Tag Archives: economics

Congestion Tolling at the Supermarket

To help explain why toll lanes might not be the great solu­tion to traffic con­ges­tion many believe them to be, Timothy Lee goes to an unex­pec­ted place to draw par­al­lels: your loc­al super­mar­ket.

Super­mar­kets are a good ana­logy, sug­gests Lee, because they oper­ate in a free mar­ket, are ruth­lessly effi­cient, intensely com­pet­it­ive, and employ ‘lanes’ (check­out queues)… but they don’t use con­ges­tion pri­cing. The reas­ons why they don’t, he says, can also be applied to traffic con­ges­tion:

First, we have strong and soph­ist­ic­ated social norms, cul­tiv­ated since we were young chil­dren, for wait­ing in lines. This bit of self-organ­iz­a­tion is extremely import­ant for the smooth func­tion­ing of civil soci­ety. We see wait­ing your turn as an oblig­a­tion we have to one anoth­er, and there­fore not as an oblig­a­tion that a super­mar­ket or trans­port­a­tion agency can waive in exchange for a cash pay­ment. I sus­pect cus­tom­ers would see people using a tolled check­out lane as break­ing an impli­cit social con­tract.

More import­antly, cus­tom­ers would be sus­pi­cious that the super­mar­ket was delib­er­ately under-staff­ing the free lanes to gin up demand for the express ones. […] In the low-mar­gin gro­cery busi­ness, it would be a pretty effect­ive way for a man­ager to pump up his short-term profits, while the long-term harm to the store’s repu­ta­tion would be hard […] to quanti­fy.

This lat­ter con­cern seems par­tic­u­larly rel­ev­ant to the case of toll roads. The rev­en­ue-max­im­iz­ing pri­cing sched­ule is not the same as the con­ges­tion-min­im­iz­ing sched­ule. An effect­ive con­ges­tion-pri­cing scheme might gen­er­ate rel­at­ively little rev­en­ue if people shift their driv­ing to off-peak times (which is the whole point). The oper­at­or of a mono­pol­ist­ic toll road will face a con­stant tempta­tion to boost rev­en­ues by lim­it­ing through­put on free lanes and jack­ing up the off-peak toll rates. The wide­spread voter per­cep­tion that they’ve “already paid for” many tolled roads through oth­er taxes isn’t exactly right as a mat­ter of fisc­al policy, but I think it’s based on a sound intu­ition: there’s no reas­on to think the polit­ic­al pro­cess will set tolls in a way that’s either fair or eco­nom­ic­ally effi­cient.

The Drinkers’ Bonus: Alcohol Intake and Increased Earnings

Drink­ing alco­hol – and the increased social cap­it­al that it leads to – may not just be respons­ible for a pos­sible increase in life span; it may increase your earn­ings, too.

In an ana­lys­is of both the Gen­er­al Social Sur­vey and the pub­lished lit­er­at­ure, research­ers for the Reas­on Found­a­tion show that alco­hol drink­ers earn, on aver­age, 10% more than abstain­ers (pdf). This is known as the drink­ers’ bonus.

Recent stud­ies indic­ate that drink­ing and indi­vidu­al earn­ings are pos­it­ively cor­rel­ated. Instead of earn­ing less money than non­drink­ers, drink­ers earn more. One explan­a­tion is that drink­ing improves phys­ic­al health, which in turn affects earn­ings (Hamilton and Hamilton, 1997). We con­tend that there is an eco­nom­ic explan­a­tion. […]

Drink­ers typ­ic­ally tend to be more social than abstain­ers. As Cook (1991) explained, drink­ing is a social activ­ity, and one reas­on people drink is to be soci­able. In the med­ic­al lit­er­at­ure, Skog (1980) showed that mod­er­ate drink­ers have the strongest social networks. Fur­ther­more, Lei­f­man et al. (1995) doc­u­mented a neg­at­ive rela­tion­ship between social integ­ra­tion and abstin­ence. Wheth­er abstain­ers choose not to be as social or wheth­er organ­izers of social occa­sions involving drink­ing exclude abstain­ers is unclear. Abstain­ers may prefer to inter­act with oth­er abstain­ers or less social people. Altern­ately, abstain­ers might not be invited to social gath­er­ings, work-related or otherwise, because drink­ers con­sider abstain­ers dull.

Corcor­an et al. (1980), Mont­gomery (1991), and Put­nam (2000) each made con­vin­cing cases that social net­works are import­ant for find­ing jobs and earn­ing pro­mo­tions. Mont­gomery (1991) explained that com­pan­ies prefer acquaint­ances of employ­ees because employ­ees screen poten­tial can­did­ates and thereby reduce the cost of search. Approx­im­ately half the work­ers sur­veyed in the Pan­el Study of Income Dynam­ics found their job through friends or rel­at­ives, and one-third repor­ted help from acquaint­ances in obtain­ing their job (Corcor­an et al., 1980). There­fore, a per­son with more con­tacts will have more labor mar­ket options (Burt, 1997). Gran­ovet­ter (1995) sug­ges­ted that a large quant­ity of weak ties or friends-of-friends may be most import­ant to gar­ner­ing the best job offers.

Thus, if social drink­ing enables great­er social net­works, it will also increase earn­ings. In terms of search the­ory: the more one drinks, the more people one knows, and the more people one knows, the lower the mar­gin­al costs of search.

