Tuesday, 30 November 2010

Sex and happiness

Despite Chris's warning, I went to Tim Worstall's book launch today. I even bought a copy which I will review on here when I have read it (on the bus tomorrow).

But before that, I learned something nice from his answer to an audience question.

Apparently, there's a clear relationship between GDP per head and population growth. We already know that people in poorer countries have more children, and the population of those countries grows faster. It seems that there's a measurable cut-off point: at $16,000 per head of GDP, fertility drops to around replacement rate (just over two children per mother) and the population stabilises.

Which reminded me of another statistic I read recently: national happiness grows with GDP until a certain point, at which it levels off and people stop getting happier. That level? $17,000 per head*.

Now perhaps there's something important in that $1,000 difference, but the two figures are within the margin of error for cross-country macroeconomic data series. Surely this can't be a coincidence: perhaps there is a common cause behind both effects?

If so, then surely the only conclusion is: People stop getting happier when they stop having unprotected sex.

Well, maybe not the only conclusion. Let's say you think the $1,000 difference is significant and in fact the fertility and happiness effects happen at different times. What is the typical growth rate of an economy at the $16,000/head level? Lithuania and Libya are the closest to this level (in PPP terms), and their growth (pre-recession) was about 8% in both cases. On this measure, how long does it take to grow per-capita GDP from $16,000 to $17,000? Nine months. So now our explanation is different. People start using contraceptives (or stop having sex) when their income reaches $16,000. Nine months later, children stop being born - and people stop getting happier. On this theory, people stop getting happier when they stop giving birth.

So it could be children that make people happy, or it could be bareback sex. We don't know, but surely it's one or the other?

Maybe not. Let's look at the statistics a third way.

Maybe people are having less unprotected sex and fewer children all the way up the income graph. So impoverished people in Burundi have lots of sex, many children - and are really miserable. The richer they get, the less sex and the fewer children they have, and the happier they become. Until we reach $16,000 - where the fertility rate levels off at 2.1 children per mother. And then people stop getting happier.

This means that both sex and children make people unhappy. The less they have, the happier they get - until they can't get away with reducing it any further, lest the population die out. At this point, they reluctantly force themselves to have sufficient sex and children to keep the grandparents happy (and provide enough workforce to pay their pension) and this is why the happiness effect levels off.

So David Cameron doesn't need the ONS to do a complicated survey to measure happiness. He can just count the new people born every year - which we already do - invert it, and there's our happiness measure.

It's no coincidence that the word "gay" also means "happy", you know.

* a more nuanced argument is here but let's not let that spoil our story

Unexpected revelations from wikileaks

Of course we are all just waiting for something really exciting to come out of wikileaks. Everything so far is either predictable (Prince Andrew criticising the Serious Fraud Office investigation of British Aerospace) or predictable (the Saudi royal family don't like Iran).

But in the meantime you can get some cute insights by reading between the lines. Here's my favourite of the day:
IRANIAN SOURCE NAMES "U.K.-REGISTERED" COMPANY [INSULTEC] AS IRAN SANCTIONS-BUSTER
4. (S) Note: A quick google check revealed several companies with the name INSULTEC in the title - these may or not be affiliated. Based on the information provided by source (currently in Iran, where he frequently travels), one possible candidate could be "INSULTEC Chitral Ltd."
I'm not sure if Google is the only authorised resource for US diplomats to find information, but "INSULTEC Chitral" returns nothing on Google apart from links to this wikileaks document.

The surprising discovery here is that the US State Department, with its world-class spies and diplomatic research resources, hasn't yet figured out how to search public online databases run by its closest allies. It takes about 30 seconds to discover (at Companies House) that there's only one UK registered company called Insultec (that last URL may be specific to my session, but you can search for the name at the Companies House link above).

I am happy to offer my research services to the State Department if they need someone to look up any more company names for them. Just drop me an email, Hillary, and I'll be your personal interface to the Internet.

p.s. there is a "Chitra Insultec Pvt Ltd" selling paint at weathertuff.com. However this is an Indian company, reselling paint invented by a UK citizen living in Australia; not, as the intelligence claims, a UK company run by directors of Indian origin. I hope we are not relying on this quality of research to save the world from nuclear war. But I fear that we may be.

Monday, 29 November 2010

Fair pricing and its effect on brand positioning

I'd guess that most of us are relatively happy to pay more to fly on or around a holiday than at other times. I'm sure those who flew home for Thanksgiving or are planning to do so at Christmas did not expect to get the same price as they would for a Wednesday afternoon in October.

Which makes this story (FT, may need subscription) slightly surprising:
India’s government has warned domestic airlines that it intends to crack down on “predatory pricing” after carriers sharply increased fares on popular routes during a recent festival, as overall passenger traffic surges.
Indian travellers were outraged during the recent festival of Diwali – Hinduism’s biggest gift-giving holiday – when some carriers charged about Rs25,000 ($550) for a last-minute Delhi-Bombay round trip ticket, a route on which fares usually range from Rs10,000-Rs15,000.
The price lift here is substantial (60-100%) but reasonably typical of the uplift on flights or train tickets in the UK market.

Is it a sign of a maturing market that suppliers start to charge prices based on supply and demand? Or is it that consumers are starting to accept price variations - with the politicians pandering to the electoral majority, who may not be representative of the market?

In any case, like all suppliers, Indian airlines must keep an ear open to what their customers regard as fair, but need not pay too much attention to the protests of those who are not actually buying the tickets. Perceptions of fairness can shift quickly in consumer markets. However, new brands, services and routes have an advantage over incumbents who have been using a particular model for years. People who have never tried a newly launched airline will have no strong price expectations. While those who have used the same one for years may feel ripped off by a price increase.

