Evan's reading list

Irrationality by Stuart Sutherland.

This is a classic book, and it’s great to see it is being re-published this month, nine years after the author himself died.

The book simply documents some of the common quirks of human behaviour that can be described as “irrational”.

It’s full of short examples of psychological tests and their implications, and it is an especially good read for economists who tend to assume that people behave according to some simple precepts of rationality.

So here’s an interesting but familiar example: the sunk cost error (outlined on page 71).

You visit the theatre but have forgotten your £15 ticket. The box office refuses to replace your ticket, but as the theatre isn’t full, offers to sell you another one for £15. Should you buy it?

The rational answer is probably yes. If the play was worth £15 before you visited the theatre, it is surely still worth £15 and so if the choice is whether you watch the play for a £15 sacrifice, consistency demands you should choose to watch the play and thus buy the second ticket.

You should treat the forgotten ticket as a sunk cost - it’s gone whatever action you take now, and so is irrelevant to the choice of how much you pay and whether you see the play.

But - as has commonly been observed - people don’t always follow this logic. They reason that the play is costing them £30, the price of two tickets, and if it isn’t worth that much, they choose not to buy the second ticket.

As the Sutherland book notes, this kind of mistaken thinking has far more significant consequences than missed theatre plays. Generals persist in following failing strategies, investors hang on to falling shares, waiting to recoup their losses.

Sutherland’s book is packed with many, many different types of irrationality, and by alerting us to common forms of misguided reasoning, it arms with some protection against our own predictable errors.

But if the book persuasively demonstrates that we do make systematic mistakes, does it debunk the whole subject of economics, a science that seems to rely on rational economic man?

The answer of course is no. The point of the assumption of rationality in economics is not that it is accurate. Of course not. The reason we make the assumption is that it makes life easier… it simplifies our understanding of the choices we make.

The assumption of rationality in economics is like the map of the London Underground: it strips out the complexity of the system, to illuminate the essence of it.

Of course, in practice we err, but it’s still useful to know what we would do if we were rational.

And then we can read books like Stuart Sutherland’s to understand some of the ways life is in truth more complicated. Just we look at proper street maps once we’ve navigated our way round the Tube system.

Irrationality, Stuart Sutherland.
Pinter&Martin £8.99

Comments   Post your comment

  • 1.
  • At 02:04 PM on 12 Feb 2007,
  • Chris wrote:

I'm not convinced about the theatre ticket example. Most of us have budget constraints, and spend our disposable income on a range of goods and services that we consider optimal. Imagine you pay £15 for the play and lose the ticket. You also planned to go out for a £15 meal the next day. Because your budget is constrained, by buying another ticket, you have to give up the meal.

In this way, under a budget constraint the second theatre ticket will always cost you more than the first, because you have to give up something else you planned to buy (and therefore, by definition, something you value higher than the money you planned to spend for it). It is perfectly rational to decide that the play was worth it at the original cost, but not at the higher cost of the second ticket.

In other words, in chosing between the play and the meal, you go for the meal. Nothing irrational about that! :)

  • 2.
  • At 03:03 PM on 12 Feb 2007,
  • Matthew Duncan wrote:

I like your article, but I am not sure if I agree with your- or are they Mr Sutherland's- conclusions.

If we assume that people have a limited amount of money, and can only afford to outlay a certain amount of that money on leisure pursuits, say £15, then it is quite reasonable to expect them not to buy a second ticket, which would take them over the threshold. Assuming they don't commisserate with themselves by going to the pub and spending another £15, they are surely practicing good economics.

This also holds true for other areas. A general in the battlefield (the example you cited) also has limited resources, resources which might be more effectively deployed elsewhere. Your sunk cost strategy, if I might allude to another one of your blogs might also constitute a 'moral hazard' as you have defined it, but I haven't really thought that one through.

  • 3.
  • At 03:24 PM on 12 Feb 2007,
  • Ian Kemmish wrote:

Is it rational to assume that holding on to falling shares is irrational?

If the shares were good value at £20, they must be even better value at £10, and therefore even more worth holding.

The fall in value *may* be an indication that perhaps you didn't do enough research before buying them, but that's a separate issue. To ditch them without doing further research is as bad as holding them without further research....

  • 4.
  • At 04:21 PM on 12 Feb 2007,
  • Andrew Lilico wrote:


I disagree. Rationality is not an "assumption" of economics in the way that you present it. Rationality, along with related concepts such as equilibrium, is an axiom. Economics built on these bases is not "physics without friction". Instead these axioms condition not only how we structure our models, but also how we interpret our data. No experiments can tell us that agents are not Rational, for the axiom of rationality is not a theory of human behaviour. It is a piece of conceptual architecture that gives our models structure.

It is absolutely indispensible to any disciplined research programme that we have such structure. For example, much modern consumer theory is illustrated by consideration of how it explains goods for which demand goes up when the price rises so-called "Giffen" goods - such as (in certain circumstances) bread or potatoes. But another explanation would go as follows: "Of course, if agents were rational, then rising prices would lead to demand falling. But clearly agents are not rational, as is illustrated by the examples of bread and potatoes." In this way, "irrationality" is not a disciplined form of explanation. It is an empty excuse for failing to have a theory of why - the grand conspiracy theory of economics, adduced to explain everything, but in fact explaining nothing.

Thus, your examples of the "sunk cost" fallacy represent a challenge for us to explain - we need to interpret this situation in other terms, e.g. by understanding the ticket-loser as having become annoyed by losing his ticket and gone off the idea of watching the play - preferring to sulk.

The Rationality-based models of standard economics are rich, insightful, interesting, and useful, and have enormously benefitted humankind. Critics must show that they have their own, alternative, discplined axioms, producing models that perform at least as well, before we should offer them a serious hearing.

