CentrePiece - eprints.lse.ac.ukeprints.lse.ac.uk/47043/1/CentrePiece_12_1.pdf · The Magazine of...

C e n t r e Pi e c eISSN 1362-3761

Immigrant identity in BritainWhen central banks talk

Educational controversies

Teaching valuesTrade winners and losers

Blair’s economic legacy

The Magazine of The Centre for Economic Performance Volume 12 Issue 1 Summer 2007

MICROSOFT’SMARKET

POWER

As one prime minister hands over to hissuccessor, there is much talk about thelegacy of the former and the challengesfacing the latter. As a contribution to thisevaluation of past and prospective policies,the Centre for Economic Performance (CEP)is producing a new series of policy analysesaround some of the defining slogans ofNew Labour: the priorities of ‘education,education, education’; the goal of ‘makingwork pay’; and the desire to be ‘tough oncrime, tough on the causes of crime’.

The first is available on our website –http://cep.lse.ac.uk/briefings/default.asp –and more will follow on these and othertopics during the course of the transitionto the new regime. Here, CEP’s directorJohn Van Reenen provides a broadoverview of Tony Blair’s economic legacy.One thing is for certain, he concludes:over the next two or three years, GordonBrown will not have the luxury of beingable to blame the policy mistakes of aprevious government for any unfortunateeconomic news.

CEP also plans to take stock of its ownperformance, looking back at some of thebig ideas that have emerged from theresearch and the stories of theirsubsequent impact on both scholarshipand public policy. For example, studies ofthe causes and consequences ofunemployment and inequality in the 1980sand 1990s not only changed the wayeconomists think about these problems,they also led to the development of labourmarket policies like the New Deal and theminimum wage – and later, following theirimplementation, to evaluations of theireffectiveness.

In the meantime, in this issue ofCentrePiece, we showcase findings from allsix of our major research programmes –labour markets; education and skills;productivity and innovation; globalisation;macroeconomics; and wellbeing.

Two articles look at current controversies ineducation: the effects on pupils’achievement of attending a grammar

school or a religiously affiliated school.Among the others, CEP researchersquestion if there really is a potential ‘clashof cultures’ in modern Britain; explore theeffects of openness to trade; examine howmarkets react to monetary policy-makers’words and deeds; and call for schools toteach values.

And our cover story focuses on a product – Microsoft’s Windows software –with which we are all familiar but whose ramifications are often obscure. A government’s policies on technology andcompetition rarely have the publicprominence of its policies on education,work and crime, but the consequences canbe equally significant.

Romesh [email protected]

CentrePiece is the magazine of the

Centre for Economic Performance at the

London School of Economics. Articles in this

issue reflect the opinions of the authors, not of

the Centre. Requests for permission to

reproduce the articles should be sent to the

Editor at the address below.

Editorial and Subscriptions Office

Centre for Economic Performance

London School of Economics

Houghton Street

London WC2A 2AE

Annual subscriptions for one year (3 issues):

Individuals £13.00

Students £8.00

Organisations (UK and Europe) £30.00

Rest of world £39.00

Visa and Mastercard accepted

Cheques payable to London School of

Economics

CEP director, John Van Reenen

CEP research director, Stephen Machin

Editor, Romesh Vaitilingam

Design, Raphael Whittle

Print, Ghyllprint Ltd

Cover Illustration, John Minnion

© Centre for Economic Performance 2007

Volume 12 Issue 1

(ISSN 1362-3761) All rights reserved.

Centre Piece

Editorial

CentrePiece Summer 2007

1

page 2 The European Commission versus Microsoft:competition policy in high-tech industriesCEP research investigates whether the software giant has abused its monopoly power

page 8 Culture clash or culture club? The identity andattitudes of immigrants in BritainAlan Manning and Sanchari Roy doubt there is a threat from minoritygroups who refuse to see themselves as British

page 14 Words and deeds: market reactions when central banks talkCarlo Rosa and Giovanni Verga measure the impact of European Central Bank communication

page 18 Happiness and the teaching of valuesRichard Layard argues that we need teachers specifically trained to teach values

page 26 Widening access to grammar schools: theeducational impact in Northern IrelandEric Maurin and Sandra McNally assess who gains from the academic track

Contents in brief...