The study is packed full of excel­lent ref­er­ences to pub­lished stud­ies (as you can tell from the above excerpt), so I sug­gest read­ing the access­ible (and very short!) report. It’s also worth not­ing foot­notes four and five, describ­ing how this is just like all invest­ments in cap­it­al, in that an optim­al level exists: “you must drink more than 21 drinks per week to earn as little as a non-drink­er”.

via @phila_lawyer

Equal Societies Good for All

The more unequal a soci­ety’s income dis­tri­bu­tion, the more health and social prob­lems ail both the rich and the poor.

With this the­ory brought to his atten­tion through the “quite fas­cin­at­ing book“ The Spir­it Level, Nic­olas Bau­mard dis­plays the evid­ence to sup­port the the­ory that eco­nom­ic inequal­ity is bad for all inhab­it­ants of a coun­try before con­sid­er­ing some pos­sible explan­a­tions, and look­ing at what this means in terms of poverty and cli­mate change.

It is com­mon know­ledge that in rich soci­et­ies the poor have short­er lives and suf­fer more from almost every social prob­lem. In [The Spir­it Level], [the authors] demon­strate that more unequal soci­et­ies are bad for almost every­one – the well-off as well as the poor […]. The remark­able data the book lays out and the meas­ures it uses are like a ‘spir­it level’ which we can hold up to com­pare the con­di­tions of dif­fer­ent soci­et­ies. The dif­fer­ences revealed, even between rich mar­ket demo­cra­cies, are strik­ing. Almost every mod­ern social and envir­on­ment­al prob­lem – ill-health, lack of com­munity life, viol­ence, drugs, obesity, men­tal ill­ness, long work­ing hours, big pris­on pop­u­la­tions – is more likely to occur in a less equal soci­ety.

Base­ball fan? Bau­mard also points out that “the more equal the salar­ies in a base-ball team are, the bet­ter its per­form­ance”.

India and the Definition of Middle Class

A newly pro­posed inter­na­tion­al defin­i­tion of the middle class for devel­op­ing coun­tries, pro­duced by the Cen­ter for Glob­al Devel­op­ment for the World Bank, has some sur­pris­ing con­clu­sions for India.

The report, pro­duced by the pres­id­ent of the Cen­ter for Glob­al Devel­op­ment, Nancy Bird­sall, sug­gests that “middle class” is defined as every­one with an income above $10 a day, exclud­ing those in the top 5% of earners in the coun­try… mean­ing India has no middle class.

This is a com­bin­a­tion both of the depth of Indi­a’s poverty and its inequal­ity. China had no middle class in 1990, but by 2005, had a small urb­an middle class (3% of the pop­u­la­tion). South Africa (7%), Rus­sia (30%) and Brazil (19%) all had siz­able middle classes in 2005. […]

In socio-polit­ic­al terms, the middle class is tra­di­tion­ally that seg­ment of soci­ety with a degree of eco­nom­ic secur­ity that allows it to uphold the rule of law, invest and desire sta­bil­ity. They do not, unlike those defined as rich, depend on inher­it­ances or oth­er non-pro­duct­ive sources of income. […]

OECD coun­tries define their poverty lines as 50% of medi­an income which works out […] to about $30 day. In the US the poverty line for a single indi­vidu­al in 2008 was $29 per day and for each indi­vidu­al in a four-per­son house­hold was about $14 per day.

How­ever, people in devel­op­ing coun­tries liv­ing on even $10 a day still have extremely low social indic­at­ors. Eco­nom­ist Lant Pritch­ett has shown that infant mor­tal­ity of house­holds in the richest quin­tile in Bolivia was 32 and Ghana 58 per 1,000. Few­er than 25% of people in the richest quin­tile in India com­plete 9 grades of school, Pritch­ett showed. “An upper lim­it of the 95th per­cent­ile, while on the high side, is just about suf­fi­cient to exclude the coun­trys richest,” Bird­sall adds.

via The Browser

Abstraction to Increase Effort (and Spending)

When there is a medi­um placed between our effort and a desired out­come, we strive to max­im­ise this medi­um regard­less of wheth­er or not it leads optim­ally to that out­come (think points or vir­tu­al cur­ren­cies as a medi­um when attempt­ing to obtain goods).

That’s my attempt at a con­cise sum­mary of the find­ings from a study coin­ing the phrase ‘medi­um max­im­isa­tion’.

This example taken from the paper (pdf) and presen­ted by The New York Times may help:

Stu­dents were giv­en a choice between two simple tasks. One would take six minutes, and the stu­dents were told that they would get a gal­lon of Haa­gen-Dazs vanilla ice cream as a reward. The oth­er would require sev­en minutes of work, and the pay­ment would be a gal­lon of Haa­gen-Dazs pista­chio.

Not sur­pris­ingly, since the second option involved more work and a less pop­u­lar fla­vor, only about a quarter of the stu­dents chose it.

But the research­ers also repeated the exper­i­ment with a couple of tweaks. In the new ver­sion, the six-minute task led to a pay­off of 60 points, and the sev­en-minute task brought 100 points.

The research­ers then told the stu­dents that any­one who fin­ished with between 50 and 99 points would be giv­en a gal­lon of vanilla ice cream. Any­one with 100 points would get pista­chio.

Prac­tic­ally, there was no dif­fer­ence between the two exper­i­ments. But the out­comes ended up being very dif­fer­ent.

In the com­ments of a pre­vi­ous post of mine look­ing at the denom­in­a­tion effect, the idea that “the great­er the level of abstrac­tion, the more ready we are to spend” was mooted. So it seems to be the case here.

via @BFchirpy