Wednesday, 17 April 2019

A Review of 'Priceless' by William Poundstone

This is an interesting book about price.

Its written for business people. Chapters are short and punchy with a form as regular as poetry; a set-up, investigation, explanation and then musing on meaning.

"One of America's longest-running Guy Grand pranks takes place every day just off 1-40 in Amarillo. A giant steer statue us blazoned with a sign advertising a FREE 72 OZ STEAK."

From here we descend into the Gospel of Poundstone, or more like the Torah of Poundstone as the Gospels themselves are micro stories of the lives and investigations of academics, economists, psychologists and business people investigating the nature of prices.

There is a study, or a series of studies, or a 'latest study', which takes place in a lab, in the U.S. or Israel. These labs are like temples of logical knowledge which issue out these subtle riddles of behaviour or thought which are meant to reveal something about humanity and the way it thinks. In one scene a guy in the U.S. constructs a room in a building that _invisibly tilts on its axis due to hidden workings_, this is to measure if or how much people are aware of their actual balance without meaningful cues. The room is set to lure people in with fake free eye tests and then to move around secretly and see if they notice.

(People sense something is wrong, but they often don't relate it to movement.)

The riddle is considered, examples are given.

Some thought is dedicated to the overall meaning of the consequences should the riddle be found to be a full truth. But only for a few paragraphs. The chapters are short, like blog posts, magazine articles. They are *punchy*. Constructed like SEO-optimised trade-speak clickbait  blog posts. Artfully constructed; bit of human interest, bit of academic soap opera, bit of theory, dab of mathematics, knowing tie-up at the end.

So those are the onrushing pebbles of Poundstones argument. They never entirely or purely cohere into a directly-stated synthesis or final statement. But it is not hard to work out his meaning. His argument is that the prices for a lot of things are very close to being imaginary.

You probably have a rough personal intuition of how imaginary or made-up most prices are; well double that, or maybe triple it, or increase it by a factor of something. It is all light and air, or so it would seem.

Humans judge by ratio very keenly and are totally bonkers when it comes to abstract values. If you want to change someone’s thoughts about pricing, change the ratios that are present when someone thinks about it. Therein lies almost all of everything anyone ever did to find a way to raise a price.

Humans assign value by primate-relevant qualities. Considering Price, alone, deeply, actually brings you back into contact with the whole of the human experience, because when assigning an abstract value to something, it seems that what really matters is human context. People draw from their entire physical and social experience of the world when negotiating with each other.

In particular, people are much more loss-averse than they are success hungry, and they are much, _much_ more averse to letting someone treat them in an 'unfair' way than they are desirous of getting free, or extra, resources.

So people will literally turn down a free lunch if it comes with an assumed loss of relative status or participation in a deal in which someone is seen to have an unfair advantage over them, even if in rational absolute terms they only gain by the exchange.

So, 'value', though seemingly abstract, is like a puncture hole though the skin into a deeply human world of social networks, personal status, testosterone levels, fear of loss, desire to impress and just a very apelike, living, human experience of reality.

Heuristics are a thing. Humans use simple, usually ratio-based intuitive shortcuts based on environmental factors in order to work out the value of thigs. When considered singly, isolated away from their ecology of thought and zoned-in on in a lab, all of these heuristics break down. They are 'non-logical' and can be made to be wrong with tricks of logic.

The question of how powerful this effect is in real-life is a primary moral axis of the book. Unusually, Economists and Humanists are on the same side in this one, they both want the decisions people make to 'make sense' and to be fundamentally rational, or at least, reasonable. Even if they break down in the lab it is hoped human decision making works 'as a whole'.

Poundstone seems to be on the other side, the general drift of his argument is that prices are deeply and worryingly ethereal and much of human value-assignment is basically a holodeck episode.


Priceless absolutely abounds in factoids, little slivers and pieces of counter-intuitive, strange or just interesting discoveries and secrets. Here are a few;

There is/was a weird culture war/intellectual war between Economists and Psychologists. Economists seem to move in an ethereal realm of pure figures and absolute values on which the great machinery of their thought is based. Psychologists are a bit dirty, manipulative and fluid, they chew away at the absolute values Economists need to build their palaces, revealing them to be little more than dreams.

So in the partial history of the relations between Psychologists and Economists that makes up a strand of the book, Economists have a habit of just straight-out dismissing anything Psychologists say with "Oh, _psychology_)", at least for the start of the book. So that is an interesting example of subjects or cultures deciding to just not talk to each other. Or at least, not to listen.


'Anchoring', shaping assumptions on price by opening with a super-high figure, works much much much better than most people intuit it does. It works on the average Joe, and it works on experts, it works on people who decide quickly, and it works on people who consider deeply. The depth to which you have thought about something, alone, does not meaningfully change the extent to which Anchoring will work on you.


"Fisher aspired to predict prices from supply and demand. His thesis described how to do that, and Fisher went so far as to build a price-generating machine (see page 225). It was a tank of water with a flotilla of half-flooded wooden "cisterns" connected by a system of levers. Adjustments to "stoppers" and levers fed in data on incomes, marginal utilities, and supplies; then prices could be read off scales. Gibbs must have been pleased. The device prefigured, if not parodied, the direction of twentieth-century economics. ("Press stopper I and raise III," read part of Fishers instructions for the thing. "I, II, III now represent a wealthy middle class and poor man respectively...")


"I absolutely thought [pay] would go down because the disclosures would be so embarrassing," recalled Graef Crystal, an architect of those disclosure rules. "But it turned out that when somebody is hauling in $200 million, he's not embarrassable."


