate your flowers, given that your neighbors are not environmentally conscious enough to notice whether you are using ecologically thrifty plants? Etc. An interesting counter-argument is “it takes too much mental energy to identify occasions when I won’t be caught.” A boring counter-argument is “I’ll surely be caught.” The latter violates an explicit assumption of the problem.
Should Rationalists Tip at Restaurants?
Related to: Robin Hanson on Freakonomics
DISCLAIMER: This is an exploration of a theoretical economics problem. This is not advice. I have not made up my mind. Please do not cite this post as support for your plans to indulge in mayhem or selfishness.
Suppose you want to lead a life that, compared to the average resident of an industrialized country, is relatively altruistic. Say you value your own comfort and pleasure at roughly 104 × the comfort and pleasure of a stranger who currently lives across the world from you. You are likewise relatively patient, risk-neutral, and scope-sensitive compared to your peers; if you believe that there is a 10-3 chance that 109 people will die in a specific kind of catastrophe that would occur in the year 2031, then you would factor that into your plans at a discount of only two orders of magnitude. In other words, you would be indifferent between (a) the risk of the future catastrophe and (b) the certainty of 104 people dying today. Moreover, you are reasonably well-calibrated; if you say something will happen with 95% certainty, it actually does happen about 90 out of 100 times. Finally, suppose you have an IQ of roughly 130, and no crippling physical or mental disabilities. My numbers are meant to sketch a crude portrait of one plausible kind of amateur rationalist; if you don’t like the numbers, please mentally substitute your own and move on, because the exact numbers aren’t my point.
My point is this: does it make sense to follow social norms like tipping, waiting in line, making small talk with strangers, and paying taxes when you’re reasonably sure that nobody who’s important to you is watching? Although such norms are cheap to follow in terms of individual situations, they recur frequently—if you make a habit of skewing your time toward bars and clubs that take cash under the table, you’ll put a serious dent in your entertainment budget. Repeat this across several different areas of your life, and you’re looking at significant resources. As an amateur rationalist, with the time and money you save by cheating on ‘responsible’ habits, you should in theory be able to get slightly more work done and/or donate slightly more money to efficient charities. You might even be able to invest the time and money into bootstrapping your own economic productivity, allowing you to accelerate your donation timetable by a few hundred $/yr2 , and thereby, over the course of your life, save another hundred lives or so. Assuming you never get caught, isn’t that worth the mild inconvenience you cause to several thousand strangers? Assuming you do occasionally get caught, does that really tip the balance back toward following the rules? Remember that you’re well-calibrated, so while you do make mistakes, you should in theory be able to identify situations where you are extremely unlikely to get caught, cheat in all such situations, and then only get caught on extremely rare occasions.
It certainly *feels* good to tell stories about how social niceties pay off. Following these kinds of rules and making arguments that we should follow them both signal a certain kind of pro-social, well-adjusted, trustworthy, successful attitude. Plus, it would be nice if everybody got what they deserved; i.e., if not tipping actually led to something like bad karma. Even without mystical influence, not tipping might lead to individual punishments if, e.g., it’s impossible not to feel guilty about it, or it subtly alters your personality for the worse, or if people inevitably catch you when you least expect it and then impose costs on you that are worth more than the cumulative money you saved on that occasion plus the money you saved on all the other occasions when you weren’t caught.
Those kind of mechanisms seem unlikely to me, though. Because of wishful thinking, though, I would imagine that we tend to overestimate the chance that something like karmic balance actually obtains. What do you think?
EDIT: Please try to cope with The Least Convenient Possible World when imagining examples, rather than simply picking average examples that suggest, e.g., that you will certainly be caught, as this is contrary to the spirit of the exercise. E.g., for tipping, imagine dining alone at a chain restaurant with high turnover while traveling through a town you rarely visit. For avoiding small talk, imagine you are on a light rail in the exurbs with your laptop, typing up some notes, a full twelve minutes from the nearest likely source of enough extra passengers to disturb your personal space, alone in a car with a nice old man who wants to talk to you, apparently ad infinitum, about the 1954 Brooklyn Dodgers’ infield. If you don’t want to eat at restaurants, that’s fine—pick some other relevant pleasure or luxury that an amateur rationalist might indulge in. If you own a car, should you ever refuse to let someone in who’s waiting to merge, assuming that maintaining your current speed in your current lane is unlikely to increase your risk of an accident? If you rent or own a house in the western United States, should you ever use scarce water to irrigate your flowers, given that your neighbors are not environmentally conscious enough to notice whether you are using ecologically thrifty plants? Etc. An interesting counter-argument is “it takes too much mental energy to identify occasions when I won’t be caught.” A boring counter-argument is “I’ll surely be caught.” The latter violates an explicit assumption of the problem.