The likely outcome of this dynamic is that existing airlines are under more pressure to keep prices low, and will sell out all their tickets quickly. New airlines can charge more, will mop up the excess demand and, as a result of the higher prices, may be perceived as a higher quality service. This will give them a competitive advantage when demand returns to its usual lower level from February onwards.

Why are bund yields rising? Because the ECB is doing something right

Rising German bond yields over the week since the Irish bailout have been interpreted (here and here) as an increase in default risk to the German government.

But surely there is another "risk" much more likely to explain this increase. And that is the simple risk of inflation.

The Irish bailout must make inflation more likely - through one of two routes.


One: ECB liquidity support, i.e. lending to Irish and other banks, might not be repaid - which will result in an effective increase in money supply unless the ECB then tightens monetary policy to reduce it, which would be politically quite difficult. The ECB lending is secured on bank assets, but we know that those assets might not be worth 100% of their nominal value. So in the case of a bank default, the ECB has printed a billion euros to buy an asset which repays less than a billion euros of principal.

Two: monetary policy may be deliberately loosened, either to help reduce the pressure on sovereign borrowers or simply to increase inflation and growth.

In either case, the money supply increases, eurozone inflation is higher, and the real value of bunds falls. This can easily account for the small increase in yields which we have seen.

The euro has also fallen around three per cent in the last week, which provides further support for this theory. The idea that the German state is going to default is as unlikely as a default anywhere in the world, with the possible exception of Switzerland.

And that eurozone inflation? Much needed, as you can find out from reading the last two years of Scott Sumner's blog. Although he's recently pointed out that it's much better to express the policy goal as "higher wages" than "more inflation".

Another small note on this same subject: German business confidence is at its highest level in 20 years. Maybe this is an result of the implied monetary easing; more likely it's due to growth in China and sustained recovery in Germany itself. But either way, it's another reason for bund yields to go up and certainly doesn't imply that the German goverment can't pay its bills.

Sunday, 28 November 2010

The economics zeitgeist, 28 November 2010


This week's word cloud from the economics blogs. I generate a new one every Sunday, so please subscribe using RSS or the email box on the right and you'll get a message every week with the new cloud.

The words moving up and down the chart are listed here (this link may be temporarily down due to a server move - should be back within a couple of days).

I summarise around four hundred blogs through their RSS feeds. Thanks in particular to the Palgrave Econolog who have an excellent database of economics blogs; I have also added a number of blogs that are not on their list. Contact me if you'd like to make sure yours is included too.

I use Wordle to generate the image, the ROME RSS reader to download the RSS feeds, and Java software from Inon to process the data.

You can also see the Java version in the Wordle gallery.

If anyone would like a copy of the underlying data used to generate these clouds, or if you would like to see a version with consistent colour and typeface to make week-to-week comparison easier, please get in touch.

Saturday, 27 November 2010

The price of Amazon Prime

An article in BusinessWeek about the success of Amazon Prime contains this interesting revelation:
One challenge was selecting the annual fee for the service; there were no clear financial models because no one knew how many customers would join or how it would affect their purchasing habits. The team ultimately went with $79 mainly because it's a prime number.
We see this problem a lot with our clients - new product or service launches are one of the main occasions when people hire us for pricing advice. But I have never yet seen anyone argue for a price level because of whether it factorises!

Other interesting points: Amazon Prime in the UK costs £49 (not a prime number) and is one of the few items that is cheaper in the UK than the US. It also provides next-working-day delivery in most cases, rather than the two-day US service. But then, the UK is a smaller place and perhaps delivery services are cheaper in general here.

Finally, a funny quote from, of all people, a professor of psychology and marketing:
"I don't think it's a bargain at all," says Kit Yarrow, a professor of psychology and marketing at Golden Gate University who recently got a free Prime trial and cancelled it after a month. "Really what people are paying for is immediate gratification."
I don't agree. What I think people are paying for is the pleasure of not having to think about, or calculate, the varying utility of different shipping speeds. And, as Dan Ariely hints in a different context this week, the ability to get fast shipping without feeling guilty about it.

Friday, 26 November 2010

Security theatre versus terror TV

Bruce Schneier (via Farnam Street) makes the by-now-unoriginal observation* that:
...we pick a defense, and then the terrorists look at our defense and pick an attack designed to get around it. Our security measures only work if we happen to guess the plot correctly. If we get it wrong, we’ve wasted our money. This isn’t security; it’s security theater.
Probably true. But then, terrorism isn't exactly "real" either: by design, it's a theatrical exercise too. Or perhaps a reality TV show.

The clue is in the name - terrorism isn't designed to kill people, it's designed to make them scared. Thus, if we design this game for terrorists to play, they can win it just by smuggling a bomb through the security measures, regardless of whether it goes off.

Notice that all the recently-discovered terrorist plots - the shoe bomber, the underpants bomber, the soft-drinks bomber, the printer-ink bomber - have failed? If the goal is simply to make us worry, make us remember that Islamic fundamentalism exists, and to waste millions of hours of our time in airports, it succeeds without having to kill anyone. The terrorists and security authorities are in a cooperative game and I'm not sure if we really want to disturb the equilibrium too much. If we give either party a reason to start doing something different, we might not like the results.