  • 5.
  • At 04:55 PM on 12 Feb 2007,
  • Tim wrote:

I don't agree with your analysis of the theatre ticket example. When you are faced with deciding whether or not to buy the second ticket you are are not at the same economic standing as when you bought the first ticket - you are £15 poorer. If you are of limited means, being £15 poorer may well influence your decision. If you are whealthy, the fact that you are £15 poorer is less significant. The economic theory would predict that less well off people would baulk at buying a second ticket whilst the wealthy would buy the second ticket - I suspect that that is usually what happens in reality.

  • 6.
  • At 06:37 PM on 12 Feb 2007,
  • Bedd Gelert wrote:

As part of training in Banking I had to study some Investment as well as Economics, and the 'sunk cost' concept is a great one. As is the accounting concept of 'contribution' to profit on each extra item of widgets sold.

Often a 10% increase in sales = 100% of a company's profit. Tricky to understand until you think that 'Dixons' profit figure is often similar to the amount received on 'extended warranties'.

By the way, Evan, what is that concept where a large group of 'amateurs' can 'guess' the value of something more accurately than a far smaller group of 'experts' ? Nowt to do with economics, but would make a very interesting post if you could explain how [why?] that works. If indeed it does; it may be just an urban legend.

  • 7.
  • At 06:45 PM on 12 Feb 2007,
  • Dave Pritchard wrote:

Surely paying for a second ticket is only rational if the value of the second 15 pounds to you is the same as the value of the first 15 pounds. This is presumably true if it's a marginal cost, but not (for example) if you're a postgraduate maths student who can only spare 15 pounds per week for entertainment: the utility value of money then isn't linearly proportional to the amount of money. (Analogously, someone who would die within a week without a life-saving operation which would cost, say, 100 times thrir current savings might be rational to spend that money on Lottery tickets: their expected financial profit would still be negative, but the utility of this money only kicks in if it's enough to save their life!)

  • 8.
  • At 06:50 PM on 12 Feb 2007,
  • Rohan wrote:

Another issue of course is to assume consumers have the perfect information to make rational decisions. A large amount of the public policy thinking, especially in the UK and Australia, is based on choice. Choice is seen as being the driver of better outcomes (for the individual) and efficiencies (for the state).

However this is often delivered in markets where information is likely to be more limited or incomplete especially for certain groups of citizens (e.g. lower income) - they are therefore are unlikely to be able to make rational choices in such circumstances.

The anecdotal from the expansion of choice in the NHS (choice of hospitals) seems to show this in action, and the debate simplified without the true principle of consumer/citizen choice being tested.

Should we abandon the principle of providing people with rational choices about what services they use on the basis that some groups will have an unfairer 'choice' than others?

  • 9.
  • At 12:11 PM on 13 Feb 2007,
  • Andrew wrote:

Hi Evan - I was reading the above, and I thought I'd write to say I disagree.

Rationality is not an "assumption"
of economics in the way that you present it. Rationality, along with related concepts such as equilibrium, is an axiom. Economics built on these bases is not "physics without friction". Instead these axioms condition not only how we structure our models, but also how we interpret our data. No experiments can tell us that agents are not Rational, for the axiom of rationality is not a theory of human behaviour. It is a piece of conceptual architecture that gives our models structure.

It is absolutely indispensible to any disciplined research programme that we have such structure. For example, much modern consumer theory is illustrated by consideration of how it explains goods for which demand goes up when the price rises so-called "Giffen" goods - such as (in certain circumstances) bread or potatoes. But another explanation would go as follows: "Of course, if agents were rational, then rising prices would lead to demand falling. But clearly agents are not rational, as is illustrated by the examples of bread and potatoes." In this way, "irrationality" is not a disciplined form of explanation. It is an empty excuse for failing to have a theory of why - the grand conspiracy theory of economics, adduced to explain everything, but in fact explaining nothing.

Thus, your examples of the "sunk cost" fallacy represent a challenge for us to explain - we need to interpret this situation in other terms, understanding the ticket-loser as having become annoyed by losing his ticket and gone off the idea of watching the play - preferring to sulk.

The Rationality-based models of standard economics are rich, insightful, interesting, and useful, and have enormously benefitted humankind. Critics must show that they have their own, alternative, discplined axioms, producing models that perform at least as well, before we should offer them a serious hearing.

  • 10.
  • At 01:18 PM on 13 Feb 2007,
  • Ed wrote:

Does this mean that when you buy a £15 gig ticket (but not the £30 ticket to the same concert as it's not worth it), if the same ticket is selling for £60 a few weeks before the concert on the internet, you should sell it?

Andrew: I think what you say is a very intelligent -- and somewhat more sophisticated - way of looking at rationality in economics.

Of course, at all times we should be asking "what is the underlying model here?" -- and if our simple model fails to predict behaviour, then we should be asking ourselves what else might be going on.

So, when an army general (or President) persists in throwing away the lives of good troops, after a military campaign that has gone bad, we should not just assume he is irrational. We should ask whether there is perverse personal incentive for him to throw the dice again in a vain attempt to salvage his career, knowing that his reputation is down the pan anyway if it all goes wrong.

All true.

But this interesting question about the general's behaviour is prompted by the much simpler economics, of knowing the "rational" behaviour as naively assumed, and observing the deviation from it.

Until we can walk at that comprehensible level of behavkour, we can't run with all the more sophisticated stuff.

And on top of that, it is interesting by way of background to the economics, to at least know what common patterns of behaviour arise in a variety of psychological tests.

Anyway, very good stuff.

  • 12.
  • At 01:31 PM on 13 Feb 2007,
  • Neil P wrote:

I'd be interested to see the economists take on casino gambling, where games like roulette and blackjack offer a return of less than the real statistical odds or 'expected value'.

I understand that the desire to win is rational since, crudely speaking, more is preferred to less. However, it seems irrational to risk an amount of your cash when the expected return is perhaps 98% of your stake.

Surely the assumption of rationality goes out of the window for this whole multi billion pound industry?

