Friday, May 31, 2013

Influence, Chapter 6

Chapter 6: Authority

Ah, the power of authority.

A shocking experiment!
This chapter begins (of course) with an account of Milgram's famous electric shock experiment, in which one subject (the "teacher") was instructed to deliver increasingly high voltages of electric shock to another subject (the "learner") for each wrong answer the learner gave.  As the shocks reached intolerable levels and the learner cried out in increasing agony, the scientist pressed the teacher to continue administering the shocks ("The experiment requires that you continue.").

In fact, the learner was a confederate, just pretending to be shocked.  The real experiment was about the willingness of a subject to do what an authority figure (the scientist) told him to do.
The result was that "teachers" became increasingly uncomfortable, but instead of refusing to continue, 65% of them continued to the end.

Ah, the power of --- editorializing?

Sigh.  I am tired of the standard interpretations of this experiment. Cialdini draws the usual, strongly negative conclusion that people are wrong to be such slaves to the authority of the scientists. Rise up, timid sheeples!  Don't just blindly follow orders!  You can think for yourselves!

But in a sense, this experiment couldn't be suggesting something more opposite.  Because actually it wasn't a bad idea to trust these scientists!  They weren't monsters after all, shockingly (!), even though they really, really seemed to be.

Authorities are authorities for a reason
Why do we listen to scientists? Maybe because we have accumulated a large body of evidence saying they are trustworthy.  Did you ever hear of a Yale psychologist actually torturing people?  Our prior is based on the fact that they don't really do horrific experiments.  This is an appropriately strong prior that perhaps shouldn't be overridden by apparent evidence to the contrary from inside a psych experiment.

(And gosh, a psych experiment! Are things ever as they seem in a psych experiment?  Not that the subjects knew this, though.)

Why follow orders?
Maybe you think they know what's best for you. Maybe they can hurt you if you don't do what they say.  Maybe by following orders, you are absolved from some portion of responsibility for your actions.  Maybe everyone else is following orders and going with the flow might be a good idea.  These are all fine reasons.

The question is not whether we tend to follow the orders of authority figures. (We do).  The question is whether our priors are well-calibrated on which authority figures to follow what kinds of orders from.  Scientists, governments, bosses, our we give them the proper weight on average?  It's not enough to look at a few decisions ex post and declare people stupid for following orders.  (And it's definitely not enough to look at an experiment in which people weren't stupid to follow orders and declare them stupid for following orders.)

What? A Doctor?
Of course, when a trait is correlated with (justified) authority, it can potentially be mimicked by non-authority figures hoping to ride the wave.  Anyone can dress like a police officer or call themselves a doctor to gain an air of authority!

But while this might look like authority having "undue influence" to the naive social scientist, it's a bit of a masquerade.  The real man behind the curtain is Dr. Incomplete Info, also known as Doctor What the Heck's Going On? or simply Doctor What for short.  You can trust Doctor What -- after all, he's a doctor! -- to mess things up sneakily.  His specialty is sneakily taunting his victims with all the great outcomes they could achieve If Only.  Like any successful man behind the curtain, Doctor What is seldom blamed directly for his deeds.

I'm a stellar candidate for this job.  If I'm the best person for the job I should get the job. Why can't I even get an interview?  If Only I could convince my prospective employer. (Life is so unfair).
Look at these stupid buffalo being herded off a cliff.  If Only they realized the dangers of herding behavior.
Look at this dupe trusting a salesman pretending to be a doctor. If Only he realized that the appearance of authority can be deceiving.

When you know the truth, you feel strongly that the world ought to align with what you know, and any misalignment is stupid or unfair.  Because the truth is true. The truth is right.  But behavior isn't a function of the truth, it's a function of the information people happen to have.  How could you expect people to act on information they don't have?  Perhaps they couldn't possibly know the truth at the time of their decision, or perhaps it is simply too costly for them to gather the information.

Coexisting peacefully
Because economists are interested in optimal decision-making, and because decision-making depends on the available information, we are hyper-aware of information and the role it plays.  Economics has uncovered many ways to facilitate information flow, but in the process, we have also been forced to recognize that information is often too costly to be worth collecting.  This can include the cost of walking downstairs to figure out if the mailman has come yet; the cost of getting your trading partner to reveal how much he really wants the microwave you're trying to sell him; the cost of sorting through all possible job applicants to find the true best fit for the job...

In an imperfect world, such impossibility results bring peace of mind.  They tell us that even if we do the very best we can, we have to live with imperfection, so we may as well accept it as a fact of life.  Faced with a suboptimal outcome, people's first instinct is to complain, to search for a culprit, someone to blame.  But things can suck without anyone to blame.