What's more, there are plenty of beneficiaries from this process. The toothpaste, razor and bottled water markets have been boosted significantly in the last ten years. And airlines are now making millions in checked-baggage fees - or alternatively, have a powerful price discrimination tool - which never existed before.

So maybe security theatre - no matter that sophisticated observers recognise exactly what it is - is just what gives terrorists the level of visibility they need without having to sacrifice hundreds of lives for it. Here's to shared illusions...and happy Thanksgiving!

* To be fair to Bruce, he was one of the ones saying this kind of stuff before we were bored to death of it.

Biases about economic facts

Those who defend rational expectations have some explaining to do, in the face of data like this:
...six in ten Americans think most of the money spent to rescue banks will be lost forever. Six in ten think the economy shrunk over the past year. One in two think federal income taxes have gone up in the past two years.
Wrong. Wrong. And wrong.
In theory, it would still be possible for rich, sophisticated arbitrageurs to bet against the public in all these areas, and thus bring the overall path of markets back in line with the predictions of any given model. But most of the wealth of the economy is (directly or indirectly) under the control of the people who have these wrong impressions; so I think it unlikely that rational expectations can effectively operate. Especially since the errors are not random, but systematic; Derek Thompson in the article above has five explanations for why this might be.

David Laibson gave a fascinating talk on this topic at the Geary Institute on Tuesday; he introduced the concept of natural expectations, which combine two different ways of predicting the future: first, an intuitive extrapolation of current trends; second, a rational expectations component which matches the real data. Laibson's forthcoming paper (with Andreas Fuster and Brock Mendel) explaining this model is here.

Thursday, 25 November 2010

The dreaded Could of the economist

Robert Peston inadvertently highlights a common journalistic trap today - a trap laid for them by a frequent and annoying habit of economists.

David Walker, author of last year's Walker Review on bank governancehas written in the FT that:
"[A]ny attempt to require banded disclosure for UK banks in isolation would be commercially sensitive vis à vis their non-disclosing competitors elsewhere. It could also stimulate higher executive turnover, and (as a perverse unintended consequence) lead to higher remuneration as a defensive retention measure." [emphasis mine]
What exactly do "could" and "would" mean in this context? They have a very specific meaning in economist rhetoric.

Would means: is guaranteed to have the following effect on an unobservable variable. Note that Walker can confidently state that this is "commercially sensitive" because there's no way to measure, confirm or deny whether that is the case.

Could means: creates an incentive for an observable change, but nobody knows how strong an incentive. So, the unobservable variable of commercial sensitivity might encourage people to move jobs - but it could be such a small factor compared to everything else that there would be actual effect in the real world. As soon as the change becomes observable (higher executive turnover) then the language has to shift to "could" instead of "would".

Here is the problem: conventional economic theory makes it quite clear in which direction an incentive points, but has no way to determine how strong the incentive is. So we can agree that disclosure creates some incentive towards increased turnover, and therefore some incentive for higher pay - but it may be incredibly weak. On the other hand, disclosure also creates some incentive towards public shame, and therefore some incentive for lower pay! But equally, we don't know which one is stronger, or whether either of them is actually strong enough to make any difference to the decision of any real person.

So, economic theory cannot tell us what will happen in this case. Hence, economists use could to indicate that something might happen, but without giving any practical guidance to its likelihood. Only empirical data or a deeper theory can illuminate this, and rhetorical economics generally stops before getting to that.

In this case, as often happens, Walker chooses which coulds he wants to highlight, and journalists inadvertently translate could into would in their interpretation. In this case Peston himself hasn't fallen into the trap, but David Cameron has. Or more likely, Cameron has selectively used the coulds that suit him, so he can justify a politically convenient policy.

And David Walker? Well, he hasn't really changed his mind - his FT article repeats a point already made in his report, suggesting the UK should pressure other countries to implement their own rules at the same time as the UK does. But by using the economist's ubiquitous could - which is logically and intellectually respectable but rhetorically dangerous - he runs the risk that less subtle readers will take his potential, maybe weak, maybe counterbalanced, incentive effects and run away with them, over the horizon into convenient deregulation.

Update: I was going to say that the biggest could in economics is probably the effect of income tax on productivity and hours worked - higher taxes could have a disincentive effect, but nobody really knows. And they could indeed have a positive incentive on work through income effects...but nobody really knows. No sooner do I publish the post than I see Karl Smith correctly highlighting this as the main difference between Paul Krugman and Greg Mankiw - both eminently respectable economists who agree on textbook economic theory (if not on which textbook to read it in) but virulently disagree on most politically sensitive subjects. The deadly Could is at the core of this disagreement.

Sunday, 21 November 2010

The economics zeitgeist, 21 November 2010


This week's word cloud from the economics blogs. I generate a new one every Sunday, so please subscribe using RSS or the email box on the right and you'll get a message every week with the new cloud.


I summarise around four hundred blogs through their RSS feeds. Thanks in particular to the Palgrave Econolog who have an excellent database of economics blogs; I have also added a number of blogs that are not on their list. Contact me if you'd like to make sure yours is included too.

I use Wordle to generate the image, the ROME RSS reader to download the RSS feeds, and Java software from Inon to process the data.

You can also see the Java version in the Wordle gallery.

If anyone would like a copy of the underlying data used to generate these clouds, or if you would like to see a version with consistent colour and typeface to make week-to-week comparison easier, please get in touch.

Saturday, 20 November 2010

SJDM posts

DraftFCB's Institute of Decision Making is at the SJDM conference too, and tweeting on tag #SJDM if you want to keep up with some of the interesting research results in real time.

