The Trust Manifesto

“We are as gods, and we might as well get good at it.”
— Stewart Brand

No Comments

My friend Gabe had a summer job with the local Weights and Measures
department. I don’t think he busted too many people with bogus scales
or gas pumps. They use these really elaborate devices for measuring
gas pumps.

The temptation for a gas station to cheat its customers must be
extremely high. It is hard to get an exact average for profit per gallon
of gas because it changes each week, and varies dramatically by
state and whether the gas station is self serve or attended. But a
typical station might make just 10 cents of profit on a $2.49 gallon
of gas. If that is the case, then by altering the pump to dispense just
4% less gasoline, the owner could double the profit of the entire
gasoline operation. Even if he took a more cautious route and
altered the pump for just six months following each inspection
(because gas pump and restaurant inspectors don’t get around as
often as one would hope), then he would have 50% more income
for his family without taking much additional risk.

So what should we do? Should we all buy these measuring
trailers and bring them everywhere we buy gas?

My answer is that you should check your tire pressure. What?
How can checking your tire pressure solve this problem? Well, it will
not solve the problem of the dishonest pump, but it is much, much
more likely to increase the distance that your car will travel per dollar,
which should be a more important goal than how much gasoline you
get per dollar. The same principle applies to people who drive across
town with underinflated tires to buy gasoline that costs five cents less
than the gas closer to their house. It just doesn’t make sense for many
reasons. For one thing, inflating your tires properly boosts your gas
mileage for thousands of miles, whereas finding cheap gas improves
your cost per mile for only hundreds of miles.

A recent study showed that 27 percent of cars and 32 of light trucks
had severely underinflated tires (more than 8 pounds). Underinflated
tires can use up to 15% more gasoline, plus they wear out much faster
(at maybe $400 a set), and they are more likely to cause fatal
accidents.

This is a fairly mundane example; many of you have never worried
about gas pumps being incorrect, and many never will. This is good.
But there are many analogous situations which have much more
profound consequences. I think almost all of them are instances of
people losing sight of a larger goal, and trying to optimize some smaller
piece when there is an easier, more effective piece to work on. In the
case of driving the greatest distance for the least money, getting as
much gasoline as one pays for, or getting a good price per gallon, seems
like a natural place to look to optimize things. But the tires are an easier
thing to fix, are much more likely to be causing the problem, and the
fix has longer lasting benefits. All it takes is to relax one’s ego and
allow for the possibility that someone might be ripping you off a dime
a gallon every so often. This is not always an easy thing to do.

In future postings, I will write about terrorism. Terrorism interferes
with our desire to live long and happy lives. There is no question about
that. But taking many precautions against terrorism is not the best
use of time if you are trying to prolong your life, just as testing the
volume of gasoline coming out of a pump is not the best way to ensure
your car goes the most miles per dollar. So why not attack the easier,
and more likely problems, and not worry about the intractible and
improbable ones?

1 Comment

I was at Border’s this week, and during the holiday season, they allow various
non-profits to come in and do the wrapping. This is a great move for Border’s;
normally their employees perform this service at no charge, and so this frees
them up to run more registers. It is also great for the shoppers; the lines at the
register become shorter, and you can still get your books wrapped for free if
you have the nerve to not put money in the jar. And of course it is great for
the non-profit, because people usually throw some money in.

I was wondering whether it is better to set a price for wrapping a book, or
just to allow people to donate whatever they want. First, one microeconomics
review. You may remember that people have different preferences, or utilities
for goods, so for example, someone might be willing to pay $100 for a Dell
computer, and someone might be willing to pay $500 for it, and someone might
be willing to pay $1000 for it. Those prices obviously depend on how much one
needs or wants the item, and how much money one has. If Dell prices the
computer at $499, then the first consumer would choose to not buy it, and the
second and third consumers would buy it. The difference between what they
were willing to pay and what it actually cost is called consumer
surplus. In one case, it is $500-$499=$1, and in the other case it is
$1000-$499=$501. A seller’s profits can generally be maximized if they can
find out how to legally segment the market and charge each customer close
to what they were willing to pay, thus reducing consumer surplus. Airlines
know that business customers might be willing to pay more than vacationers,
so they off discounts for Saturday night stayovers on the theory that many
business customers will not want to do that. This means that people sitting next to
each other on the same plane can be paying wildly different amounts for what
amounts to the same service. This is also the theory behind offering long distance
calls for less on weekends and after 11 PM. The question is how we can coerce each
shopper in the bookstore to pay close to what they think the wrapping service
is worth to them, without too many people lying about it by not paying anything.

I wrote a little program in Excel to simulate six scenarios: free will donations,
and $1, $2, $3, $4 and $5 per book. We will assume that 50 cents worth of paper
and tape is used per book, which should be ample. In the free will case, every book
gets wrapped, and the non-profit could actually lose money on a given book
if the donation is less than 50 cents. In the $1 case, the people who are not willing
to pay that much will not come to the stand, so the number of books wrapped
goes down, and we make exactly 50 cents profit on each book. In the $2 case, the
people not willing to pay that much will not come, so we wrap even fewer books
than in the $1 case, but we make $1.50 per book profit, which is three times as
much as when we were charging $1. And so on for $3, $4, and $5. It actually
sounds pretty appealing to charge a large fixed cost per book, because there is
less work and much more profit per book. Here is the chart of profits:

You may dispute my assumptions for how much paper costs, or how preferences
for payments are distributed randomly. If so, it is easy to write your own simulation
with your own assumptions. But I will bet that my assumptions are not that far
off. What this clearly shows is that charging $2 per book is better than charging
$1 a book, and so on, all the way up to $4, and then the total profit starts to fall
again. But, no scheme is as profitable as just leaving a jar open and trusting people
to toss in whatever they want.