Well, in this instance you can blame Doctor What.  But even better would just be to accept the reality of the situation.

Economists have coexisted peacefully throughout time with the Doctor in all his various incarnations.  They see through his curtain, which is to say they know when he's up to no good...but there are no answers there, only unanswered questions...

The bottom line
Cialdini says we should be wary of fakers and mistrustful of authority.  And while it's true that paying extra attention may help us to distinguish some fakers, that's no magic bullet.  Effort is costly, and for any level of effort we put in, there will still be some fakers who manage to make themselves indistinguishable from genuine authorities.  It is equally intelligent to trust every person who appears to be an equally convincing doctor, regardless of the actual truth.

Monday, May 13, 2013

The Value of College

Tony has a nice post about the value of college:

A huge part of the value of college its option value: i.e., finding out (a) is a diploma is worth it for me? (b) what kind of diploma is worth it for me?   Even if the answer is that the student should drop out and do something else, going to college is valuable because it helps the student figure this out.
Do high dropout rates imply that college is a bad investment?  No!  You can't determine whether college is worth enrolling in by looking at dropout rates.  To the contrary, 46% dropout rates are good evidence that people are learning a lot about the suitability of college for them.  If they knew up front that they were going to drop out, they might not bother attending in the first place.

College is a great facilitator of information flow.  It is a chance to efficiently learn about a great many things.  You learn about subject material, yes, but also about your own affinity and aptitude for various subjects. You learn a lot about yourself, and about other people, too.  It's relatively easy to find friends (and more) in college.

Making the best use of college is a tricky business.  For one, it's too easy to take college for granted.  College campuses dramatically lower the cost of interacting with a variety of subjects and people.  In fact, because they are part of a trend of decreasing costs in life up to that point, it's natural to think the trend will continue for years to come.  And then you might not focus enough on making low-cost investments in your future, instead of living it up in college.

The college atmosphere can trick you into thinking it's a good idea to major in something that won't help you after graduation.  There are few expectations imposed upon college students, which is how they come to have so much time for exploring.  In such a low-cost environment, nearly anything has a positive return.  The real world, by contrast, expects people to do stuff that generates value for other members of society.

Friday, May 3, 2013

Influence, Chapter 5

Chapter 5: Liking.

You will not be surprised to hear that liking people increases our willingness to comply with their wishes.  But the strength of the effect is perhaps surprising.  And the ways likeability can be manufactured, say by a salesperson or other compliance professional, are worth a chapter for sure.

Physical Attractiveness
Yep.  Good-looking, well-groomed people are strongly favorably treated in politics, in job interviews, in the courtroom, and just about everywhere else.

I'll see you in court! 
In a Pennsylvania study, researchers rated the physical attractiveness of seventy-four separate male defendants at the start of their criminal trials...The attractive defendants were twice as likely to avoid jail as the unattractive ones.  In another study...a defendant who was better-looking than his victim was assessed an average amount of $5,623 [in damages]; but when the victim was the more attractive of the two, the average compensation was $10,051.
Maybe attractiveness is correlated with other underlying traits that partially explain this, I dunno. But strong attractiveness effects are also easy to identify in experiments.

Improper schooling!
Research on elementary-school children shows that adults view aggressive acts as less naughty when performed by an attractive child and that teachers presume good-looking children to be more intelligent...
Whoa! Well, there's probably something going on here too.  Although...I mean nobody likes to talk about it, but...shouldn't more attractive people actually tend to be smarter? Beauty and intelligence are both positive attributes.  In the marriage market, beautiful people should tend to match with people who are higher in both attributes.  Their children should be more beautiful and more intelligent than average.

I also wonder whose job it was to rate elementary school kids on "attractiveness." (Grad students! Grad students!)

Vote for the ugly ones!
Unsurprisingly, attractive politicians do better in the polls.  As an aside, a consequence of selection for physical attractiveness in politics is that uglier politicians should on average be better along other dimensions that are valued in politics (speaking ability, bargaining power, sanity...).  This follows from a point Jeff Ely has made a few times (see for instance  (To check your understanding: why do I expect more attractive politicians to be worse in other ways, while I expect more attractive people to be more intelligent?)

In fairness, likeability is genuinely important for a politician's effectiveness; however, if we tend to overweigh attractiveness, you should respond by downgrading attractive politicians.  The next time you're on the fence in the voting booth, maybe you should pick the uglier one!


We like people who dress like us, talk like us, have similar background and interests.