Today's highlights for me were:

  1. Martin Hilbert of USC has built an information processing model which can potentially explain seven different cognitive biases. In the model, external signals are stored in memory; and then retrieved again when required to make a decision. Assume that the channels into and out of memory are subject to random noise, and then place the following two constraints on the noise: first, that there is less noise than signal (i.e. our beliefs are more likely to be close to reality than not) and second, that the noise is symmetrically distributed (unbiased). From these assumptions, we can derive Bayesian likelihood, placement bias, subadditivity, hard-easy, overconfidence and conservatism effects. Martin also predicts a seventh bias which has not yet been observed, which he calls the exaggerated-expectation bias. We look forward to this being experimentally tested.

    My view: it's important to find underlying mechanisms which explain multiple biases. The noisy-channel model is interesting and plausible, but I am not convinced that this specific model is applicable in a wide range of decision scenarios. It seems quite specific to Bayesian-style probability scenarios. I think Martin has a more general version in mind, but I'm not sure what it is.

  2. David Tannenbaum looked at the ethical values of consumers. He measured experimentally the degree to which people hold "protected" values - that is, values which they are not willing to trade off against others. Typical unprotected values might be price, quantity or quality - the kind of things we trade off in our product purchases all the time. Protected values are typically based in ethics - for example, has the rainforest been cut down to make the desk I'm buying? His experiments suggest that consumers starts with a lexicographical decision-making process - where consumers seek a do-no-harm choice which will make no impact at all on their protected value - and if no such choice is available, they revert to a traditional one, where they are willing to make trade-offs if the price or quality difference is sufficient.

    My view: I am not sure if it is correct to model protected values as all-or-nothing, rather than just as very strong preferences. No matter how deeply I care about the rainforest, I might not be willing (or able!) to spend $1 million to buy a desk which does not damage it. However, to a first approximation this is probably a useful model. There are useful applications for brands here: one implication could be that a brand which offers a do-no-harm option (the no-rainforest choice) can take advantage of a high willingness-to-pay among the consumers who care about that value. An analysis of bundling shows that firms are likely to sell products which have high quality in multiple dimensions at once - so you can buy organic chocolate which also tastes good and has beautiful packaging, but nobody sells organic chocolate which tastes worse than a regular Cadbury's bar. I'm not sure if this bundling effect operates differently for protected values, but David's comment after the talk suggested that it might be.

  3. Dan Cavagnaro showed a useful methodology for running experiments to distinguish between different theoretical models, called Adaptive Design Optimisation. In this mode, instead of having a fixed set of experimental tests designed in advance, the software automatically selects in realtime a test which will maximise the resolution of the distinction between the options being tested, given previous responses.

    My view: an extremely powerful approach, indeed one which I'm incorporating (in a different form) into our software offering. A simple example: if I'm testing willingness-to-pay for a product, I might start by testing consumers at a £50 price point, planning to test £30 and £70 afterwards. However, if I find that all of my initial test subjects are willing to pay £50, there is little point testing the £30 level. Instead, the test software should adjust to test an £80 price point and then perhaps £120. I'm not sure whether Dan's software will be precisely applicable to what I do, as it is about distinguishing between models rather than distinguishing between parameter values. But it's good to know someone else is applying a similar approach - and, it appears from his results, with striking success.

  4. Sig Mejdal lightened the tone with a fun discussion of how he's persuaded the St Louis Cardinals to hire him as a quantitative analyst, to help them select the best new players from each year's draft. It took him three years to convince them, despite Michael Lewis's bestselling book Moneyball which showed how the Oakland A's did the same thing eight years ago with amazing success.

  5. Jennifer Trueblood described a "quantum" model of judgment, applied to how jurors decide the guilt of an accused person after hearing prosecution and defence arguments. I usually have a real problem with people making this kind of dodgy analogy from physics (especially quantum theory) to human behaviour. However, I read this as being merely an adaptation of quantum theory's mathematical techniques, rather than an attempt to claim Heisenberg's Uncertainty Principle actually applies to juries (which is the kind of thing you'll hear occasionally from social scientists).

    My view: the mathematics was too dense to convey in a 20-minute talk - even for me, with training in quantum physics - but it could be a useful model. I do have doubts on two points: one, I don't see how the model would plausibly arise from our mental capabilities; and two, even if it does, it will be very hard for most researchers to apply this model in practice, due to the complexity of the mathematical concepts used. Still, I am hoping to find out more before the weekend finishes.

  6. Laurence Maloney proposed that essentially all probability biases can be derived by assuming that people remember the log of probability rather than the probability itself, and that a constant error term is applied to the log.

    My view: this could be a really fundamental result, if further research bears it out. There is a plausible cognitive mechanism for it (sampling behaviour which is proportional to the log of the population size) and it can be easily applied. Worth watching.

  7. Anuj Shah gave a very entertaining talk about some work he's done with Eldar Shafir and Sendhil Mullainathan on why people take out payday loans and otherwise behave against their long-term interests. He's a great speaker and someone to watch in the popular behavioural press in the future. The hypothesis is that if people are resource-constrained (poor) they have fewer resources available to help them make the right decision, meaning they are likely to overborrow and will end up even poorer in the long term.

    My view: Good, solid results which I think shed light on a potential different explanation. It could be that when we have money available, we find it much easier to envisage the alternatives we could be enjoying, and regret not having them. If I don't have any money - or ability to borrow it - I won't much miss the restaurant meal or holiday that I can't have. But if I could put that vacation on my credit card, and am simply choosing not to, the potential enjoyment is much more salient and I'm quite likely to cave.
In other good talks, Dan Egan showed how clients of Barclays Wealth think that all other clients of Barclays Wealth are more pessimistic than they really are; Katherine Thompson surveyed some clever methods of eliciting discount rates; and Pavel Atanasov measured how ruthless people are when playing The Price Is Right (answer: not as much as they should be, but more ruthless against those of the opposite sex).