In my simulation, one hypothetical bastard gave 39 cents to charity, and one
fictional scrooge gave 1 cent to charity. Money was lost on those two, which would
steam some people who do not believe in the Trust Manifesto. But the free will scheme
allowed the non-profit to effectively charge a price closer to each shopper’s utility value,
which thus reduces the total consumer surplus for what those shoppers would pay for
wrapping services. Maybe those “pay what you want” restaurants are not so crazy. I
do think that the fact that it is a charity doing the wrapping boosts the rate of honesty.
A “pay what you want” parking meter would probably not have to be emptied very often,
but you never know.

No Comments

Most people have probably been in a McDonald’s or Burger King that has a self service
soda fountain. Some of you may have even filled up your cup twice at one. Let’s do
an economic analysis of this idea. First of all, we need to make some assumptions.
We will only look at the indoor sales, because the drive through customers cannot
pour their own drinks. We will assume that indoors, a restaurant sells 400 small
drinks, 400 medium drinks and 400 large drinks in a day. As you probably know,
the wholesale cost of soda is pretty low, and the advertising for the drink may cost
as much as the syrup for the drink. Also, a fast food restaurant typically has a very
small gap between the prices of its drinks, so that you are encouraged to upgrade
to a medium or large for just a little bit more.

Let’s assume that if people are allowed to serve themselves soda, then 100 of the
people who would have bought a large soda will instead by a small, and go back
and fill it up again after finishing their first cup. So our sales then become 500
small, 400 medium, and 300 large. If it takes 30 seconds to fill and cap each cup,
then it is about 7 cents of labor for an $8 an hour employee. The average retail
prices for McDonald’s soda were found on the web.

To summarize, self service soda would save $80 a day in labor, but there would
be $40 a day in losses if 100 people bought small sodas instead of large ones
and refilled their cups. (The loss comes entirely from the 40 cent price difference;
the amount of syrup consumed is identical in both cases.)

The net savings is $40 per day for the restaurant. That does not sound like alot,
but it adds up to $14,600 per year. That would pay for an extra person to be
working during busy periods, or could represent a 10% boost in profits for the
restaurant.

Since McDonald’s has so many restaurants, it would mean a $461 million savings
across the entire chain. Since McDonald’s is a franchise, the $461 million would
not all go to the parent company, but would be the sum of all the benefits for
all the owners. If they chose to donate all that savings towards Ronald McDonald
Houses, for instance, that would be a very well funded charity.

No Comments

Exercise caution in your business affairs;
for the world is full of trickery.
But let this not blind you to what virtue there is;
many persons strive for high ideals;
and everywhere life is full of heroism.

Max Ehrmann, Desiderata, 1952.

No Comments

Better Days – Goo Goo Dolls

And you ask me what I want this year
And I try to make this kind and clear
Just a chance that maybe we’ll find better days
Cuz I don’t need boxes wrapped in strings
And desire and love and empty things
Just a chance that maybe we’ll find better days

So take these words
And sing out loud
Cuz everyone is forgiven now
Cuz tonight’s the night the world begins again

And it’s someplace simple where we could live
And something only you can give
And thats faith and trust and peace while we’re alive
And the one poor child that saved this world
And there’s 10 million more who probably could
If we all just stopped and said a prayer for them

No Comments

On a recent trip to India, I was amused by the list of prohibited items
on domestic flights. They are presented on a computer monitor in
alphabetical order, so there are some weird juxtapositions. For example,
I remember the sequence “… , blasting caps, curry powder, dynamite, …”.
I do not want to get on an airplane with anyone who needs to be
reminded to not bring blasting caps and dynamite, or with anyone who
can take over an entire airplane just with curry powder. I think someone
has been watching too many Steven Seagal and Jackie Chan movies.

Amtrak used to have a hysterical list of prohibited items, but I can’t
find it online. I remember that it included power lawnmowers. They
have a pretty predictable set of prohibited things listed on their web site,
and you can fill out a form to get a paper copy of the full list in the mail.
This is why they are bankrupt. (If you have the Amtrak list, please let
me know.)

On my way back from India, I talked to my cousin who works for an
airline. He had just returned from a one day course on prohibited items
which was held in Malaysia. (No web based training needed if you work
for an airline, I guess.) He said two major problems are people bringing
parachutes and gasoline on board. Apparently people who own
parachutes never let them out of their sight – kind of like Yo-Yo Ma
and his cello, only with more serious consequences. Still, I’d
be more comfortable sitting on a plane next to Yo-Yo Ma than some
D.J. Cooper wannabes with parachutes.