This one was an eye-opener for me.  Yes, car salesmen are trained to mirror posture, mood, verbal style.  But apparently they are also trained to manufacture a persona that coincides with whatever they can learn about their potential customer.

This includes looking for evidence about a buyer from the car that he drives into the lot.  If there is camping gear in the back (or the car betrays other signs of an outdoorsy lifestyle), the salesman can mention that he loves to get away from the city whenever possible.  If the license plate is out of state, he might ask where the person is from and report that he (or his wife) was born there.  If there are political bumper stickers or a Jesus fish or...well, you get the idea.  A car can say a lot about a person before the salesman has even made first contact.

Up until now, I wouldn't say that I have trusted salespeople, exactly.  But their trustworthiness just got seriously downgraded, which is saying something.

Here is some good advice:
Because a veneer of similarity can be so easily would be wise these days to be careful around salespeople who just seem to be just like you.
Or more generally, in light of all the discussion in this chapter, be wary of salespeople who you just seem to like for some reason.  In fact, I will coin my own pithy moral:
Be wary of the salesman you seem to (be)  like.


According to Cialdini, we are "phenomenal suckers for flattery," even when we know the flatterer has something to gain by flattery, even when the flattery is completely inaccurate(!)

Praise makes us feel liked.  And we reciprocate with a powerful urge to like the praise-giver back.

This is a great time to mention that at Economonomics, we genuinely love our small but thoughtful and intelligent readership!  And remember, comments are both the best complements and the best compliments to a post you have enjoyed!

(Every comment is a complement and a compliment, whether it contains compliments or not.  In fact if you did leave a few words of praise, that would be a complementary complimentary compliment.)

Contact and Cooperation

Being in cooperation with someone, even being forced to cooperate, strongly influences us to like that person.  They become allies.  We begin to trust them, and to internalize their goals and desires in our own utility functions.

I infer from the tone that Good Cop/Bad Cop was unfamiliar to readers in 1984 (when Influence was originally published).  These days, it's such a staple of crime drama that even the interrogated characters know about it.  In any case, the point is that after the Bad Cop rips into the subject, the Good Cop comes in and adopts the air of a trustworthy ally who thinks his colleague is taking things too far.

Similarly, a salesman will pretend to "battle with his boss."  The salesman seems to be on your side, trying to get you a good deal.  It's an incredibly effective tactic.

Conditioning and Association

Advertisements are full of beautiful people.  This is partly to get our attention, but partly also to make us like the product by association.  Mere association with something positive increases our opinion of the product, even when the associated feature has nothing to do with the product itself.

This isn't hard to believe or understand.  As a general heuristic, positive things tend to be associated with each other.

Heads we win, tails they lose!
Apparently we also take measures to associate ourselves with positive things while avoiding the negatives.  Cialdini did a neat experiment in which he asked students at Arizona State to describe the outcome of football games.  When the team lost, fans tended to say things like, "They lost the game."  But when the team won, it became, "We won!"  This is perhaps best exemplified by this gem of a response from one student:
They threw away our chance for a national championship!
Oh, sports fans.  Cialdini also describes an interesting extension where students took a general knowledge test before being asked to describe the game.  The test was rigged so that some succeeded and others failed badly.  The people whose pride had just taken a blow were much more likely to associate themselves with a victorious team, or distance themselves from a losing team.  In other words, association with success is used as a substitute for success.


Time to disassociate! (Or, How to Lose Credibility and Alienate Readers)
So far, I think we've been having a pretty good time with this chapter.  But the author is about to go horribly, horribly astray.

I'm flipping through the chapters as I write these reviews.  And this is the unfortunate part of the review where I just flipped to a page covered in angry margin notes.
In the aftermath of the [1980 Olympics victory of the US hockey team] over the Soviet team, scalpers were getting a hundred dollars a pair for ticket stubs...Although the desire to bask in reflected glory exists to a degree in all of us, there seems to be something special about people who would wait in the snow to spend fifty dollars apiece for the shreds of tickets to a game they had not attended, presumably to "prove" to friends back home that they had been present at the big victory...
Lying to friends...really?  And he goes on...
Unless I miss my guess, they are not merely great sports aficionados; they are individuals with a hidden personality flaw---a poor self-concept.  Deep inside is a sense of low personal worth that directs them to seek prestige not from the generation or promotion of their own attainments, but from...associations with others of attainment. 
Oh lord. This is what happens when you take a psych theory too far.  There can be a very large difference between an effect discovered in the lab, and the net effect of all the forces that characterize real world behavior.