page 12 Sharing the fruits of tradeCEP research explores the impact of tradeopenness on countries, communities,regions and sectors

page 24 Faith primary schools: betterschools or better pupils?Stephen Gibbons and Olmo Silvaexamine pupil achievement in religiouslyaffiliated primary schools

page 28 Blair’s economic legacyJohn Van Reenen on why Labour isstruggling to convert economic success intothe political currency of popularity

page 8 Culture clash or culture club?

page 24 Faith primary schools


The antitrust cases against Microsoft in the United States andEurope have been the most high profile implementation ofcompetition law in the last 20 years. Christos Genakos, Kai UweKühn and John Van Reenen look at the key economic issues,notably what they imply for the conduct of competition policy inhigh-tech industries dominated by rapid innovation.


3

In the various Microsoft cases,antitrust authorities in the UnitedStates and the European Union(EU) took on one of the mostvaluable companies in the world

and its CEO Bill Gates, the world’s richestman. After five years of investigation, inMarch 2004, the European Commissionheld Microsoft guilty of abuse of itsdominant market position under Article 82of EU law and imposed the largest fineever for such an antitrust violation inEurope: €497 million.

The Commission found that Microsofthad abused its monopoly of personalcomputer (PC) operating systems in twoways: ‘deliberately restrictinginteroperability between Windows PCsand non-Microsoft work group servers,and by tying its Windows Media Player, aproduct where it faced competition, withits ubiquitous Windows operating system.’(Work group servers are computers thatallow people to share files and printing,store and protect large amounts of data,access the internet, etc.)

The Commission also demanded majorremedies, including compulsory licensingof intellectual property: ‘within 120 days,to disclose complete and accurateinterface documentation which wouldallow non-Microsoft work group servers toachieve full interoperability with WindowsPCs and servers’; and ‘within 90 days, tooffer to PC manufacturers a version of its

Windows client PC operating systemwithout Windows Media Player.’

This degree of intervention is highlyunusual and has led to a continuedconflict about the implementation of theremedies. The case also raises animportant question about the conduct ofcompetition policy in high-tech industriesdominated by rapid innovation.

Market powerIn the server case, which we focus onhere, the Commission’s basic argumentwas that Microsoft extended its marketpower from PC operating systems (ofwhich Windows controls over 95% of themarket) into a complementary market –that of the operating systems for workgroup servers. How did it do this?

For server operating systems to beeffective, they must be able tocommunicate easily with the PC operatingsystem – what is known as ‘efficientinteroperability’. Microsoft’s control of thePC operating system meant that it couldlimit the efficient interoperability betweenWindows and rival companies’ serveroperating systems by manipulating theinterfaces responsible for connectingWindows with other software.

The Commission argued that Microsofthad both short-run (‘static’) and long-run(‘dynamic’) incentives to ‘foreclose’ rivalsfrom the server operating systems marketin this way. The dynamic reasons are

The EuropeanCommissionversus Microsoft:competition policy in high-tech industries

probably most important, as Microsoft wasclearly concerned that a strong presenceof rivals in server operating systems couldthreaten the profits it enjoyed from itsWindows monopoly of the PC market inthe future.

For example, customers could reducetheir reliance on PCs by runningapplications like spreadsheets, databasemanagement and banking software mostlyon servers, leading to a decline ofMicrosoft’s longstanding monopoly. Byextending the Windows platformdominance from PCs to servers, Microsoftcould extinguish this future threat.

Various internal emails by Microsoftsenior executives suggest that this strategywas not the overzealous imaginings ofEurocrats. For example, in 1997, Bill Gateswrote: ‘What we’re trying to do is use ourserver control to do new protocols andlock out Sun and Oracle specifically… thesymmetry that we have between the clientoperating system and the server operatingsystem is a huge advantage for us’.

This may have just been cheap talk,but as Figure 1 shows, Microsoft’s share ofthe server operating systems market didrise dramatically during the late 1990s:from about 20% at the start of 1996 to

over 60% in 2001. By this point, Novell,the combined UNIX platforms (IBM, Sun, etc.) and Linux could muster onlyabout 10% each in market shares. TheCommission argued that at least some ofMicrosoft’s swift rise to power was due toanti-competitive actions.

There has been much debate overwhether Microsoft was dominant in thework group server market. But the keyissue in a ‘leveraging’ case like this iswhether Microsoft had power over PCoperating systems. Given their 95%-plusmarket share, even Microsoft’s lawyers didnot try hard to contest this point.