Here is a list of what is banned on Canadian, Indian or US flights:

• - adhesives
• - aerosol (except hair spray)
• - ammunition
• - antifreeze
• - arsenic
• - axes
• - bacterial cultures
• - baseball bats
• - BB guns
• - billy clubs
• - black jacks
• - blasting caps
• - bows and arrows
• - box cutters (you would think “duh”, but read on)
• - brass knuckles
• - C-4 (which you would be arrested for possessing anywhere)
• - car batteries (that would keep the iPod going for a while)
• - catapults (would that even fit through the door?)
• - cattle prods
• - charcoal briquettes (the food sucks on planes, so I like to barbecue)
• - chlorine for pools and spas
• - compressed air guns
• - compressed gas cylinders
• - cricket bats
• - crowbars (some people don’t know how to travel light)
• - cyanide (don’t leave home without it)
• - darts
• - detonating cord
• - drain cleaners
• - drills, unless they are to be used with a prosthetic device (hip surgery on the plane?)
• - dyes
• - dynamite
• - enamel
• - fertilizer
• - fiberglass repair kits (I like to fix Corvette fenders on the longer flights)
• - firearms
• - fire extinguishers (um, shouldn’t there already be some accessible onboard?)
• - fireworks
• - flame retardant compounds (which I wear when others are carrying dynamite)
• - flares, in any form (how many forms are there?)
• - flare guns
• - fuels
• - gasoline
• - gas torches
• - golf clubs
• - gun powder
• - hammers
• - hand grenades
• - hatchets
• - hockey sticks
• - hunting knives
• - ice skates
• - laboratory specimens
• - lacrosse sticks
• - laquer (I like to paint those Corvette fenders after fixing them)
• - lighter fluid
• - lighters
• - liquid bleach (would solid bleach be fine?)
• - liquids that are unidentifiable in unmarked containers (just write Orange Juice on the container and you are all set)
• - lye
• - mace
• - meat cleavers
• - mercury (not a good idea to be carrying anywhere)
• - military explosives
• - mines (in case you want to close off a harbor at your destination?)
• - night sticks
• - nunchakus
• - office machine cleaners
• - parts of guns (some friends could each bring different parts and build one in flight)
• - pellet guns
• - peroxide (board a brunette, disembark a blonde)
• - pesticides
• - plastic explosives (if I’m not mistaken, you shouldn’t be carrying that on the street, either)
• - pool cues
• - radioactive material
• - replicas of firearms
• - restraining devices
• - road flares
• - rodent poisons
• - sabres
• - saws
• - scalpels
• - scissors with pointed tips, except if they are ostomy scissors
• - screwdrivers, except those in eyeglass repair kits
• - scuba knives
• - sheet explosives
• - shoe polish
• - ski poles (kind of large for the overhead bin or under the seat anyway)
• - sling shots
• - smoke detectors
• - spear guns (there was actually a guy carrying a loaded spear gun standing next to us on a ferry, and we alerted the captain who asked him to put it away)
• - spillable batteries except those in wheelchairs
• - spray paint
• - starter guns (how will we race up and down the aisle?)
• - stilettos
• - straight razors
• - strike anywhere matches
• - strong magnets
• - stun guns
• - swords
• - tear gas
• - throwing stars
• - tools, including but not limited to wrenches and pliers
• - wrenches and pliers (which I think is part of “including but not limited to wrenches and pliers!”)
• - transformer robots that form into toy guns (I kid you not)
• - turpentine and paint thinner
• - umbrellas containing sword blades or arrows (Yo! Who carries an umbrella containing sword blades?)
• - viral organisms (unless they are inside a paying passenger, I imagine)
• - weed killers
• - welding gas

Conspicuous in their absence are machetes, which are more dangerous than
screwdrivers. Trust me.

The list of what is allowed is equally odd:

• - cigar cutters (anyone who saw Larry Drake cut a dude’s finger off with one in Darkman would not be keen on having these on board. Yes, the same Larry Drake who played Benny on LA Law)
• - pickles (maybe Heinz lobbied for this)
• - “reasonable” reading material (what does that mean? No Ann Coulter books?)
• - rugs
• - up to one gallon of liquor in each container (a gallon of Bacardi 151 would probably propel the plane for a few miles)
• - up to four pounds of dry ice (if put in water, would force a pretty quick landing)
• - “unlimited numbers” of unused syringes

Some of the stuff is so obscure that I would not even know if I was carrying it:

• - compound 3 weed killers (are compounds 1, 2, and 4 OK?)
• - coshes
• - eyelash curlers (aren’t eyelashes already curly?)
• - explosive auto alarms (huh?)
• - knuckle-dusters
• - kubusaunt
• - kubatons
• - kukris (I must have slept through the Ks in vocabulary class)
• - leather dressing
• - rice flails (doesn’t sound so dangerous)
• - skeandhus
• - slurries

You would think that confiscating dangerous things at airports would
be pretty rare, but it isn’t. On eBay, TSA sells Swiss Army knives and
Leatherman tools in huge lots … 50 or 100 at a time. There were
about 12.5 million confiscations in 2005, including 8.1 million lighters
(that’s understandable), 3 million sharp objects (sort of understandable)
1.6 million knives and blades (less understandable), 819,000 tools,
362,000 flammables and irritants, 21,939 instances of ammunitions
and gunpowder (wow!), 19,499 box cutters and 19,183 clubs, bats
and bludgeons.

I don’t think that box cutters would be a super effective way to take
over a plane any more (they stopped being effective after the first
few attacks on September 11th, when the heroic passengers fought back
and the fourth plane went down in Shanksville, Pennsylvania). But
damn, 19,499 people tried to board a plane in 2005 with box cutters, and
the year isn’t over yet. That is absolutely amazing.