People collect stuff.  
Which level would you like me to defend this on?  How about this: it's fun to collect stuff.  No?  Okay then, how about this: There is a thriving resale market for sports memorabilia. You don't even have to enjoy it!  According to Antiques Roadshow, those $50 dollar Miracle on Ice ticket stubs are worth over $500 today.   Furthermore, an unused ticket is worth over $2,500 today.

Yes, the unused ticket is worth more. Even though, unfortunately for Cialdini's theory, it cannot be used to prove to your friend that you were at the game.  Quite to the contrary, I suppose an unused ticket would prove that you wasted a ticket instead of sending your friend to the Miracle on Ice.  (And yes, you would be seen as a total jerk, because sports fans are not known for their keen ability to distinguish between ex ante and ex post reasoning).

Now one might ask, where does this value come from then?  Even if it's smart to buy a $50 stub for resale purposes alone, doesn't someone have to value it for its own sake?  Isn't it possible that the $500 resale price is largely underpinned by the motives Cialdini describes?

The bizarre truth is, it is perfectly possible for ticket stubs to be valuable even if no one actually values the stubs for their own sake!  What? Intrinsically worthless pieces of paper going for $500?


How about I'll agree not to put a whole annoying page of dots, and you can pretend you didn't already see the punchline in your peripheral vision?


Technically this bill is worth considerably more than $500 nowadays, but you get the idea.
Yep.  That's basically what money is.  It has value today because people expect it to still have value tomorrow.

My point is not to argue that Miracle on Ice ticket stubs don't have intrinsic value to people; to the contrary, I'm sure they do.  But good luck figuring out which part of the $500 is "intrinsic" value, and which part is value based on the fact that people expect them to have value tomorrow.  Which implies, especially: good luck arguing that a particularly high price for ticket stubs is evidence of deep emotional insecurity.

Even if he's right (he's still wrong)
I'm actually not done demolishing Cialdini's argument.  For the final pass, let's suppose Cialdini is completely correct that sports fans and memorabilia collectors really are largely driven by a desire to associate with the triumphs of others. A hidden personality flaw, he says!  Low personal worth, he says!

Holy absolute versus marginal fail, Batman!
Just because you really enjoy a good steak doesn't mean you are really unhappy without steak.  If we collect a bunch of steak lovers do we expect them to be secretly unhappy whenever they aren't eating steak?  No.  Similarly, you can have a great life and still really enjoy the vicarious successes of other people's lives.

Slightly more technical: In a lab, you can vary people's baseline utility while effectively holding constant the expected marginal value (conditional on baseline utility) of associating with success in sports.  But in the real world, people select into sports according to marginal values, not baseline utilities. Sports fans are people who get particularly high marginal value from associating with sports.  This doesn't imply they have low baseline utilities, even though in the lab the people with the highest marginal values were the people with the lowest baseline utilities.

Are you not entertained?
Now, let me shift gears a bit and make a broader point about entertainment.  I'm not much of a sports fan but I would bet the real, dominant reason for sports fanaticism has got to be less about associating with success of others, and more about mentally experiencing the drama and success oneself.  There's a fine line between these, but sports are best understood as an immersive experience analogous to a good book or movie.  Good drama immerses you and invests you in the outcome.

The real appeal of sports, or for that matter most entertainment, is that the drama in sports and movies and books is so much more dramatic than your own drama.  Have you ever gone up against a league of the world's best athletes and vanquished them in a glorious display of skill and strength and other traits that you value?  Have you ever launched an against-all-odds rebellion against the evil Empire and blown up the Death Star using mystical Force powers?

Stories are made less compelling by the fact that they aren't "real," but more compelling by the fact that awesomer things happen in them than in reality.  

Sports are so popular because they are social, immersive, and considerably more "real" than most other forms of mass entertainment.  Movies and novels are successful because the author is totally unconstrained by reality and thus free to choose the optimally dramatic sequence of events.  Video games are successful because they pull you into the action and give you an actual role in victory.  (And the future will be full of progressively more immersive entertainment experiences).

Summing it up
The point of all this is, I think the immersive entertainment effect has got to be orders of magnitude stronger than the associative effect.  You have to add them together to ascertain a sports fan's motives, and when you do, the associative effect gets swamped.  Yes, we especially like to associate ourselves with teams when they are successful. But somehow the Cubs still have fans.

If you've made it this far, I salute you.  Or rather, I salute you, dear reader who has made it this far.  I hope you have been entertained.  Incredibly, that only covers the first half of my angry margin notes on page 203...but we can talk about stage mothers and rock-music groupies another time.