And it's nice to meet some people in real life whom I've previously encountered only in journal papers and the occasional TED talk.

Friday, 19 November 2010

Sarah Lichtenstein tribute

In St Louis for the SJDM conference and the first event is a very interesting 2-hour precis of the career, contribution and personality of Sarah Lichtenstein.

I hadn't realised she (and her long-term collaborator Paul Slovic) had such a critical influence on the development of the Judgment and Decision-Making field. I've come across her work before, of course - if you have read much in this field you are unlikely to have avoided her. But I often read and remember a result without taking note of who was responsible for it, and I've noted many of the key experiments and theories that she and Slovic carried out, without realising they all came from the same team.

I'm not sure whether the tribute was due to some particular occasion or anniversary, but you'll be pleased to hear she is still alive and well, as is Slovic and the several other pivotal researchers who spoke about her work. Slovic, John Payne, Chris Hsee, Eric Johnson, Elke Weber, Robin Gregory, Ellen Peters and Maya Bar-Hillel all spoke, with closing comments from current SJDM president Valerie Reyna.

Many of them took the opportunity to talk about more recent work and show how Sarah's original research inspired it. The key insight of Lichtenstein-Slovic's work was to show preference reversals. In the classic case, subjects are given two options, A and B, and choose B. But when asked to put a monetary value on each one, they put a higher value on A.

From this, they developed a few basic principles - summarised neatly by John Payne as "the three C's": contingent decision making (decisions depend on context and circumstances), constructed preferences (and beliefs: the idea that preferences are not defined prior to choosing, but arise in the process of considering a choice), and choice architecture (the idea that the framing and presentation of a question influences the answer people give).

Chris Hsee showed some cute examples of preference reversal and outlined a theory of why preferences reverse. Namely that incremental differences are easier to compare when you see two options at once, and so they dominate, while category differences provide more insight when you only see a single option and are asked to put a monetary value on it. For example, if offered a choice between $120 next month versus $100 today, I am likely to see the monetary difference as dominant and pick $120. But if asked to simply put a value on "$120 next month" I may focus on the "next month" part and give the whole package a low valuation, maybe $80. When offered "$100 today" a subject is likely to see "today" as a valuable attribute and may well value the whole offer at $100.

Eric Johnson drew a few examples from William Poundstone's book Priceless, which I read earlier in the year and think is an excellent overview of psychological pricing research. I remembered the examples, but once again had not noticed that they were referring to Sarah Lichtenstein. And Ellen Peters showed the results of some strikingly counterintuitive experiments. Subjects were offered a choice between two (nearly identical) erasers and asked to pick the one they preferred. When a large number 1 was placed between them, 67% of people chose the eraser on the left. When a large number 9 was displayed, 65% chose the one on the right. A remarkable result.

The work of this discipline, following on largely from Lichtenstein and Slovic's experiments more than 40 years ago, is crucial in providing the microfoundations for cognitive economics. Congratulations to Sarah - and Paul - for so many years research and for their inspiration of generations of JDM researchers.

p.s. any readers in or near St Louis who are interested in meeting this weekend, or coming to any of the SJDM sessions, let me know and I'll see what I can do.

Tuesday, 16 November 2010

Was Britain really better in the 1940s?

Andrew Sullivan approvingly quotes Barry Eichengreen on Britain:
[Britain] failed to develop a coherent policy response to the financial crisis of the 1930’s. Its political parties, rather than working together to address pressing economic problems, remained at each other’s throats. The country turned inward. Its politics grew fractious, its policies erratic, and its finances increasingly unstable.
In short, Britain’s was a political, not an economic, failure. And that history, unfortunately, is all too pertinent to America’s fate.
Now there's no doubt that Britain's political parties (once they got over six years of intensive cooperation during the war) did engage in a debate about how to run the country. Partly about its role in the world, but more importantly the nature and extent of the welfare state, the kind of industrial policy to follow, and how the education system and the country's infrastructure should be built. Its most important decision about its "role in the world" was that it shouldn't be running other people's countries any more - a choice which I believe most people now agree with.

In fact, the fractious political debate was very healthy for the UK. Even if it led politicians to focus on other issues and prevented them from arresting the UK's economic decline - and it's unclear that they could have done so, anyway - it was absolutely necessary for us to debate the correct policies for Britain in a changed world. I wouldn't want to go back to the 1940s, in political, economic or social terms.

Two more things about this:

First, David Cameron has picked up the theme and is talking about it too. Prompted by that, Nick Robinson is talking about this issue over the course of the week.

Second, note Eichengreen's clever rhetorical device. By asking the question about America and not Britain (Is America Catching The "British Disease?") Eichengreen tricks us into not questioning the premise as we should. If we are busy thinking about whether America today matches the supposed pattern of 1940s Britain, we don't stop to ask ourselves whether 1940s Britain really matches it. Take notes, polemicists - if you are trying to persuade people of something, just ask a slightly different question and let your real argument be implicitly given:

  • Instead of "Are markets irrational", ask "Are irrational markets really the main factor behind the financial crisis?"
  • Instead of "Is Barack Obama a Muslim?", "Are Obama's Muslim roots really relevant to US defence policy?"
  • And how about "If Jesus was black, what about Santa Claus?"