Some of the rules in India are amazing: “passengers are permitted to
wear their head covering including those who wish to keep their faces
covered during the screening process” and “Sikh passengers are
permitted to carry a religious Kiran [up to] 9 inches in length”.

When you add up all the exceptions, you can have a passenger wearing
a mask, carrying a 9 inch knife and a parachute, 100 syringes and a
gallon of grain alcohol. As long as he’s diabetic and religious. What’s up
with that?

Just after the first Persian Gulf War, I flew to Kuwait on Kuwait
Air. It was absolutely the most thorough security check I have ever
seen. They made me drink out of a water bottle I was carrying, and
they just about destroyed some nice gifts that fellow passengers
were carrying for their friends. That airline had just lost 75% of
their airplanes when they were invaded by Iraq, and I guess they
did not feel like losing any more. Still, I could not help but notice
how zealous the security guards were; much more serious than the
disengaged ones often seen at US Airports in those days. Then, after
takeoff, I discovered the reason why: the security guards who had
manned the final checkpoint were on board the flight with us. With
that much skin in the game, they were highly motivated to make it a
safe flight. They were also wearing parkas on a flight between tropical
cities, which was a little odd. I am assuming this was covering some
bulletproof clothing and who knows what hidden guns.

The list of prohibited items for the last US presidential inauguration
was pretty short but covered the gamut: lasers, mace, firearms and
alcoholic beverages (that’s fine), and posters, signs, and placards (that’s
not so fine in a democracy).

No Comments

When I was a student, I twice worked on projects that were about
seemingly innocuous topics, but which made a connection to what most
people would define as stereotyping.

This posting could be a lightning rod, so let me get a few things out of
the way. I consider myself a very open minded person (unfortunately,
almost everyone considers themselves to be well above average in open
mindedness and in driving ability, and we know this can’t be true for
all of them. But for me, it is true. No, really ). I have been to exactly
20 countries, and have met mostly nice people plus a few jerks in all 20
of them. I have friends of every major religion, political orientation,
sexual orientation and IQ. I am an active voter, but neither a Democrat
nor Republican, and although I have strong opinions about the current
occupants of the White House, I think that Democrats and Republicans
have had a similar number of great and stupid ideas when averaged
over the past 100 years.

The first project showed me that pattern recognition and
stereotyping are uncomfortably close to each other, and the second
project showed me that cultural sensitivity and stereotyping
are uncomfortably close to each other. My junior year, I was taking a
graduate seminar on Artificial Intelligence, and I decided to build a
simulation of some ideas found in Brains of Men and Machines, and
Brains, Behavior and Robotics. Many years later I worked for
one of the authors of those books, by coincidence.

These books talked about a form of neural net models called CMAC. I
programmed a pattern recognition system according to the ideas in the
books. It had primitive input and output, where I fed the patterns in a
text file instead of a fancy graphical interface. I would “teach” the CMAC
some patterns, and then later feed it some new patterns and see if it
could classify them. It did a decent job, just like any of today’s neural
nets would do a good job classifying credit risks, or picking race horses
(well, maybe not that), or guiding a cruise missle. The alarming thing was
that just by changing the labels of the data, and not the program itself, I
had a working simulation of a bigot. If I showed it a bunch of automobile
data, it might learn what a sports car and family sedan was like. But if I
changed the parameters from horsepower and place of manufacture to race
and gender, then it might make some guesses about profession. This was
sobering indeed, and I have never forgotten about that. After September
11th, the government embarked on a mission to build the Computer
Assisted Passenger Prescreening System, which is a fancy term for
an automated and expensive stereotyping system to guess who might be
a threat to an airplane. The people who built it probably just dusted off
some old code from Netflix’s movie recommendation system: Those who
enjoyed Caddyshack also liked Groundhog Day…. becomes Those
who bought first class one way tickets and carried box cutters are on
their way to meet 72 virgins…..

Nearly 15 years later, I was in business school, and we did a project on
cultural sensitivity for international business. What could be more politically
correct than trying to accomodate other cultures when doing business,
instead of being an “ugly American”? We used the book Do’s
and Taboos Around the World which despite the gimmicky title was pretty
interesting. My group put together a presentation, and when we looked at it,
it struck us that rather than being politically correct and sensitive, it almost
sounded downright offensive. The rules of thumb intended to promote
cultural sensitivity were stereotypes. They probably had validity in many
instances, but there is a very fine line between being sensitive to how
business is conducted in another place and some very negative stereotyping.
What if Country A were known for its punctuality, and Country B were
known for a more lazy attitude? It is easy to imagine both useful and
harmful inferences that come from this generalization. The useful ones
might be called “cultural savvy” and the harmful ones might be called
“bigotry”, but it is the same mechanism that is creating both of them.
It is something to think about.

People who have read alot of my postings so far have detected much
use of probability and statistics. I think that trust has to do with
accurately predicting probabilities of certain events, and then basing
decisions and policies with those probabilities in mind, to yield
maximum gain. And I think that people often fail to do that correctly,
and that’s why I write.

