The Economist 29Feb2020

54 Business The EconomistFebruary 29th 2020

2

Bartleby Whenrankleadstorancour

I

n davidmamet’sfilm,“GlengarryGlen

Ross”,a groupofAmericanproperty

salesmenareforcedintoa contestto

maximisesales.Thetoptwowillget

prizes;thebottomtwowillbefired.The

playcomesacrossasa critiqueofthe

corruptingeffectof“dog-eat-dog”capi-

talismandputtingperformanceaboveall

else.Butiscompetitionbetweenem-

ployeesaneffectivewayofimproving

overalloutcomesforbusiness?

JanWoike,fromtheMaxPlanckIn-

stituteinBerlin,andSebastianHafen-

brädl,oftheiesebusinessschoolin

Barcelona,trytoanswerthequestionin

anarticle*fortheJournalofBehavioural

DecisionMaking. Theytestedwhether

performancerankinghelpedorhindered

groupeffort.

Theirapproachwastousea “public

goods”gameinwhichparticipantsare

giventokenswhichtheycaninvest.They

hadthechoiceofinvestinginanindivid-

ualprojectorinvestingcollectively.Two

differentversionsofthegamewere

played.Inbothgamesreturnswere

higherif everyonecollaborated.Butin

oneversion,investingintheindividual

projectimprovedtherelativerankingof

theparticipant,eventhoughthereturns

toboththeindividualandthegroup

werelower.

Participantsinthegameincluded

somestudentsandsomeexperienced

managers.Theresearchersobservedno

significantdifferenceinthewaythetwo

groupsplayedthegame.Whatmattered

wastheformoffeedback.Inoneversion

ofthegame,individualsweretoldhow

welltheyscored andhowwelltheywere

performingrelativetotherestofthe

group.Inanother,theywereinformed

abouthowwellthegroupasa wholewas

performing,relativetothemaximum

possiblereturn.

Predictably,thesecondfeedbackmech-

anismledtomoreco-operation.Less

obviously,informationonindividual

performancerelativetofellowgroup

membersledplayerstofavourmovingup

thepeckingorderovernotjusttheir

group’scollectivereturns,butalsoover

theirmaterialwellbeing.Theywerewill-

ingtoforgoguaranteedfinancialgains;

achieving“status”wasmoreimportant.

Astheauthorsnote,thisresulthas

implicationsformostorganisations.

“Rankingfeedback,whichisoftenusedin

organisationalsettings,promptspeopleto

perceiveevensituationswithco-operative

outcomestructuresascompetitive,”they

write.Peoplemaynotbeinnatelyco-

operativeorcompetitive;theymaysimply

respondtocuessetbytheorganisation

theyworkfor.

Destructivecompetitionwouldbea

particularproblemforthosecompanies

whichuseso-called“agile”management

approaches,inwhichstafffromdifferent

departmentsareorganisedintoteamsand

askedtoworktogether.Insteadofbeing

agile,suchteamsmaywrestlethemselves

toa standstill.

Theresearchalsoraisesmoreques-

tionsabouta managementapproach,

dubbed“rankandyank”,underwhichall

employeesareratedyearlyandthose

whofallintothelowestcategoryare

liabletolosetheirjobs.Rankingsystems

ofthiskind,associatedwithJackWelch’s

tenureasbossofge, anengineering

giant,from1981 to2001,havebeenthe

subjecttoincreasedacademicscrutiny.

Studyafterstudysuggeststhattheyhurt

overallperformance,notleastbylow-

eringproductivity.

Businessesneedtocompetewith

theirrivalsbutwithinthefirm,co-oper-

ationisnormallymuchmoreusefulthan

competitiverivalry;a housedivided

againstitself,cannotstand,asAbraham

Lincolnsaid.Competitiverankingseems

notjusttoreduceco-operationand

fosterselfishnessbutalsotodiscourage

risk-taking.Suchfindingshave

promptedmanybossestoyank“rankand

yank”.Microsoftabandonedit in2013.

TheEconomistisa genuinelyco-oper-

ativeplace(althoughBartlebyislocked

ina DarwinianstrugglewithSchumpeter

fortherighttoa full-pagecolumn).If it

wasn’t,journalistswouldbereluctantto

passoncontactsorstorytipstotheir

colleagues,andsectioneditorswould

constantlyrubbishthesuggestionsof

theirpeers[asit is,weonlydoit occa-

sionally,ed.].

In“GlengarryGlenRoss”twoofthe

salesmenconspiretorobtheoffice,steal

someofthebestsalesleadsandsellthem

toa rivalbusiness.If yousetupa dog-eat-

dogsystem,youriskhavingthehounds

turnaroundandbitetheirowner.

Hownottogiveemployeefeedback

.............................................................

* “Rivals without a cause? Relative performance

feedback creates destructive competition despite

aligned incentives”

formation campaign designed to influence

America’s election in 2016. When subse-

quently asked by Mr Levy whether he

thought she had “let him down”, Mr Zuck-

erberg offers only a pause, followed by a

non-committal response.

The author’s access risks putting him in

thrall to his subject. He is not afraid to

chronicle Facebook’s failures. But his tone

is occasionally fawning. He recounts how

Mr Zuckerberg reacted to a question about

the wisdom of Instagram’s founders selling

their photo-sharing app to Facebook “as if

he were a chess grandmaster, startled by a

move from an inferior player who sudden-

ly shifted the board to his disadvantage”. At

times Mr Levy can seem too quick to accept

the tech industry’s macho self-image, for

instance in his description of an internal

team charged with driving new users to Fa-

cebook as “a data-driven Dirty Dozen

armed with spreadsheets instead of com-

bat rifles”.

In recent years Facebook has hired le-

gions of moderators to check up on its us-

ers, and fortified them with automated

monitoring systems. But its chief defence

against accusations of harm is one to

which Mr Levy seems mostly sympathetic:

that from the crooked timber of humanity,

no straight thing was ever made, not even a

social network. It is a belief that Mr Zucker-

berg seems to hold sincerely. It is tactically

useful, too, because while it contains more

than a grain of truth, it also minimises the

firm’s culpability.

In the end, Mr Levy sees Mr Zuckerberg

as a Utopian genius undone by the world’s

lamentable wickedness; a man who “set

out to connect a world that was perhaps not

ready to be connected”. Not everyone will

be so generous. 7