  • 13.
  • At 02:41 PM on 13 Feb 2007,
  • Chris wrote:

Neil P, I may be on thin ice with the definitions here, but aren't the assumptions (if that's the word!) of rationality and risk averseness two different ones?

In other words, isn't it possible to be rational, just not always risk averse? Attempts at measuring individuals dislike of risk shows they associate different levels of cost with different size risks. So you might flip a coin over a beer, but not your house.

Does it become irrational if, at the lower end of the scale, risk averseness turns into risk seeking, and someone is actually willing to pay the casino to arrange some controlled risk taking for them?

  • 14.
  • At 10:08 AM on 14 Feb 2007,
  • Ed wrote:

Aside from the people that gamble as a fun way to expose themselves to risk, surely people are gambling on themselves more than the odds? You are betting that you are better at this game (say, poker) than your opponent, rather than gambling that the coin will come down heads.

  • 15.
  • At 11:25 AM on 14 Feb 2007,
  • Luke wrote:

Economics has no problem with what you call 'risk seeking' and what it dubs 'risk loving' behaviour. It is just as easy to model people who are risk averse with regards to money as it is someone who is risk neutral or risk averse, you merely have to adjust their behaviour with regards to expected behaviour.
For example, in a game of double or quites, a risk averse person will not bet, a risk neutral person could take it or leave it and a risk loving person will want to bet.

  • 16.
  • At 01:44 PM on 14 Feb 2007,
  • Joe Jarvis wrote:

Well, war is irrational for starters.

But i completely agree with the author on the subject of the ticket. The ticket has cost you £15, so in being at the theatre you have already lost £15, so paying a further £15 is rational, due to its value.
As for the arguements, "well they could be someone who only has £15" is rubbish, if you are going to be anal, then an arguement could be, "the parking costs £5 and the travel expenses were £10!"
Accepting the extra £15 is rational as its not a complete waste of time and moey.

Glad to see that Irrationality inspires so much discussion! Not sure if I'm allowed to post this, but there's a long extract on our website for anyone who may be interested in more examples of irrational behaviour...
All the best, Martin (publicity-seeking independent publisher)

  • 18.
  • At 12:59 PM on 15 Feb 2007,
  • Neil P wrote:

Some good points, I'm familiar with the concept of risk averse/neutral/loving behaviours, however my point was about the specifics of blackjack and roulette where the house has an edge over the player and the payout is roughly 98%, meaning the odds paid out are lower than the statistical odds of winning.

The above points about behaviours in a game of poker or 'double or quits' make perfect sense to me because these are 'zero sum' games, where all of the money gambled can be won by the players - no one external to the game is taking a cut.

The question I raise therefore is this: Is it not irrational for even a risk loving person to accept a gamble where the odds offered are less than the statistical chance of success? i.e. where they have no chance of winning in the long run

  • 19.
  • At 07:42 PM on 15 Feb 2007,
  • carol wrote:

Surely a rational decision is one that would benefit the individual? *Most* actions taken by individuals are 'selfish' in respect that the individual needs to gain something from the action that they do. Therefore if a person thought they would gain more from purchasing a second ticket than they would from not seeing the play, they would do so. However if they thought that they would gain more from spending the £15 elsewhere, they would do the latter. I appreciate this may be a rather cynical interpretation but is it not true, what was the last selfless act you did? Economics is based upon the behaviour of mankind and in the modern day world the behaviour is ever more self-orientated. An approach to analysis which should at least be considered?

  • 20.
  • At 10:52 AM on 16 Feb 2007,
  • Ben H wrote:

Surely, the only way to see if not buying the second ticket is irrational is to know the consumer preferences (utility function). The sunk cost represents only an income effect, and so the consumer will consumer less the theatre tickets (good x) and "all other goods" (good y) assuming that they are normal goods. If the budget constraint and indifference curves are such that utility is higher by no longer consuming any theatre tickets (as can very easily be drawn), then this is perfectly rational.
A far stronger example of "irrationality", though undoubtedly still with flaws, is the observation that people keep money in low interest savings accounts while owing money on high interest credit cards. To explain this, we must make use of behavioural economics.


You are doing a good job explaing economics to laymen although I am afraid only economists read your blog anyway.

You stress that rationality is paramount in economics. That is true but economics moves on and we have today many models which allow for irrational behaviour. Probably you could present some of this work in this place.


  • 22.
  • At 12:39 AM on 17 Feb 2007,
  • David Gwilliam wrote:

It is sensible to not pay for the second theatre ticket. You would not be in the mood to enjoy the play and you would be giving patronage to a business which had treated you badly.

  • 23.
  • At 04:17 PM on 17 Feb 2007,
  • Chris wrote:

Neil P, the point is that a risk averse person will demand to be rewarded for a risk (asking for a return higher than zero), while a risk loving person will be willing to pay for it (return less than zero). Hence he will accept that, once in a while, the house takes all on the roulette table.

  • 24.
  • At 06:24 PM on 21 Feb 2007,
  • David McDonagh wrote:

The very idea of irrationality seems inept, as all we do has some sort of rationale. The idea that we are irrational if we do not think in terms of opportunity cost is inept as any error will have a rationale; & maybe some future Von Wiester will come up with an insight that makes opportunity cost as inept looking as historical costs look today. The great epistemplogical problem leads one to think that mere error is no ground for thinking one is irrational, as any thought could be an error [for all we know]. we learn from error just because it does have a rationale. Thus irrationality is a mere expletive, referring to nothing, & is best dropped an an idea.

  • 25.
  • At 05:47 PM on 24 Feb 2007,
  • Arjun Sen wrote:

I´m not convinced either. Apart from the observation of one of the other commentators that the choice of paying again is likely to be influenced by disposable income, I want to pick up on the ´Generals following failing strategies´and investors selling in falling markets analogies.

I don´t think they hold true because the theatre show doesn´t have the same consequences. It is optional to survival.

For example, when Hitler persisted in forcing General Von Paulus to stay fast on the line in Stalingrad without adequate supply lines in the winter of 1941-42, he made a grave error. The German defeat at Stalingrad was the turning point on the Eastern Front.