Imagine you noticed a pattern in a coin that was repeatedly flipped. Let’s
say the coin is slightly damaged and lands on heads 51% of the time, and
tails 49% of the time. You would have to watch it for a while to detect that
pattern, but that information would be useful to have. If you ran a casino,
such odds would be enough to make a decent profit over time, although
sometimes you would lose and have to pay out. To choose another gambling
example, suppose you noticed that a poker player always rubbed his chin
when he had a very strong set of cards. It could be that sometimes he
rubs his chin when he does not have a strong set of cards, but if you noticed
that he did it more often than not, it would give you an edge when playing
against him.

An enormous problem with stereotypes is that they can have predictive
power. That is why they are so difficult to overcome. They can be useful
tools to deal with the world, and that is why it is so difficult to get people
to stop using them. The downside is that they are also wrong much of
the time (just like our deformed penny is wrong 49% of the time), and
when they are wrong, they can do plenty of harm to the person being
stereotyped, and to the person doing the stereotyping.

The town I live in has more than 200 professional firefighters. Currently,
just one of them is a woman. That information can be used in all sorts of
ways, some of them good and some of them bad. If I met a man and
woman at a party, and knew that one was a schoolteacher and one was a
firefighter, I could guess with 99.5% certainty which was which.
Sometimes I would be wrong. If I was the architect of a new fire
station and I had a $100,000 budget for bathrooms, I could spend
$50,000 on the men’s room and $50,000 on the women’s room, or
I could spend $90,000 on the men’s room and $10,000 on the
women’s room. One of those plans will result in shorter bathroom lines,
at least for the next few years, but may be a discouraging sign to
women who are thinking about taking the firefighter test.

It is very unwise to be a slave to stereotypes, but it is equally dumb
to not acknowledge where they come from and why they persist. If
they had no predictive power whatsoever, they would cease to be used
(and probably be replaced by different, more useful ones, but not
just go away).

Humans are not great at intuitively figuring out true probabilities for
things, and therein lies a problem for all who hope to use trust correctly.
There are many names applied to roughly the same mechanism:
cultural savvy, bigotry, gut feel, sensitivity, stereotyping,
intuition, and pattern recognition.

Tread carefully.

No Comments

My mother told me a story yesterday about some friends of hers who
pay for their meals with credit cards, but leave tips in cash. They have
been reconciling their credit card receipts with the monthly bills, and
have found out that twice in two years, the waitron has added
a tip on the credit card in addition to the cash tip left on the table. This
is obviously inappropriate; there is no legitimate reason for the waitron
to alter the credit card bill even if the customers left no tip at all. So this
is bad stuff.

I would be pretty pissed if I discovered this happening to me, because
it is obviously more convenient to leave the whole thing on the credit
card, and the customer is going out of their way to leave cash as a
gesture of kindness. Either the customer believes that the tip will get
to the waitron faster, or in its entirety because of its being left in cash,
or they are helping the waitron to avoid reporting some taxable income
by doing this. So an act of kindness is being paid back with an act of
theft.

The question is, as in the case of counting money at the ATM, whether
it is worth being on guard for this sort of thing. I suppose my first
reaction would just be to leave the tips on the credit card in the future,
but someone could still doctor it easy enough, and turn that $11.00
tip into a $17.00 tip with one stroke of the pen. Two strokes of the
pen would make it an implausible $77.00 tip, but that’s more obvious.

Let’s imagine that people go out for meals for two that cost about the
same as an hour of their time. In other words, a person earning $20
an hour goes out for a $20 meal for two, and a person earning $80
an hour goes out for an $80 meal for two. Let’s be 20% tippers, so
the tip is $4 in the first case, and $16 in the second case. Let’s say
they go to a restaurant once a week, or 52 times a year, plus a little
more on vacation, or 60 times a year total. It probably takes one
minute to reconcile each receipt, and then 30 minutes to go argue
with the restaurant if you find something fraudulent.

So the $20 an hour guy spend a total of 90 minutes, assuming he
finds one episode of fraud a year. He’d probably be able to
convince the restaurant to give him the entire meal for free if he
busted them for this sort of thing (my mother’s friends did). So
he uses $30 worth of time to recover $20+$4+$4 (remember,
there was double tipping). It is a break even proposition. The
$80 an hour guy spends $120 worth of time to recover
$80+$16+$16, so again, it is about a break even deal.

Personally, I think that life is too short to be spending any time
reconciling credit card bills if it only means you are breaking even
on the time spent. But a rational way to handle this might be
to double check all credit card receipts for large meals (because
the return on investment for that time is greater), or meals from
restaurants that you do not frequent often (a restaurant you
go to all the time may be less likely to want to lose your business
with petty antics like this, although you never know for sure.)

No Comments

My grandfather wrote a newspaper column during World War II called “Daily Until
Victory”. Over the past year, I have been making prints of his old columns, which
are only on microfilm at this point. The column was seven days a week, and I
also found that he had many guest columnists helping him out on certain days.
I think it made for a more interesting column.

A friend of mine stationed in Iraq is a regular reader of my blog, and he’s been
corresponding with me about issues of trust. I will leave his name out of this until
he’s permanently back home.

In the spirit of my grandfather’s columns, the rest of this posting is directly from
Iraq. Thank you, sir. Your Trust Manifesto shirt was mailed last week.

“So I have been reading your blog, and it got me thinking about the kinds of trust I
see around me everyday. As a soldier in Iraq, I have a completely different environment
than most, but since I actually started analyzing what I was seeing, I determined that
it is not just here, but really all over the world. Hear me out:

And that is why trust is important. “

No Comments

A guy named Toshio Yamagishi has spent many years doing research
on issues of trust. He is particularly interested in differences in trust
between different societies, and gives prescriptions for what Japanese
society should do to prepare for the future. I have started to read one
of his books, and it is completely fascinating. I am sure that many of the
experiments he describes in the book will be material for this blog.