Monday, 15 November 2010

New York, Wednesday-Friday this week

I'm attending the SJDM conference in St Louis this weekend; and will be in New York from Wednesday afternoon until Friday afternoon. I would quite enjoy meeting and chatting with any US-based readers, either in NYC or at the conference. If you fancy meeting up, post a comment here or email me.

Sunday, 14 November 2010

The economics zeitgeist, 14 November 2010

This week's word cloud from the economics blogs. I generate a new one every Sunday, so please subscribe using RSS or the email box on the right and you'll get a message every week with the new cloud.


I summarise around four hundred blogs through their RSS feeds. Thanks in particular to the Palgrave Econolog who have an excellent database of economics blogs; I have also added a number of blogs that are not on their list. Contact me if you'd like to make sure yours is included too.

I use Wordle to generate the image, the ROME RSS reader to download the RSS feeds, and Java software from Inon to process the data.

You can also see the Java version in the Wordle gallery.

If anyone would like a copy of the underlying data used to generate these clouds, or if you would like to see a version with consistent colour and typeface to make week-to-week comparison easier, please get in touch.

Saturday, 13 November 2010

What have the Normans ever done to us?

I thought I'd seen it all.

But no. Commenter "zadok" on Robert Peston's blog has discovered the real culprits behind Britain's economic malaise.

From comment 5 on "Can the UK close massive deficit with China?":
The english ( anglo saxons ) are the most inventive people in the world but unfortunately,like everything,the normans highjacked their flair and trashed it.
(Punctuation and spelling in the original)

I have seen the Americans blamed; the Labour government; the Tory government; the Chinese; the rich; the middle class; the unemployed; the banks; the EU, the IMF, the World Bank, the capitalists, the socialists, the communists, and even the behavioural economists.

But it never occurred to me that it could have been...the Normans.

Thursday, 11 November 2010

The dangers of selective reporting

The Wall Street Journal's RTE blog (via Paul Krugman) has been spluttering about being misrepresented by Sarah Palin.

RTE criticised her for saying "everyone who ever goes out shopping for groceries knows that prices have risen significantly over the past year or so". In fact, RTE pointed out, inflation in food prices has been at a record low this year, of just 0.6%. General inflation is also low, at 1.1%.

Palin - and I have to give her credit for this - responded quite cleverly to RTE by citing the Wall Street Journal itself, which said in an article last week:
"an inflationary tide is beginning to ripple through America’s supermarkets and restaurants…Prices of staples including milk, beef, coffee, cocoa and sugar have risen sharply in recent months"
RTE correctly points out the facts of the matter: beef is up, but bananas are cheaper; producer prices are up but retailers have not yet passed them on; the catering industry is increasing quality and offering premium options to customers rather than forcing straight price rises on them.

But here's the thing: nobody would have guessed any of that from reading the WSJ's own article. The authors of the article, Jargon and Brat (what wonderful names), have selectively quoted statistics and anecdotal evidence to convey an impression of fast-rising prices. It would hardly be surprising if ordinary readers, let alone a Sarah Palin with a political point to make, rely on the WSJ's clear implied message rather than poking through the underlying data.

Perhaps the Journal will learn from this that their credibility will be improved if they present facts in a neutral, responsible way instead of following the lazy editorial habit of pick-the-evidence-to-support-today's-narrative. You can always select a convenient subset of data points and draw any curve you want through them. But someone who is genuinely attempting to get an accurate picture of the world will be as comprehensive as possible, and will work hard to combat the cognitive biases that influence the patterns we see. We can hope that the Journal actually is more committed to accuracy than to storytelling. But to be fair, it's very hard both to achieve objectivity, and keep the attention of readers.

It's a little like Interfluidity's message this week about economic morality. It's all very well between economists for Paul Krugman to rely on counterintuitive, technocratic, model-driven analysis. But if the message is not translated into something that real people can feel in their gut, it won't resonate. And whoever does have a message that affects people in that way - regardless of whether it's logical, self-consistent or founded in facts - is likely to get their way.

Tuesday, 9 November 2010

The economics of getting off a train

A surprising article on the BBC today, explaining How to get off a busy train. I guess the BBC does have an educational mission.

But however obvious getting off a train might be, reading the article prompted a few ideas. The article is mainly about how passengers should behave, but acknowledges the role of the train's design in influencing that behaviour. And some of those design choices are very reminiscent of the "choice architecture" discipline we know from books like Nudge.

But this is a problem we wouldn't normally associate with economics at all. It's a product design - or even an architecture - question. So is the domain of Nudge really economics? Or is it in fact design, ergonomics or something else?

The stuff of economics is normally about how we allocate our wealth and material resources; about how we respond to incentives; how we trade and deploy limited amounts of capital and labour to produce maximum utility.

Design problems, on the other hand, are about how we allocate our thinking time; how we respond to the tools and interfaces offered to us; and how we choose, arrange and structure limited physical resources to solve a problem.

Hmm...maybe not so different after all. Cognitive effort is a resource just the same as labour, iron or land. Designing a choice architecture to minimise thinking and bring about a desired result is not dissimilar from setting the right prices to reduce carbon usage. And trading goods to increase utility has a lot in common with how our attention is pulled between competing attractions.

So we can apply economic thinking to the problem of getting off a train (for that matter, maybe we can also apply design thinking to increasing small business lending - but not today).