Had Hitler retreated into a solid defence, as his generals counselled, instead of going forward willy nilly, the Eastern Front might not have crumbled as early as it did (1943-44).

Similarly, stockbrokers selling in falling markets, if they are holding good, solid stock, are being silly´and - ahem! - irrational. We all know that you have to hold on and wait for the market to recover - as it is certain to do. The analogy simply doesn´t hold. In the case of the theatre ticket, you get to watch the show at least. In the case of selling falling stock, you suffer an absolute and unmitigated loss. Someone else, following the rules properly, is going to buy your stock -and do awfully well out of it later on.

Only in really crucial cases is the sunk cost concept valid.

I prefer, for most normal situations, the old adage 'don´t throw good money after bad.

I believe that “irrational” behaviour occurs when we find limits of “conventional wisdom” approximations. Sounds weird? Let me explain. By conventional wisdom, I mean the tool box we carry around with us to make everyday decisions. Very handy. The problem is that conventional wisdom is a set of advice that is approximate. That is, it works - but only most of the time. Some times it doesn’t. This is inevitable since it could only work all the time it were a set of undisputed first principles.

We can’t use these in our daily lives since it would take forever to work things out. Just imagine deciding whether or not you should buy share options in your company at specially discounted rate. You’d have to know about option pricing (yawn), forecast the company’s prospects for the future and work out current market sentiment. Instead, you hear that the finance director has just filled his boots with them. Conventional wisdom says “follow the expert” and that you should buy too. It would be right!

The whole point of conventional wisdom is that it is a set of handy rules that you can apply quickly to any situation. Every common situation has a bit of conventional wisdom attached to it. If it doesn’t it is not common enough - otherwise we would have come up with one! Consider the theatre tickets example. Conventional wisdom: “£30 is a bit steep to go to the theatre”.

The irrationality occurs when we find a place where the approximation fails. In this case: “but that shouldn’t stop you going if you lost the first £15 on the way to the theatre”.

I have an irrational fear of flying. My conventional wisdom is: “If it feels dangerous and is out of my control – be scared as hell.” This approximation fails in a plane since I always forget the refining caveat: “… but not if it is in control of someone more competent than you”.

Why can’t we adapt each piece of conventional wisdom when we find out the areas where it breaks down? Well, we would end up with such long and complicated caveats that the whole usefulness of the simple rule would breakdown. “Follow the expert … but only if he can be trusted … not when he is the finance director and is pulling the wool over shareholders’ eyes … and is planning on cashing out tomorrow before all the employees buy in … etc” How would you remember all of that?!

  • 27.
  • At 03:04 AM on 28 Feb 2007,
  • JMW wrote:

Maybe they should offer you a ticket for less than £15, to make better use of their space? If you allow every ticket to include a reservation, then those who bought their tickets earlier will have payed more to have first choice.

  • 28.
  • At 12:55 PM on 06 Mar 2007,
  • Dan Sheppard wrote:

I found the "sunk cost" idea as being sometimes a misconception a very valuable idea at work, and quite a revelation. (Though in reality things are always more complicated, of course!).

Also I thought Andrew's contribution earlier was brilliant.

An economist's perspective on a matter is almost always valuable, but I do worry that in certain fields it's becoming the only perspective considered. Given its axiomatic simplifications (inherent in any discipline) I think that this can become damaging when it comes to dominate fields such as politics and social disciplines.

The biggest problems I see over and again is the assumption of total ordering of choices; the ability to conceptualize utility functions which extrapolate experience; and the ability to choose cost-free,

  • 29.
  • At 06:09 PM on 06 Mar 2007,
  • Robert Strawson wrote:

I don't understand the UK economy. We import a lot and provide much less than that in services and exports. Does the country simply go into debt more each month or have I missed something. If we just go into more debt surely there must be an end to this no one will lend any more. An entry on this would be very interesting.
The economy is claimed to be strong but that looks very weak to me. Hoping I have missed something.

  • 30.
  • At 10:03 AM on 07 Mar 2007,
  • Lossaversion wrote:

Irrationality - mmmmm as I live in a world populated by Homo Sapiens and not Homo Economicus we have to accpet that people are "normal" (a wide spectrum) and not rational in the economic sense. The integration of Psychology into Economics has changed the nature of the game and the fields of Behavioural Economics and Behavioural Finance are being followed - especially in the financial markets and in some cases policy e.g the default opt in choice in pensions is from Behavioural Economics not standard economic theory. People have limited capacity to assess outcomes (especially the more complicated the decision becomes e.g pension investing) in terms of risk and reward and focus on gains and losses and are prone to act differently i.e risk-averse when faced with gains but risk preferring when faced with losses (a development known as loss-aversion) - I have seen thuis in the presentations I have given on behavioural economics/finance (no plug intended)

People have difficulty in calculating the "optimum and choosing it" and our psychological drivers will always play a significant role in our decision-mnaking process - the key is to understand what these drivers are so that we can control them better and then perhaps move towards if we wnat to, the Vulcan like decision-making set out by standard ecoonmic theory - bear in mind Adam Smith was the first Behavioural Economist his Theory of Moral Sentiment published in 1759 preceded Wealth of Nations by 17 years. Standard theory is being enriched by other disciplines and Economics and this is where the the real value lies.

  • 31.
  • At 11:05 AM on 07 Mar 2007,
  • Geoff wrote:

For what it's worth, I know several people who holiday annually in Las Vegas. They have a spending budget for gambling, and they spend it all (usually)because they enjoy the thrill of an elusive potential big win. If you look at gambling as a form of entertainment (£15 pounds for a theatre ticket?), and you therefore feel that it's money well spent even if you lose, then there's nothing irrational about taking a chance on a slot machine or a roulette wheel.