In Yamagishi’s words:

It is well understood that a society cannot function if there is no trust among its
people. Trust is a lubricant of social relations, making relations possible among
people and organizations. An absence of trust will reduce greatly the efficiency of
interpersonal relations, including social and economic relations. In this sense,
trust is privately possessed social capital making our society and easy place to
live.

He goes on to write “the central message [is] that collectivist society produces
security but destroys trust.” He defines a collectivist society as one in which
group members cooperate at a much high level with each other than with
people outside of the group.

This thesis sounds paradoxical at first, but the more of his book you read, the
more it starts to make sense. Examples of collectivist societies (using his
definition) would be organized crime families, members of New York’s Diamond
Dealer’s Club, OPEC and Japanese businesses.

He talks about how in Japanese business, there is alot of time spent up
front building the trust relationship. That is what permits entrance to the
collectivist society. The Japanese businessperson prefers to do business
with one of the other members of the collectivist group. Transactions are
more efficient than in less trusting styles of business, and huge deals can
be executed within the collectivist society with just a phone call. This
sounds very much like the Diamond Dealer’s paper, which I have read but
not yet written about here.

Yamagishi writes “… people who are immersed in secure relationships
with compatriots … have difficulty forming trust toward people in general.”
He goes on to say that relation fortification has been the main
focus of trust research in the past, and that his book focuses on relation
extension. That’s interesting to me, because the Trust Manifesto is
not about being trustworthy, or about trusting the people you have always
trusted. It is about carefully analyzing when trust should be extended to
others who you are uncertain about. I have been using the word extended
for many years when describing my idea on trust to people, because it is exactly
the right word, and I am glad to see the related word extension in Yamagishi’s
work. (To be fair, the book was translated from Japanese, so all we know
for sure is that the translator and I think alike.)

Yamagishi says that the role of trust is to be a “constructor of spontaneous
relations”. Bingo!

The book goes on to talk about how the stable relations among trusted
partners have benefits, but typically hinder the amount of trust that is
extended to outsiders. In economic terms, the closed organization lowers
transaction costs but increases opportunity costs. In the Japanese
business example, the auto maker knows that his steel supplier
is completely trustworthy and will honor an order that was delivered
over the phone, but he does not know if he is doing business with the
most efficient steel supplier that might be available.

There will be much more from this book, and you can download a
copy yourself.

No Comments

My friend Gary has this brilliant observation about how many problems
in life are attributable to people making an assumption that other people
think exactly like they do. It does not sound all that profound at first,
but once he said it to me, I noticed how often it explains something that
is going on in the world.

If Gary’s observation is true, then it would imply that you should be
really careful of people who surround themselves with barbed wire
or demand that contracts be signed for trivial things. Maybe they are
very honest people who have been burned too many times, or maybe
they are paranoid, or maybe they are dishonest people who believe
that others will act like them.

Likewise, I think that I would be more likely to trust someone who
is extremely trusting, not because I want to return the favor, but
because they (perhaps naively) are thinking that I am going to behave
as trustworthy as them, and that’s a signal to me.

Think about this theory; it is applicable to many things. It is a huge
blind spot in human behavior. Think about how many businesses
fail because the owners think that the general public has exactly
the same taste that they have. Think about how many armies
have been decimated because they thought that another army
was going to think like they thought.

No Comments

There is a term in economics and finance called simultaneous transaction which some
consider a synonym for riskless transaction. I suppose there are at least four ways in which
a trade could happen: I could pay you for something, and you later send it to me, or you
send me something, and I later send payment for it, or we could swap the item and the
payment at the same time, or we could use an intermediary that we both trust more than
we trust each other, and they could serve an escrow function, meaning that once both
payment and goods were received, they would be released to their destinations.

Almost all transactions on eBay follow the first model; payment is sent and then the goods
are sent. I think that people are very used to that model because that is how mail order
has worked for 100 years. An individual would send payment to a big firm such as L.L. Bean,
and L.L. Bean would then mail the goods. In general, a big mail order firm with a reputation
to protect would seem more trustworthy than an anonymous individual customer. I am
sure that there are plenty of exceptions where money was sent to Post Office boxes and
no goods were ever shipped, but buyers tend to be more wary sending money in situations
such as that one.

eBay carried over this traditional payment model, but it seems sort of arbitrary,
because either a buyer or seller could have the higher feedback number on eBay. I
have often thought that on eBay the party with the lower feedback number should
send the payment or goods first, regardless if they are a buyer or seller. (Escrow
services on eBay start at $15.00, and that seriously cuts into the efficiency of the
transaction for all but the highest priced items on eBay.) I recently bought a lens on
eBay, and my feedback number was about 535 and the seller’s number was about 6
when the auction started. In this case, I would have been more comfortable buying if
the seller had shipped the lens to me before payment, and the seller should have
little to worry about because of my high feedback number. (The lens arrived and is
absolutely perfect, by the way.)

L.L. Bean was founded in 1912, but decades before that, Montgomery Ward and
Richard Sears had mail order businesses. In those early years, the businesses did
not have the advantage of a long track record, and they actually sent goods and
got paid later.