The BBC's train-exiting experts recommend three tactics:

  • Keep moving steadily: speed is the desired outcome here, but comes at the cost of additional thinking to keep physically coordinated. As with any economic supply curve, there is an increasing marginal cost: each bit more speed brings a higher and higher extra cost of coordination.
  • Get off before getting on: of course this is obvious...to you, I hope, my reader. But not to every user of London Underground, it seems. How could we solve this problem with economics? Assuming the transaction costs of paying people to get off the train are too high, perhaps we can create a cognitive incentive. There are two at play here. Passengers have been trained to apply a social pressure to each other...anyone trying to get on before others have got off will pay the cost of many stares and tuts from other travellers. And the thinking cost of getting onto the train is higher if you don't wait for it to empty first. Why then do some people jump on first? Because of their fear of the failure of the third tactic...
  • Spread out evenly: A version of a well-known economic trap applies to trains too. Hotelling's Law shows that two suppliers competing in a market (archetypically, ice cream stalls on a beach) will not move to the optimal position for customers, but to the best position for competitive defence against each other - right next to each other, with no differentiation. And when passengers are standing in a train carriage, the optimal social solution is for them to space themselves evenly [at least, if we assume declining marginal utility of space]. But the selfish choice for at least two of them is not to move right down the carriage, but to leave extra breathing space while blocking other passengers nearer to the doors. This benefits those two a little at, a high cost for the rest. We therefore have a strong incentive to become one of those two people, which in turn results in the problem of people rushing to get onto the train first - if necessary, before disembarking passengers have had a chance to get off.

    The solution here? Apart from social pressure again, we can use the useful economic tool of arbitrage. If two people are standing apart in the middle of a carriage, they are creating a space which someone crushed into the door would be happy to use. So we should allow - indeed, encourage - those at the door to move past everyone else and take the space. A notice on the door instructing people to do so would be sufficient. What's more, in most markets the threat of arbitrage is sufficient to stop most arbitrage from actually happening. And this case is no different: once passengers come to expect the annoyance of people pushing past and standing on their toes, they will pre-empt it by moving down the carriage and using up the space themselves. Thus neatly resulting in a socially optimal solution at minimal cognitive cost.
An excessively complex analysis of a simple problem? Certainly (though in my defence, the BBC started it). Then again, that's what economists do. Enough practice at modelling the simple problems means the answers to complex ones start to come naturally.

The broader lesson: cognition is an economic resource, and information changes cognitive incentives. Adding these two insights into the standard theory creates a powerful extension to economic thought - potentially a whole new discipline - cognitive economics.

Monday, 8 November 2010

How are beliefs about growth formed?

Two articles this evening lead me to ask the question: how do we predict GDP growth?

Before reading on, why not ask yourself this question: what do you expect next quarter's GDP growth figures to be? How about the next 12 months? And why? I'd be interested to see some of your answers in the comments to this post - please also say which particular GDP figures you're predicting (personally I'm most familiar with the UK and US figures, but would be interested in comments from the eurozone and other regions too)*.

I'm not going to test you on the accuracy of your forecasts: I'm more interested in the prediction itself, and your reasoning, than whether it turns out to be right.

My prompts for thinking about this are:
  1. a paper Roger Farmer sent me this evening, introducing the concept of a belief function describing what people expect next year's growth and inflation figures to be
  2. David Smith's blog excerpt of his Sunday Times column today, in which he discusses this quarter's UK GDP figures, and the differing unemployment picture between the UK and the US
  3. (Honourable mention also to Scott Sumner's post on British austerity, which is more about the balance between monetary and fiscal policy but is also implicitly about the effect of growth expectations on current growth.)
So here's my question. Are your expectations of next quarter's GDP growth mainly influenced by:
  1. This quarter's GDP growth?
  2. Your personal experience of the economic environment? (Chris Olivola mentioned some research to me the other day: if you ask people whether they believe in global warming, they are more likely to say yes on a hot day than on a cold day. He said this is like asking someone how the economy is doing, and they answer based on how much change they have in their pocket).
  3. What you read in the media, or hear from your friends, about GDP growth?
If you consider that a manager deciding whether to employ more people - or sack their current staff - is going to be partly influenced by their expectations of economic growth, you can see the importance of this question.

My hypothesis: point 3 has the biggest impact. I don't think most people in business know the current GDP figures. And I think our direct experience of economic events is strongly filtered through the lens of our beliefs; which are in turn influenced by the information we receive from outside. If the media, and our friends, think there is still a recession, then we will expect growth to be slow. And this, to some extent, will mean that growth really is slow.

Of course expectations are only one factor that influences GDP - inventories, replacement cycles of durable goods, fiscal policy, wealth and incomes, exports, and exogenous events such as weather or the Olympics are all important too. The balance of these different factors is one of the main research subjects of empirical macroeconomics. But the effect of expectations, especially on investment but also on employment, is strong.

Thus, we want to understand how people form their opinions about future growth. How much of their beliefs do they gain from media messages? How much from social contact? (And how much from direct evidence gathering?) How were those media and social messages formulated? How did those journalists or friends pick up their beliefs in turn? What is the lag time? How fast do beliefs spread, and what are the conditions that influence that velocity?

These are critical research areas for cognitive economics. I don't know of any programme of research across these areas, so maybe I need to start one.

* I'll post my own predictions tomorrow

Sunday, 7 November 2010

The economics zeitgeist, 7 November 2010


This week's word cloud from the economics blogs. I generate a new one every Sunday, so please subscribe using RSS or the email box on the right and you'll get a message every week with the new cloud.