Economists, exponents of the dismal science as they are, don't think of gamblers as having fun: they think of them as people with lopsided (not in any derogatory sense) utility functions for whom the 3% chance of winning a roulette jackpot is worth more than the 97% chance of losing their relatively small stake. That's not necessarily stupid: it's just something to keep under control. In fact, it's a lot more sensible than a flutter on the National Lottery.

  • 32.
  • At 11:36 AM on 07 Mar 2007,
  • Lossaversion wrote:

Apologies to all for previous and future grammatical mistakes etc in my submissions.

If Mr Davies wants evidence that irrationality from an Economics view has been pervasive throughout history - there are 2 excellent books

1) Security Analysis by Graham and Dodd; and

2) Graham's Intelligent Investor

The financial markets are one of the best places where it really is a case of the more things change the more things stay the same i.e. where the "separation of brains from capital" (Buffet?) is all too frequent an occurrence. Dragon Den's Farleigh is probably all too familiar with the reality gap the markets frequently create and the opportunities to exploit it. However to do so takes deep pockets and patience otherwise, as LTCM (a hedge fund that included not 1 but 2 Nobel laureates in Economics amongst its partners)discovered, it can really be a case of the markets being irrational longer than you can be solvent (Keynes?).

  • 33.
  • At 07:05 PM on 11 Mar 2007,
  • Lossaversion wrote:

For a couple of cases of "trrationality or loss aversion one has to only look at the ecent tech bubble and the current mania re man made climate change. Both have litle evidence to support them yet resources have and will be wasted oursuing them.

Man made climate change is the current irational exuberance made worse by the backing of politicians and of course UN. Based on the C4 programe last Thursday climate cannot be infkluenced by Homo Sapiens only the big yellow thing in the sky can - for the evidence see the Great Climate Change Swindle from Channel 4.

Water vapour is the dominant greenhouse gas and non-human CO2 production is 3 times that of human CO2 production - oh and by the way CO2 production by the eary lags temperature change by about 800 years - all of which is data based.

Time for a rational discussion with all the facts not one where the high horse riders dominate using research and evidence where the cenral assumption is CO2 causes climate change and ignores the real cause wich is the sun

  • 34.
  • At 04:42 PM on 13 Mar 2007,
  • Lossaversion wrote:

Nice to see the first Behavioural Economist being honoured on the back of the £20 note!!!!

  • 35.
  • At 10:57 PM on 15 Mar 2007,
  • Matthew Flinton wrote:

Ladies and Gents

Such erudition!

Keeping this simple.

Evan, bought the book, it is as brilliant as you say. My advice to other comment posters is buy it and read it as there are a lot more illuminating examples to get your teeth into.

Any other top tomes to recommend?

Many thanks.

  • 36.
  • At 11:05 AM on 19 Mar 2007,
  • Lossaversion wrote:

"The greatest trick the devil ever pulled was convincing the world he'she didn't exist."

The Trap is an excellent BBC 2 programme that highlights the paradox of public policymaking that relied on the irrational view that people behaved in a "rational self interested manner." Disasters using thsi premise included the Vietnam war, US and UK health policy and many others.

This assumption has created such a mess so that when things go wrong the policymakers tweak the formula rather than realise that the model is wrong (dissonance and loss aversion working together)

Its now time to move away from this assumption of rationality and move towards a more reality based one that people also act in a manner that involves equity and fairness - roll on Behavioural Economics - time Mr Davies too moved with the times and overcame his loss aversion and dissonance.

As for tomes read a small one called The Psychology of Investing by Prof Nofsinger - which highlights that people have difficulty in calculating and choosing the optimum even Nobel laureates in Economics

Of course Economists should be aware of this because they must be familiar with Adam Smiths Theory of Moral Sentiments (1759)

  • 37.
  • At 11:06 AM on 23 Mar 2007,
  • Economyst wrote:

I think we are confusing the english meaning of the term irrational and its economic context. Simply put we can define irrational behaviour as a deviation of what we expect in our economic models. The real question is as alluded to above is whether our models are wrong or that our understanding of rational behaviour is wrong. Rational behavious is based on what we expect a normal self interested person doing in a given situation. The problem is we never know what their self interest is.

  • 38.
  • At 11:21 AM on 23 Mar 2007,
  • Ming wrote:

Irrationality in one view could well be rationality in another. One can only see that if one is capable of taking both views

  • 39.
  • At 11:56 AM on 23 Mar 2007,
  • chris wrote:

An economics lecturer recommended this book to our class ten years ago. And what a great book it is.

I think a lot of people are too eager to pick holes in the theatre ticket example. Mainly because Evan was trying to describe it as breifly as possible. If you read the book you will see that he covers a number of assumptions such as disposable income.

If I remember correctly, one point he makes is that if someone loses a £15 theatre ticket and subsequently buys a second ticket, they may come to the conclusion that the ticket ended up costing £30 - and as a sidenote they will probably not enjoy the play as much, as pointed out above.

However, if the same person lost £15 quid instead, they wouldnt consider the ticket to now cost £30.

The net result is the same, but the perception is much different.

Anyway, I read the book ten years ago and I think that was the point he was making! Maybe I should read it again myself. I also remember some very interesting chapters on misinterpreting statistics.

  • 40.
  • At 11:41 AM on 30 Mar 2007,
  • Tom Cloke wrote:

Interesting debates going on here - but unsurprisingly some people are missing the point of what rationality is as defined by Sutherland & the many years of social & cognitive psychology research. Rationality is decision making using all the available information at the time, but is influenced by thinking errors that all people make. To clarify the theatre ticket example - the price of the ticket has not changed, rather the perception of the value of the ticket. The subjects in the study were observed to "distort" the information available and as such were not acting rationally. What this book does very well is highlight the errors we all make that skew our perception of situations and make our behaviour more or less rational - therefore economics is just as susceptible as any other area to the usual flawed human judgement. No matter how many statistics you use, the perception & filters viewing them will skew the rationality. Those responses that disagree - read the book first, study the views before resorting to self-justification. And yes, that one is in there too... :)

  • 41.
  • At 03:12 PM on 03 Apr 2007,
  • Tom P wrote:

This isn't exactly new. Behavioural Economics and Behavioural Finance are fairly well-established areas now. As someone points out above the auto-enrolment element of the proposed new Personal Accounts system is based on behavioural insights (inertia etc), and just the other day I managed to find a 'Behavioural Finance For Dummies' type leaflet on the retail investors bit of a fund management website.