Economic incentives toforecloseSo did Microsoft have an economicincentive to foreclose competition throughleveraging? The key question is as follows:when firm A, the monopolist (Microsoft inPC operating systems), faces firm B in acomplementary market (server operatingsystems), in what circumstances will firm Aexclude firm B from the adjacent market?Microsoft’s essential argument rested onthe Chicago School view that amonopolist does not have incentives tomonopolise a complementary market sinceall profits can be extracted at least aseffectively by increasing the price of themonopoly product.

The Chicago argument – known as the‘one monopoly profit theory’ – is thatdegrading interoperability would costMicrosoft lost revenues as consumers would


4

not be willing to pay as much for aWindows operating system due to its lowerperformance with non-Microsoft servers.Instead of going to the expense ofmonopolising the new market throughreducing rivals’ quality, Microsoft couldsimply charge a higher price for its PCoperating system and extract all the profitsfrom the server market in this way.Consequently, the Chicago argument is thatMicrosoft must have benign reasons, suchas its desire to end the excessive profitsearned by other server vendors or thesuperior efficiency of Windows technology.

The modern economic theory offoreclosure suggests many reasons whythis critique might break down. InMicrosoft’s case, it is useful to distinguishbetween dynamic and static incentives.

Dynamic incentives toforecloseThe lack of any long-run incentive toforeclose in the ‘one monopoly profittheory’ arises from the assumption thatthe monopolist has a permanentunchallenged position with no threat offuture entry to the primary market. This isunlikely to hold for Microsoft’s position inthe PC operating systems market.

Although in the short run it wasprotected by strong barriers to entry, inthe longer run, there were a variety ofthreats to Microsoft’s juicy stream ofprofits. Consumers care about thesoftware applications (spreadsheets, wordprocessors, games, etc) that are written ona particular operating system. The maincompetitive advantage of Windows is thewide range of applications written on itsplatform (software developers writeprograms to work on the most popularplatforms). But major platform threatsemerged in the late 1990s associated withthe growth of the internet.

One threat was that increasingnumbers of applications could be deliveredthrough servers. Server operating systems

Figure 1:

The growth of Microsoft’s share in the work group servermarket 1996-2001

0%

10%

20%

30%

40%

50%

60%

70%

80%

2001

Q3

2001

Q1

2000

Q3

2000

Q1

1999

Q3

1999

Q1

1998

Q3

1998

Q1

1997

Q3

1997

Q1

1996

Q3

1996

Q1

■ Win/NT■ Unix■ Netware■ Linux■ Other

Sh

are

of t

he

mar

ket

By degrading the abilityof rival server operatingsystems to work withWindows, Microsoft‘foreclosed’ the market

The Commission'sremedy for foreclosurewas to ask Microsoft to

reveal information toenable server vendors

to connect properlywith Windows


5

typically run on open standards, sosoftware developers could use thesestandards rather than Windows. Thismeant that the server operating systemcould become a potential non-Microsoftplatform, directly challenging thestronghold that Microsoft had created onPC operating systems. If applications onlyneeded a slimmed down version of a PCoperating system, customers would notneed to buy expensive Windowsupgrades.

Effectively, a platform based on aserver operating system could havebecome a potential competitor for theWindows operating system. One way toprevent this danger was for Microsoft tomonopolise the server market – even if thismeant sacrificing profits in the short run.

The key idea in dynamic foreclosuretheory is that an action that shifts short-run market share can have long-runbenefits to the monopolist throughdepressing rivals’ incentives to invest andinnovate. In many cases, these argumentsmay be suspect as there is no obviousmechanism whereby this could take place.But in Microsoft’s case, the mechanism isclear and well established due to the‘applications network effect’. Shifts inshare towards Microsoft in the servermarket (current and expected) will meanthat developers start switching away from

writing to non-Microsoft platforms.Customers will shift away from rivalsbecause there are fewer applications andthis will further reduce developer’sincentives to write software. Thisapplications network effect makesforeclosure arguments much moreplausible than in other industries.

Static incentives to forecloseThe dynamic arguments for foreclosurework even though, in the short run, themonopolist may suffer some losses. Butthese arguments are even morecompelling when there are short-runincentives to foreclose. One such incentiveis the ability to price discriminate moreeffectively in the monopoly market (PC operating systems) by dominating the complementary market (serveroperating systems).