"After years of work-around like half-dozen cartons of eggs and ever-shrinking containers of milk, it bit the bullet and raised its top price to $99.99. For president Jeff Gold, it was almost like a death in the family. 'The number 99 is a magic number - deviating from that is something we absolutely are not taking lightly,' he said. 'I find significant discomfort emotionally about considering making the change.'


There is much much more, its something of a wunderkammer of these, more than it is a book almost.

There were layers to my response to 'Priceless'. I often found it 'clever', in a mildly frustrating way. The glib single-idea 'gotchas' of business writing or magazine articles. I intuitively strongly dislike what I perceived as a mild wash of slightly sneering superiority through the opening parts of the book.

Under that, its actually interesting. The basic deal is that price, which bills itself as a firm and near-abstract unit of measurement, is a deeply human construct. On the down side this means the world economy is largely a collective hallucination and could collapse or change hugely at any time. On the more positive (to me) side, Economics is not just a part of the Humanities, it’s actually a Romance. Piercing through the needle-dot of price reveals a cosmos of human emotion, human feeling and human social relationships, illogical, 'irrational', but making organic sense, not without reason or meaning and if not entirely within our control then at least within our capacity to understand.


  1. Prices represent best guesses about the uncertain future. They are shaped by relatively objective facts, of course, such as current stock levels and expert endorsement, but at a fundamental level how could prices every be anything other than ethereal, consensual agreement? Why do you find this deeply worrying? (Prices have this property no matter they way price discovery occurs; markets aggregate best guesses of people making transactions while regulated prices represent the best guesses about the future by technocrats.)

    1. (Man, I need to proofread better on Blogger: *how could prices EVER be anything... *no matter THE way price discovery occurs...)

    2. I suspect that Poundstone would disagree with you about the 'relatively objective facts', especially the experts endorsement.

      I don't like it when things dissolve into air. I want the world to make sense, be solid and provide a predictable future for the people in it. If everything is just made up then that's frightening.

      This is an old back and forth between us.

      Imagine any actual, living human in any actual high stress situation like a flood or famine. You are haggling for a tin of spam. They make and offer. You say; "Prices represent best guesses about the uncertain future. How could prices every be anything other than ethereal, consensual agreement?"

      How long until you get stabbed?

    3. To clarify, by expert endorsement being an objective fact I mean that an expert did in fact make a particular endorsement rather than that the endorsement itself represents something true about the world. For example, about The Worm Ouroboros you wrote: I found this to be an exceptional book. Your statement of approval could change the book's price for people that trust your opinion by increasing willingness to pay, increasing demand, changing forecasts of future supply, and so forth.

    4. Perhaps there are better ways to understand barter exchanges during stressful encounters than pricing.

      How useful is it to reflect on the negative reinforcement value of pain for learning when touching a hot stove?

  2. Well, some Economists use the methods of Psychologists, reinvent our wheels and call it "behavioral economics" ...

    Yeah, that style is pretty much the way social science gets transmitted to the masses since Gladwell.

  3. There is a solid, objective price for things. It's whatever price the merchant and customer agree on for that individual transaction.

    Trade is nothing more than human beings interacting non-violently with people they hate; concrete, physical, measurable events occurring in the real world. But once you start talking about what a price, or anything else, "ought" to be, you are leaving the realm of science and entering into the realm of philosophy. Economics straddles that line and often fails to notice which side they are on when drawing conclusions.

    1. (Apologies in advance for a long comment.)

      @james how do you interpret a willingness to pay experiment where people are randomly assigned to conditions with high and low numerical anchors that finds people presented with (unrelated) high numbers being willing to pay more than people presented with (unrelated) low numbers?

      One example from a field study: ... The presence of an incidental price elevated the average bid among shoppers from $7.29 (median = $7.50, standard error = $.54) when the sweatshirt was priced low ($10) up to $9.00 (median = $10.00, standard error = $.64) when the sweatshirt was priced high ($80). ...

      Nunes, J. C., & Boatwright, P. (2004). Incidental prices and their effect on willingness to pay. Journal of Marketing Research, 41(4), 457-466.

      (A web search should allow you to find the full text of that paper if you are curious about any details. This effect is widely documented both in lab and field experiments.)

      I am responding to your first paragraph, the claim that a price agreement is solid, rather than your second paragraph, about oughts. (I see no ought claims in Patrick's post or any of the comments so far.)

      Your note that prices are somehow bound to real transaction events is good though, and is why prices for goods in illiquid markets are less defined than prices for goods in markets with frequent transactions. If transactions never happen, a market participant cannot consider prospective opportunity costs, something that can occur if an expected price is known even in the absence of the participant engaging in a transaction themselves.

  4. Anchoring is one of a long list of factors that influence the price people are willing to pay. It sucks as it makes us prone to manipulation, but it’s part of being human. And many of the other factors that contribute to agreed price are just as arbitrary and unrelated to the value of the product. When it comes to more traditional negotiated transactions, as opposed to modern fixed price transactions, stupid things like height, beauty, relative social status, and the buyer’s and seller’s charisma stats are big factors in determining the price paid. Trade is human interaction and humans are messy and not 100% rational. There’s a butterfly effect involved: There are so many factors involved in a transaction that you can’t reasonably measure all of them, or even identify all of them, before the transaction takes place. Which is why, as you pointed out, we have more trust in prices in markets that have more transactions, and spend less time making a decision to buy eggs than to buy a car or house. And so the only real price is what was paid.