Can we really say anymore that rationality is an axiom? maybe bounded rationality would be fairer thses days?

  • 42.
  • At 12:27 PM on 27 Apr 2007,
  • David wrote:

The arguments both ways are well made and I personally am swayed by reference to the individuals cost constraint. However, I think we are in danger of losing focus. If both agents to the transaction cannot agree then there will be no equilibruim.

  • 43.
  • At 11:23 PM on 10 May 2007,
  • Simon Roberts wrote:

# At 06:37 PM on 12 Feb 2007,
# Bedd Gelert [Post 6] wrote:

"By the way, Evan, what is that concept where a large group of 'amateurs' can 'guess' the value of something more accurately [...] than a far smaller group of 'experts' ? it may be just an urban legend."

There's a book called the 'wisdom of crowds' and it's all about that. it's good and it's not urban legend.

  • 44.
  • At 05:48 PM on 11 May 2007,
  • Alan wrote:

The decision rule about whether to buy another ticket having lost the original one:
u(see play,initial wealth-30) > u(miss play,initial wealth-15)

is not necessarily the same as the decision rule when buying your ticket initially:
u(see play,initial wealth-15) > u(miss play,initial wealth-0)

This is because u(wealth) may not be linear, i.e. du/d(wealth) may not be a constant, and the incremental effect on utility of the £15 may be different in the two scenarios. Would people agree with this? I finished my degree last July so I might be a bit rusty...

  • 45.
  • At 05:52 PM on 11 May 2007,
  • Alan wrote:

p.s. "By the way, Evan, what is that concept where a large group of 'amateurs' can 'guess' the value of something more accurately [...] than a far smaller group of 'experts' ? it may be just an urban legend."

Are you perhaps talking about predicitve markets - where betfair-style markets are used to derive implied probabilities of given events?

  • 46.
  • At 09:47 AM on 14 May 2007,
  • Andy wrote:

What's the marginal value of having a good sulk these days? £15?

"Critics must show that they have their own, alternative, discplined axioms, producing models that perform at least as well, before we should offer them a serious hearing."

This is fine, but this shouldn't be taken to mean that us "irrationalists" can't criticise "rationalists" just because we don't have fully resolved alternatives.

The criticism of a theory should stand or fall on whether the theory holds up against new evidence... not on whether the critic has an alternative theory to peddle.

This is well accepted in the physical sciences, where it was accepted for years that Newtonian physics didn't explain a whole bunch of things... people realised there were problems even before the likes of Einstein and Bohr came along with alternatives.

There are many examples of rationality working - but many others of "the rationality axiom" causing major problems.

Wasn't it John Maynard Keynes who said "markets can stay irrational longer than you or I can stay solvent?"

  • 48.
  • At 11:09 PM on 16 May 2007,
  • Alan Sykes wrote:

re comment 45 and the large number of amateurs. This relates to a kind of law of large numbers in that the larger the sample size the more reliable the result. So whilst 10000 amateurs may all behave irrationally in their estimate, some will over estimate, some will under estimate but the average will hit pretty near the mark. Of course, it could also be about how ludicrously overpaid 'experts' are.

  • 49.
  • At 04:22 PM on 17 May 2007,
  • Andy wrote:

44. Alan, I tend to agree with you, son. Nature abhors straight lines so why do we assume linear functions?. I think Evan had to simplify the example to fit it in.

47. John Maynard Keynes was so far up himself he didn't understand human behaviour. That's why he was so poor at negotiating.

  • 50.
  • At 03:23 PM on 06 Jun 2007,
  • Juan Carlos Negrete wrote:

Some “rational” people would not buy the ticket since their income has been reduced by 15 pounds. This could be a lot since, for instance, if you had a 45 pound budget to spend this evening (and you’re so disciplined that you stick to your budget), a 15 pound loss would be a 33.33% loss! Then, as economy textbook tell, if your income is reduced you spend less on some goods (the so-called normal goods. So, being rational would imply not buying another 15 pound ticket. The person would not buy the ticket but not because it costs 30 pounds since because their income is reduced.

  • 51.
  • At 01:15 AM on 09 Jun 2007,
  • Alan wrote:

Juan - that's what I've just formalised. And I've never heard of such a thing as an economy textbook. And there's no need to hyphenate the word rational. It's part of the discussion. No need.

  • 52.
  • At 01:41 PM on 20 Jun 2007,
  • Andy S wrote:

Thank goodness rational people are so unpredictable.

"The assumption of rationality in economics is like the map of the London Underground: it strips out the complexity of the system, to illuminate the essence of it. Of course, in practice we err, but it’s still useful to know what we would do if we were rational."

Doesn't this rationally imply that economics is the science of a fantasy world? Why is it useful to know how the world should behave if it was different to the way it mignt actually be?

What happens if the key to behviour of the system, literally, lies in it's complexity?

Perhaps logically, a belief in "rationality" is a rather irrational thing too. I understand that Theorems are derived by rules from axioms -but there is no way to derive an axiom -which are basically based on "self-evident beliefs". The history of non-Euclidian geometry shows that once these "self-evidences" are questioned they can generate viable systems -even though they (initially) appear to defy all (previous) logical inferences.

On a Globe (such as planet Earth), parallel lines meet in finite space -this is a mind bogglingly absurd for those indoctrinated by the flat surfaces of Euclid.

In your article on the WTO you say:
"These are all nightmare scenarios for economists, who learn the merits of free trade in their first weeks at university rather as school children learn the times tables."