In Microsoft’s case, imagine that thereare two types of customers: large firms(which are less sensitive to the price of thePC operating system) and small firms(which are very sensitive to the price ofthe PC operating system). A monopolistwould like to charge a high price to thelarge firms and a low price to the smallfirms. Microsoft finds this hard to dobecause the large firm can always pretendto be a small firm.

But imagine that large firms also place

a high valuation on a complementaryproduct (servers), whereas small firms donot because the gains from sharingcomputing resources are much smaller. Inthis case, by monopolising the servermarket and charging a higher price for thePC and server operating systems bundle,Microsoft is able to extract more profits.Our research provides empirical evidencethat these short-run incentives exist andthat they have grown stronger over time.

RemediesSoftware markets are fast moving andhighly innovative: many new economyadvocates have argued that Europeancompetition law is inadequate in suchmarkets. In particular, Microsoft arguedthat the proposed remedies of forceddisclosure of interoperability informationwould have a severely negative effect oninnovation, as it would lead to thewholesale cloning of Microsoft’s valuableintellectual property. Whatever thesupposed short-run gains, they arguedthat the long-run costs in terms of lowerinnovation by Microsoft would swampthese purported benefits.

These are difficult areas as theCommission was under no legal obligationto consider the effects on innovation,despite their economic importance.Nevertheless, the Commission arguedthat: ‘a detailed examination of the scopeof the disclosure at stake leads to theconclusion that, on balance, the possiblenegative impact on Microsoft’s incentivesto innovate is outweighed by its positiveimpact on the level of innovation of thewhole industry’.

To assess this claim, we mustinvestigate the Commission’s remedies andtheir likely impact on innovation incentiveson Microsoft, on its rivals and therefore onthe market as a whole. The Commissionasked Microsoft to reveal informationnecessary to allow rivals to interoperatewith the Windows platform. This amountsto a compulsory licensing remedy. TheCommission conceded that Microsoftcould charge a reasonable fee for suchlicenses, reflecting the intellectual propertyembedded in the information.

There is an important distinctionbetween demanding information toenable interoperability compared withimitation. The Commission wants theformer to enable other firms to connectto Windows in the same way telecom

There are many reasonsfor believing that theremedy could have a

positive effect onindustry-wide innovation


6

regulators force fixed line incumbents toshare their network with mobile phoneoperators, even if the incumbent alsooffers these services.

If the remedy allowed imitation – forexample, a complete copy of the keysecurity features of the PC operatingsystem – there would be a strongerconcern over innovation. Consequently,the remedy did not require release ofWindows source code – Microsoft’s ‘crown jewels’.

Interestingly, Windows source code isnot what the rival server vendorswanted. Instead, they were after adetailed technical description of theinterfaces to enable them to design their own code to interoperate withWindows. Microsoft’s description of the remedy as allowing cloning istherefore inaccurate.

The impact on innovationincentivesWhat are the likely effects of the remedyon industry incentives to invest inresearch and development (R&D)? ForMicrosoft’s rivals, there are two effects:

■ Having interoperability informationincreases the value and sales of theirproducts. This will increase rivals’ returns

to R&D, as any innovation will be spreadover a larger number of units sold. Theremedy essentially reduces Microsoft’stax on rival innovation and shouldincrease incentives to innovate.

■ Rivals no longer have to incur costs toovercome technical barriers tointeroperability created by Microsoft’sdisclosure policy. Overcoming suchbarriers is innovation of a sort, but it isduplicative and socially wasteful.

There are several potential effects of the remedy on Microsoft’s incentives to innovate:

■ First, with better disclosure, rivals will beable to compete on a level playing field.To the extent that this reduces theexpected market share and increasesprice competition from now higherquality rival products, the remedy maylead to some reduction in Microsoft’sincentive to invest. But unlike its rivals,Microsoft will still obtain substantialprofits from general innovation in the PCoperating systems market, where it willcontinue to enjoy a monopoly. There istherefore little reason to expect thatMicrosoft’s incentives to innovate onoperating systems solutions wouldsubstantially fall.