I summarise around four hundred blogs through their RSS feeds. Thanks in particular to the Palgrave Econolog who have an excellent database of economics blogs; I have also added a number of blogs that are not on their list. Contact me if you'd like to make sure yours is included too.

I use Wordle to generate the image, the ROME RSS reader to download the RSS feeds, and Java software from Inon to process the data.

You can also see the Java version in the Wordle gallery.

If anyone would like a copy of the underlying data used to generate these clouds, or if you would like to see a version with consistent colour and typeface to make week-to-week comparison easier, please get in touch.

Friday, 5 November 2010

UK economics blog rankings

With the imminent release of an 'Economy' category on the Wikio blog ranking site, it may be time for me to update my list of UK economics blogs. I hadn't even heard of a few of Wikio's top 20 (though as Richard Murphy points out at that link, some of them aren't really about economics).

I'm pleased to be on the list and especially flattered by Paul Mason's comment here. Paul is perhaps the black sheep of the BBC's three main economics bloggers and all the more interesting for it, so I'm very pleased to hear he is a reader.

The most surprising omission for me is Stumbling and Mumbling - Chris is one of the best economics analysts in the UK blog scene. And I'd have expected to see Tim Worstall on there, for his traffic levels and profile at least. They both seem to be listed under Politics instead.

Congratulations to everyone on the list and I look forward to some friendly rivalry with you all in moving up the rankings.

Update: By my count, I've moved one place up the rankings already. Amusingly, the Economist's Free Exchange blog is in the list twice, once at number 3 and once at number 12. It's an excellent blog, but I think two listings is a bit much. So I'll omit the second listing and declare myself moved up one place to number 16.

Thursday, 4 November 2010

Moral posturing versus QE2

I've had a go at Allister Heath in this blog before. Looking back, I realise that it was on the same subject as today's post: quantitative easing.

As I read his column in City AM today, I wondered: when I disagree with someone this much, and yet he is so sure of himself, should I question my beliefs? So I questioned them, and read the column again, and realised: no, 90% of macroeconomists are right, and Allister Heath is wrong.

Heath argues with conviction that the Federal Reserve's "hubris" will cause inflation without helping the economic recovery. It's a version of the Austrian story: the economy must undergo a necessary recalculation, restructuring, reallocation of resources, before a recovery can happen. Stimulus will just paper over the cracks and maintain the distortions in the economy.

There are reasonable arguments that this may be true of fiscal stimulus (though on balance, I don't agree with them). But monetary stimulus (including QE) can hardly be accused of this. Monetary easing, in fact, is one of the main routes out of a distorted economy.

In an economy where prices and wages adjust instantly to changes in demand and supply, recessions would not happen - all economic corrections would take place right away, and no intervention would be needed from any kind of authorities (except to deal with externalities, public goods or other identifiable market failures). That is not the world we live in. Prices and wages are anchored to fixed points - by psychology, by menu costs, and by the imperfect transmission of supply and demand signals through imperfect markets full of asymmetric information and not-very-good negotiators.

Inflation is the easiest way we know to loosen these anchors, allow prices to adjust, and make the economy more flexible. This is because inflation gives people a reason to change their prices - and while they're changing them anyway, they may as well try to adjust them to the correct level. It provides a way for companies to reduce effective prices or wages without protest from suppliers or employees, so that they can adjust to new market realities. There are far fewer forces to stop wages from increasing than decreasing, so people in high demand will get higher salaries regardless. And people in low demand - construction workers, according to Heath - will automatically get lower wages in a time of inflation, and therefore will be more likely to keep their jobs.

Much of the theoretical work of macroeconomics over the last few decades has been about determining the strength of this effect: the Phillips curve, the concept of NAIRU and New Keynesianism are all based on the idea that inflation has an effect on growth and vice versa. Monetary policy can be seen as the optimal, market-based response to fix this market failure. The interest rate (combined with other monetary tools), exogenously set by the central bank, is the empirically determined price which society puts on the externality represented by inflation.

Monetary stimulus has another effect, which is to increase the price of assets relative to consumption (see last week's post about Nick Rowe) - which boosts investment. Investment in a recession is generally much lower than its optimal level - and if, as Heath claims, we "need to deleverage", it will be lower for many years yet. If monetary policy can combat this effect, it's a good thing.

An odd claim towards the end of the article:
"Alan Greenspan...cut rates to one percent to minimise what many wrongly saw as a risk of deflation. In the event...inflation rose from a low of 1.1%...to 2.4%"
The fact that a Fed action, intended to prevent deflation, was followed by no deflation is hardly an argument against it. We can have a debate about Greenspan's overall effectiveness, but inflation of just over 2% is a success, not a failure, of Fed policy.

Finally, Heath warns against "the huge dangers of engaging in further QE" but omits to mention what any of those dangers are - with one exception. He suggests there is "a bubble in emerging markets" - but if, as he also says, "elevated unemployment and weak growth are...unavoidable" then where else would Western investors put their money than in emerging markets? QE may have a stimulating effect on investment in those markets, but there is an almost unlimited capacity for it, and I don't think China, Brazil or India will be too worried about someone coming to build $50 billion more of factories.

To put this in perspectively, though, of the previous $1.7 trillion of QE, probably no more than $50-100 billion went outside of the US. The majority went either into the US Treasury or into Federal Reserve bank deposits, and is only slowly feeding out into private investment. Sensible macroeconomists wish it would go a bit faster, and the next phase of QE must help that process along. If Heath would prefer the economy - and millions of unemployed people - to suffer a few more years until his corrections are properly implemented, he is fortunately in a minority.