Cynthia Crossen wrote about it in the Wall Street Journal:

Both companies knew Americans would likely balk at paying strangers
in advance for things they hadn’t seen. So for many items, Ward and
Sears demanded no money until customers had received the items and
judged them satisfactory. Sears himself devised the pithy slogan that
relieved the anxieties of so many prospective buyers: “Send No Money!”

One year, a menswear sale almost bankrupted Sears. So many people
ordered suits, without deposit or obligation, that beleaguered Sears
employees simply threw any suit, regardless of color or size, into any
box and shipped it off. Then came the returns and letters of complaint –
so many that some were eventually disposed of by incineration.

Yet by 1915, Sears Roebuck, which began with $150,000 in capital, listed
assets of more than $100 million.

It is odd that Sears was so successful with a model that is precisely the opposite
of how eBay normally operates. Both models require trust, but on opposite
sides of the transaction.

When I sell treasures on eBay, I sometimes mail the item before I receive payment.
That happens less often now that Paypal is so prevalent a payment mechanism,
but I used to do it all the time while waiting for people to mail checks. It would
absolutely boggle the buyers’s minds, yet it is no sillier than what they were
doing, namely sending payment to me without having the goods. Sometimes the
item would get to them before they had sent the check out, and they would
round up to some higher amount. In fact, I only once got burned on a transaction,
and was overpaid more frequently than you might expect.

If my wife reads this posting, I predict she will post the comment: “And even if the people
never sent a dime, at least you got the shit out of my house.”

No Comments

An article in the New Yorker talks about a restaurant in New York that had a
“pay what you want” philosophy. Of course some people ate giant meals and bolted without
even leaving a tip. Some paid $200 for a meal that was worth about $80. The thing that was
surprising to me is that the concept is not a novel one.

I found a pay what you want restaurant in the suburbs of London called Just Around the Corner,
which has been there for 17 years, a similar restaurant in Singapore called Annalakshmi,
another one in Salt Lake City called One World Cafe, a pay what you want winery in Berlin,
and a chain of restaurants called Govindas that at times has used this concept.

Such a restaurant probably gets the equivalent of thousands of dollars worth of free
advertising a year, just from people speculating in articles about how stupid an idea it is,
and that must offset most of the cost of the meals that people run away without paying for.
And remember, just as in the bagel case, the true cost of those meals is closer to the cost
of the raw ingredients than it is to the retail cost of the meal. (If the restaurant runs out of
tables that would have held high paying customers or out of raw ingredients to cook for
paying people, that is a true loss of revenue, but otherwise there is no cost beyond the
ingredients.)

No Comments

“I know God will not give me anything I can’t handle. I just wish He didn’t trust me so much.”
— Mother Teresa

No Comments

About 10 years ago, our fully decorated Christmas tree toppled over. I can’t remember if it
was the cat that had climbed up it that time, but that had happened in the past. I was furious,
and vowed to get the best damn tree stand on earth, no matter what it cost. My brother in law
and I jumped in the car, and went to the closest hardware store. They were sold out. We went
to the next place and the next place, and they either had the same old sucky ones, or were
totally sold out because it was so close to Christmas.

We finally found a gimmicky looking plastic one, where one piece clamped onto the tree
itself, and it would set into a huge circular base that could not tip over under any imaginable
circumstance. There was a foot pedal which allowed easy adjusting of the angle (by one
person, no less!) In short, it was perfect. It was some preposterous amount of money,
but I didn’t give a damn. I would have sold my car to pay for it at that point.

The thing only has two pieces, and the second year, one of the two pieces broke. OK,
so it wasn’t so perfect. I called the manufacturer, and they said “do you need the top
piece or the bottom piece?” I told them that I needed the top, and they asked my address,
and that was it. A new one was whisked to me, at no charge, with no proof required
that I had ever bought one of their products, let alone that I had bought one recently.

Are you thinking the same thing that I immediately thought? If I had a friend call
a few days later and tell them that his bottom broke, I’d have a complete and free
and almost perfect tree stand.

I’ve been Googling like crazy, and cannot find this exact stand for sale anywhere.
I’m guessing that over the past 10 years, a million hoodlums called up the company
and asked for 500,000 replacement tops and 500,000 replacement bottoms, and the
company went out of business.

Seriously, this is about the most liberal replacement policy I have ever heard of.
Giving away half of an $80 item for free without any documentation is really, really
trusting.

1 Comment

Look what sometimes happens when you leave your car unlocked.

Heart-broken stranger gives away $15,000 ring
Anonymous man puts jewelry in unlocked car

Friday, December 16, 2005

BOSTON (Reuters) — Are diamonds really forever?

An anonymous gift-giver left a $15,000 diamond engagement ring to the
owner of an unlocked car in western Massachusetts with a typed note
hinting at a broken heart.

“Merry Christmas. Thank you for leaving your car door unlocked. Instead
of stealing your car I gave you a present. Hopefully this will land in the
hands of someone you love, for my love is gone now. Merry Christmas
to you,” the note said.

The three-diamond ring with a white-gold band appeared on the seat
of the man’s car at a train station in Westborough, about 30 miles west
of Boston, on December 7, police said. Four days later, the man reported
it to police.

“This appears to be random,” said Westborough Police Lt. Paul Donnelly.
“I think there was a search for a car that was unlocked.”