■ A further effect may also contributestrongly to increased innovationincentives. Through innovation, a firmcan escape harsh competition with rivalsand secure profits for a transitory period.This effect will tend to increase theinvestment incentives of all firms,including Microsoft. Economic researchis somewhat ambiguous on the netimpact of all of these effects, but onbalance, it is believed that intensifyingcompetition will usually lead to increasedinnovation.

■ Finally, Microsoft may change the qualityas well as the quantity of its R&D. Therecould be positive effects on qualitybecause Microsoft will no longer haveincentives to block innovations that raisequality but have high interoperabilitywith non-Microsoft servers. There issome evidence that Microsoft hassacrificed its own innovative potential toprotect the Windows desktop monopoly.This was known within Microsoft as theWindows ‘strategy tax’ – the need to close down research lines that,although leading to innovative products, could potentially weaken thelock-in of Windows.

In summary, there are likely to bepositive effects on rivals’ innovation fromthe remedy and ambiguous effects onMicrosoft’s incentives. While the eventualoutcome is uncertain, it is far from clearthat the remedy will reduce industry-wideinnovation. On the contrary, there aremany reasons to believe that it could havea positive effect on aggregate innovation.

Interoperability at whatprice?Following the Commission’s decision, themost contentious issue has been theconditions under which the interoperabilityinformation should be licensed and what information was necessary to achievefull interoperability. The Commission leftthe exact conditions out of its initialdecision because it involved intricate reviewof technical information, which wasdelegated to an independent monitoring trustee.

Microsoft’s initial suggestions wereunacceptable to the industry, theCommission and the independentmonitoring trustee appointed to overseethe remedy. Microsoft proposed that theinterface information could only be

Competition policy can deteranti-competitive behaviour

without the need for evertaking legal action


7

purchased as one bundle and specified alicense fee for each rival software copyshipped in the order of magnitude of theMicrosoft software itself. This would haveclearly continued the exclusionary effectsimply through high prices.

Many industry insiders doubted thatany innovation of significance wasembedded in the interfaces themselves.They argued that just changing thelanguage of the interface would not be asubstantive innovation that had materialvalue and that therefore it should not be compensated.

Indeed, to the extent that Microsofthas innovation embedded in processesthat use the interfaces, such innovationsshould not matter for the assessment ofthe license fee because the interfacesthemselves do not constitute theinnovation. Typically such information inother software sectors is licensed at onlynominal fees.

Another contentious issue was theamount and type of information thatMicrosoft had to provide. To interconnectwith Microsoft’s software, rivals neededinformation about how exactly theinterface works. When the trustee foundthat Microsoft was not giving sufficientinformation to make this possible, theCommission stepped in with a ‘statementof objection’ and eventually a further largefine for non-compliance. But this tug-of-war has led to considerable delay in theeffective implementation of the remedy.

ConclusionsWhat more general lessons can be learntfrom the Microsoft case about antitrustenforcement in high-tech markets andelsewhere? First, it is worth rememberingthat the case has gone on for nine yearswith four statements of objections issuedand still no final resolution. This is partly areflection of the complexity of thetechnical issues, the legal necessity of due process and Microsoft’s financialstrength. Many of its server rivals havelong since died.

An obvious problem is that the legaltimescale is so long compared with therapid evolution of these markets. By thetime a remedy is in place, the marketplacehas moved quickly beyond the problemsover which the case was fought: even ifthe judgement and remedy areappropriate, is it ‘too little too late’? In our view, the Commission’s decision on

Microsoft has had some impact since it isprospective and hence gives theCommission power over future versions of Windows.

Although caution is always warrantedbefore intervention, antitrust authoritiescannot take a completely laissez faireapproach to innovation markets. Much ofthe positive impact of competition policy isthrough deterring anti-competitivebehaviour without the need for evertaking legal action. And since softwaremarkets are replete with examples ofsimilar issues, the case may havecontributed to higher deterrence againstanti-competitive exclusionary behaviour.

A second observation concerns thestatus of foreclosure theory. Part of theCommission’s case was an explicitconsideration of economic incentives andan analysis of the effects of the remedy oninnovation. These are clearly importantfrom an economic perspective, eventhough European legal practice is oftenambivalent about getting into these issues.

Despite the difficulty of bringingempirical evidence to bear, considerationof innovation and foreclosure wasunavoidable in making a credibleeconomic case. One of the challengesfacing modern economics is to developguidelines for the type of empiricalevidence that could be used to test thelikelihood of foreclosure being a problemin different markets.