Isn't a belief in the free market also rather irrational? Where on earth does such a thing exist? How could it possibly work fairly in a world of great divergence -with different levels of political and economic power?

How can there be a free market when one has a globalised economy based on large scale corporaions?

Doesn't the breakdown of the WTO talks, followed by the development of unilateral agreements -simply prove that that the little guys have to do what the big guys want -or the big guys won't play? So how can "democracy" and "free-trade" function under such a system? Inequality seems built into the system -and it would appear irrational not to accept this.

Ofcourse, once one admits one is beaten every which way -then one is perhaps pleased for every freecrumb that one can beg..... but it still don't make the system fair and equitable. So shouldn't economists look at the real world instead of constructing their dream scenarios?

Could somebody please answer a simple (fundamental) question: Can an "economy" increase real wealth -or only redistribute wealth.

If an economy (or an economist) really can increase wealth, then how can one resolve the apparent conflict with the laws of thermo-dynamics that say that energy and matter cannot be increased or decreased -only transmuted?

If economies can only redistribute wealth -then isn't this more a political question than a rational science?

  • 56.
  • At 12:30 PM on 05 Jul 2007,
  • Chris Frazer wrote:

In the theatre ticket example. You lose out either way. If you don't watch the play, then you've spent £15 on something you have not experienced and therefore not got your moneys worth.

Although if you buy the extra ticket for £15 then you will at least see and experience the play which you value at £15. Although obviously the other £15 ticket is wasted.

I reckon you'll find in most circumstances people will choose to buy the extra ticket.

  • 57.
  • At 10:02 PM on 22 Jul 2007,
  • Tony Coren wrote:

Evan - re Irrationality
Let's revisit the theatre example. You have in good faith paid for your ticket which left behind cannot be used by anyone else. The theatre choses in bad faith to disbelieve you and deny you entry unless you pay twice over. Therefore it is entirely rational to withhold further funds. You may even refuse future patronage at that particular theatre. As an economist you should appreciate this is market driven economics - for if enough clients took such a course of action (and clearly stated such an intention from the outset) then the theatre would accordingly be likely to adjust its policy in the clearly rational light of market forces

  • 58.
  • At 03:32 PM on 03 Aug 2007,
  • Mark wrote:

In response to comment number 55. Trevor, I am a little concerned that i haven't fully understood your point, so apologies if this is an insufficient explanation.

Essentially the answer is yes, an economy can - and does - increase real wealth. It can - and does - also redistribute wealth. A fundamental concept in economics is that of efficiency, of which there are a number of different definitions, such as productive, allocative and Pareto. It is through the efficient use of resources that more is created with less being wasted [that rhyme was unintentional - sorry]. In this way the question about the transmutation of energy and matter is kind of explained, i.e. you are reducing the amount of enery that is generated in the productive process in the forms of wasted heat, sound, inefficient labour rather than machinery etc.

The economy then redistributes wealth in order to place resources where they will be most efficiently utilised. The abstract, stylised model of a perfectly competitive market achieves this redistribution in a very simple and efficient way through price signals - though it is important to note that this is an abstract concept, for edification, not a perfect description of the real world.

The economist can possibly help this process, but really economic growth is an organic process generated by the millions of marginal decisions constantly made by actors within the economy - the micro foundations. A political element does indeed enter as you suggest, but I was always told to think of it in the following way: ideally the politicians would decide how they would like something to be, since they theoretically represent the wishes of society, the economists then use models of the economy and reccommend how to achieve this and politicians would enact their first or second best solutions. This seldom seems to happen as politicians and politics can be fickle games. The economists mantra 'Talk is cheap' is suggestive of the relationship between the two professions.

Try reading David Landes 'The Wealth and Poverty of Nations' for a very good historical overview of global economic development and interaction. Hope this helps.

  • 59.
  • At 12:20 PM on 09 Aug 2007,
  • Mark wrote:

" is an especially good read for economists who tend to assume that people behave according to some simple precepts of rationality."

Yes, as the recent BBC documentary 'The Trap - What Happened to Our Dream of Freedom?' pointed out, the only two types of people who always (or nearly always) behave in their own rational best interest are economists and psychopaths!

The idea that human beings consistently conform to their own rational self-interest would seem ludicrous to the average 5-year. How anyone including economists or worse politicians (including Blair!) ever came to believe the crackpot hypotheses put forward by people such as Francis Crick - brilliant scientist though he was - is incredible, and frankly dangerous.

Thankfully, these theories seem to have been thoroughly debunked - but has this news made its way to the upper-reaches of the Labour party?

  • 60.
  • At 04:07 PM on 09 Aug 2007,
  • John Turner wrote:

I used to work in manufacturing and often had to decide on whether to operate the plant and machinery to reduce the overall costs or risk stocking items I might not sell immediately. The philosophy then was much like your (Mr sutherlands theory ) of sunk costs. I likened it to 'Well I have the plant and machinery available so I might as well make use of the operators as I have to either pay them or have them idle'. The problem was that the extra product made by keeping the efficiency of the shop floor up, eventually cost the company in revalued product causing a profit they could not afford in taxation.
Its a bit akin to the theatre ticket in that ' Well I'm here and it has cost me to get here and if I dont pay the extra £15 for the ticket I will have
a) wasted my time which in real terms is a cost and
b) It has cost me to travel to the theatre and will cost me to get home. but there is the c element...
c) what do I do with myself if I dont watch the play? In business terms these are all sunk costs if you do not maximise your failing. in other words bad planning causes sunk costs. So you have to recognise them and recover as much of the sunk cost as you can.

The main idea is that we use rational tools to explain rational behavior. Of course there are cases which individuals don't behave rationally in which case our tools are not sufficient enough. Thus we make some sacrifices and one has to see whether the end result is worth the sacrifice.

Regarding the theater example, either action (read strategy) can be explained (in fact predicted) using economic analysis since either decision can be rational given the consumer's utility and budget constraint.