The 37-year-old man decided to keep the ring after a jeweler appraised
its value at $15,000, police said.

1 Comment

“Love mankind, trust the majority, never owe anyone.”
— Fortune cookie

No Comments

Here’s how the second wealthiest American manages, according to
an article in the November 12-13, 2005 Wall Street Journal.
(There is a weekend edition now, hence the schizophrenic publication date.)

None of [Buffett's] conversations were lengthy. He spent most
of the time listening, not advising. “A few of the Berkshire CEOs
think that if they talk to you, you’ll tell them what to do,”
Mr. Buffett says. A prerequisite to a Berkshire purchase of any company is
trusting that company’s managers to make decisions, he says.

Buffett’s philosophy makes infinite sense, but is not ubiquitous. Maybe
that is a factor in why he is so wealthy and others are not. Shouldn’t
trusting be a prerequisite for hiring or promoting anyone? Why would
you put someone in a role like that unless you trusted them to make
decisions? Admitedly, some roles require different amounts of trust.
Pouring ice cream cones is different from being a bodyguard for a
governor. But for any given job, it seems as though the people should
be trusted a priori. And yet companies spend all kinds of time and
money to have so called “checks and balances”.

Buffett continues:

Charlie and I are the managing partners of Berkshire. But
we subcontract all of the heavy lifting in this business to
the managers of our subsidiaries. In fact, we delegate almost
to the point of abdication: Though Berkshire has about 180,000
employees, only 17 of these are at headquarters.

Charlie and I mainly attend to capital allocation and the care
and feeding of our key managers. Most of these managers are
happiest when they are left alone to run their businesses, and
that is customarily just how we leave them. That puts them in
charge of all operating decisions and of dispatching the excess
cash they generate to headquarters.

Most of our managers are independently wealthy, and it’s
therefore up to us to create a climate that encourages them
to choose working with Berkshire over golfing or fishing. This
leaves us needing to treat them fairly and in the manner that
we would wish to be treated if our positions were reversed.”

I would say that until they get close to retirement age, most
employees do not have the choice between their employer and
“golfing or fishing”, (although if I did, I’d probably choose my
job over golfing or fishing, which I find tedious). But all employees
do have the choice between their current employer and some
other one, whether they like to think so or not, and therefore
at least a subset of Buffett’s philosophy needs to be considered
in virtually any work situation.

No Comments

I recently shipped an eBay item to Italy, and the item was very heavy, so the buyer
chose to ship by surface mail. This means that the package goes on a ship apparently
going one tenth the speed of Christopher Columbus. After seven weeks, he and I
decided that the package was lost forever. I seldom use postal insurance (more on
why in a future posting), but in this case, I decided to because of the value of the item,
and my image of longshoremen in Port Elizabeth, New Jersey and Naples, Italy
opening the box to see what was in it.

So I had to learn about how to file an insurance claim with the United States
Postal Service. The process is set up such that it would be worth almost no
one’s time to follow through on such a claim. The forms are only available
in a post office rather than online. Only one person at my local post office
is authorized to hand out the form for some reason, and she goes to lunch
when I am on my lunch hour, so it is difficult to see her. Once you get the
form, it has to be filled out in quadruplicate by pressing very hard
with a pen.

But the part that was really distressing and silly was that you have to show
a receipt for the item that you sent to prove that it was worth what you said
it was. In my case, I was not the original owner of the item, so I did not have
a receipt for it. The guy in Italy won the auction for about $150, and I insured
it for $150, and that was that. I asked three people at two post offices about
that case, and they mumbled stuff about not knowing what would happen.

I think that the USPS is making a big mistake here. In the age of desktop
publishing, I can clearly cook up an authentic looking receipt for just about
anything, in any amount I choose. So the receipt is proof of nothing, and
only serves as an inconvenience for honest folks. But more importantly,
what is the insurance premium that I as a consumer am paying? It is
effectively a wager. It is a bet on my part that the USPS will lose my
package. By accepting the insurance premium, they are betting that
they will not lose the package. The USPS knows enough about its loss
rate that it can set the insurance premium and make a small profit
on that additional service. What is actually in the box is immaterial.
If I shipped a box of air, and wanted to make a wager that the USPS
would lose it, I could pay an insurance premium of $20 for coverage
of $2000. In other words, I think that there is a greater than 1% chance
that my box will be lost or stolen, so my expected return is greater
than $20 on the average. The post office, on the other hand, believes
that the rate of loss is less than 1%, or else they would be charging
a higher insurance premium. What is actually in the box is completely
irrelevant. If there is a $2000 bet riding on that box, then the Post Office
should treat it as such. In fact, their job is even easier in the case of a
box of air, because it is subject only to theft and not breakage. If the
Post Office is that confident in the security of the mail that it knows
it will make a profit from the insurance, then it should encourage
people to overinsure their packages. It could double its profit on
insurance services if everyone declared their items to be worth
200% of their true value. It is totally unimportant what the actual
value of the items is. There is the ethical issue of lying, of course. A person
would be lying if they said their $300 camera was worth $600. But
since they bought $600 worth of insurance on it, the post office makes
extra profit in the more than 99 times out of 100 when the package
does not get lost. No need to waste time with receipts, phony or real.

Just as I was about to submit the claim for to the post office, an e-mail came
from Eugenio in Italy. The box had just arrived. Bravo!

No Comments