  • 62.
  • At 05:29 PM on 05 Oct 2007,
  • John wrote:

Re:Comment 55

Economies CAN create wealth by using more or better Factors of Production (Land, Labour, Capital & Enterprise) Population increases; the unearthing of new raw materials; building of a machine that is capable of producing valuable goods/services for a number of years using materials that on their own provide little benefit; all increase wealth without conflicting with the laws of physics

  • 63.
  • At 12:46 PM on 16 Oct 2007,
  • Robbie wrote:

Buy 2 extra tickets - then sell one of them to someone else for 30 pounds.

Yes, the group of amateurs guessing better than individual experts is well covered in the "Wisdom of Crowds".

What would the crowds advise with the theatre ticket problem though? Ignore the £15 sunk cost and spend £6 at the local movie hall?

  • 65.
  • At 05:56 PM on 24 Oct 2007,
  • Chris wrote:

Consider the following alternative situations presented in descending order of preference:

1. I spend $15 and go to the theatre.
2. I keep the money in my wallet and miss the show.
3. I burn $15 and miss the play.
4. I spend $30 and watch the play.

Clearly since situation 3 is preferred to situaton 4 and situations 1 and 2 cannot be achieved, it is perfectly rational
to chose 3 rather than 4.

Consider however a different ranking of the alternatives:
1. I spend $15 and go to the theatre.
2. I keep the money in my wallet and miss the show.
3. I spend $30 and watch the play.
4. I burn $15 and miss the play.
In this case it is irrational to miss the play because I lost my ticket.

A. Much depends on what alternative is feasible. Therefore the arguments of previous commentators which stress the importance of constraints seem valid to me .
B. Everyhting depends on whether you value 15 additional dollars more/less than the pleasure of watching the show.
C. The example does not have anything to do with choice under uncertainty; therefore all previous comments on that matter seem to me rather out of context.

  • 66.
  • At 03:16 PM on 26 Dec 2007,
  • Chandri wrote:

I've almost completed reading this book. This is by far the most interesting and thought provoking books I have ever read. The "sunk-cost error" example is just one of numerous examples outlined in the book. It talks in details about a lot of psychology experiments, simple probability teasers, logical arguments. It has changed the way I think and argue. Stuart is truly a genius! Thanks Evan for suggesting this book. Please recommend more of these "must-read" books.

  • 67.
  • At 03:52 PM on 27 Dec 2007,
  • Stewart wrote:

Interesting enough to make me purchase and read. is that rational

  • 68.
  • At 05:12 PM on 27 Dec 2007,
  • John Macmin wrote:

Buy low, sell high. Sounds good advice if you want to make a profit. Yet most people's instinct is the opposite. They buy high because the market's rising and they gamble on it continuing to rise, and they sell low because it's falling, and they gamble on it continuing to fall. If you're a day trader then the strategy of selling when the market's falling makes sense, if you understand the risks - and it is a risky strategy. But for most investors, who are relatively risk averse, and who are investing for the longer term, and wish to iron out the fluctuations of the market, and profit from the long term trend, which historically is growth, selling when the market's falling is complete folly.

  • 69.
  • At 06:15 PM on 27 Dec 2007,
  • Sam Pam wrote:

I like this article and as an economics student I've learnt to respect the rationality hypothesis. It is only rational to know how rational people think and make decisions.
You raised the issue of the sunk cost paradox. I feel it goes against your logic. If I understand it well, it's a situation where you continue spending on something that will more probably not yield any more returns. Britain and France did this in the defunct Concorde, which was running on heavy loses. I find that really irrational.
Instead of buying a second ticket for the same theatre performance, I'd rather put that same amount into something else that will of course be more gratifying. The sunk cost paradox only works for casinos and gamblers, who are generally famous for the irrational decisions they make.

  • 70.
  • At 07:26 AM on 14 Jan 2008,
  • David Hatchman wrote:


A very good article but I feel the sunk cost illustration misses the point and confuses perception of value with ability or willingness to pay.

Sunk cost is when the costs expended on a venture exceed the expected return from the venture - so the venture is abandoned i.e. sunk.

The theatre ticket example only works if the customer perceives the value of watching the play to be worth 30.00 to him, if it’s not then rationally the “Play Watching Project” is now over budget, the anticipated costs of completing the project now exceeds any benefits attributed to it so it is sunk. Project Managers are quite at home with this model.

The whole idea of “what’s that worth” and “rationality” is the fuel that drives Marketing as a discipline – the problem is that our language tends to only describe behaviors that are difficult to understand as “irrational” when in fact the observer is unable to see and describe the rationality. I suppose what we are talking about is aspiring to fulfill “needs” is somehow rational and fulfilling “wants” which include elements of fashion, status and prestige being irrational. An economist might start to introduce utility theory at this point and wrestle with the situation.

On the whole the sciences described, Marketing, Management, Psychology and Economics are young, still in their developmental phase so discussions threads like this can only benefit them. Let’s face it if human behavior, markets and socio economics was holistically understood we would have solved all the social economic problems and not be having this discourse.

  • 71.
  • At 01:00 PM on 15 Jan 2008,
  • Rob wrote:


I think your definition of a sunk cost is incorrect. A sunk cost is any cost which has been expended which is now irrevocable and has nothing to do with expected returns.

The decision on whether or not to watch the play depends on whether the individual decides that out of the budget he currently holds, spending £15 on the play is worthwhile when compared to what else he could buy with his budget. Whether the play is worth £30 or not should not come into his (rational) decision making process.

  • 72.
  • At 10:23 AM on 28 Jan 2008,
  • James wrote:

hey Evan, are you planning to add a new book to the reading list anytime soon? I enjoyed Irrationality and am looking forward to the next book you recommend.

Post a comment

Please note Name and E-mail are required.

Comments are moderated, and will not appear on this weblog until the author has approved them.

Required (not displayed)

The BBC is not responsible for the content of external internet sites