The Commission has been muchcriticised in its use of foreclosure theory inmerger cases. For example, the proposedmerger of General Electric and Honeywellwas blocked after it had been cleared bythe US authorities, only for the judgement(although upheld) to be severely criticisedby the Court of First Instance in 2005.And in 2002, the Court actuallyoverturned the Commission’s blocking ofthe Tetra/Sidel merger in 2001, which wasbased on ‘over-speculative’ theories.

In a sense, foreclosure theory in amerger case is inherently speculative.Opponents of the merger must producearguments that a particular type offoreclosure behaviour is more likely tooccur as a result of the merger, althoughthere are no exclusionary practices in thepre-merger situation.

The evidential position is better in anabuse of dominance case because theexclusionary behaviour is already allegedto have happened, so there can be an

empirical discussion over whether thebehaviour has in fact occurred, whether itcould be justified in terms of efficiency,and whether there has been any materialeffect on the marketplace as a result ofthis behaviour. This was the case withMicrosoft where evaluation was possible.Furthermore, Microsoft’s exclusionarymechanisms were lent credibility by theinternal emails, the kind of evidence thatis rarely seen.

Unfortunately, although foreclosuremay be easier to detect in a case overabuse of monopoly power compared witha merger, remedying the problem is muchharder. In a merger, there is always theclear choice of simply blocking theproposed transaction. Remedies for anexisting monopolist are harder to frameand even harder to enforce.

The Commission and Microsoft havebeen wrangling for a long time over theterms of the disclosure remedy and it isstill not perceived to be effective.Microsoft’s main rivals have reached out ofcourt settlements, so the concern may bethat smaller firms and potential newentrants could be the main parties tosuffer. We are unlikely to have heard theend of this case.

This article summarises ‘The Incentives of a

Monopolist to Degrade Interoperability:

Theory and Evidence from the Personal

Computer and Server Market’ by Christos

Genakos, Kai Uwe Kühn and John Van

Reenen, CEP mimeo, and ‘Some Economics of

European Commission versus Microsoft’ by

Kai Uwe Kühn and John Van Reenen,

forthcoming in Cases in European

Competition Policy: The Economic

Analysis edited by Bruce Lyons (Cambridge

University Press).

Christos Genakos, a research associate in

CEP’s productivity and innovation

programme, is at Cambridge University.

Kai Uwe Kühn is at the University

of Michigan. John Van Reenen is director

of CEP.


Culture clash or culture club?The identity and attitudes ofimmigrants in BritainDoes Britain face a real threat from immigrantsand minorities – particularly Muslims – whorefuse to think of themselves as British? Alan Manning and Sanchari Roy haveanalysed data on the national identity andvalues of both immigrants and British-bornpeople – and they conclude that fears of a‘culture clash’ are seriously exaggerated.

Many people inBritain considerimmigration tobe one of themost importantissues facing

the country today. Their concerns seem tobe as much about the social impact ofimmigration as its economic impact, whicheconomists typically conclude are small. Itis not entirely clear what socialconsequences of immigration are feared –sometimes it is simply the dilution of‘traditional’ culture; sometimes it is a moremelodramatic fear that Britain is becominga mix of mutually incompatible cultures,whose irreconcilable differences could endin a serious ‘culture clash’.

Of course, these concerns are notnew. For example, in April 1990, Norman(now Lord) Tebbit cited his infamous‘cricket test’. The former cabinet ministertold the Los Angeles Times: ‘A largeproportion of Britain's Asian populationfail to pass the cricket test. Which side dothey cheer for? It's an interesting test. Areyou still harking back to where you camefrom or where you are?’

But the current concerns seemheightened, largely because of fears aboutthe integration of Muslims into Britishculture. There is widespread belief that agrowing fraction of Muslims who live (andin many cases were born) in Britain do notthink of themselves as British, have noaspiration to do so and do not want theirchildren to do so either. Instead, it is feared,they subscribe to some other identity,creating little enclaves that resemble, as faras is possible, the countries from whichthey came or a model of the good societyvery different from what is generallythought of as ‘Britain’.

Such fears tend to be magnified bythe statements by some British Muslims,which appear explicitly to reject a Britishidentity and affirm another one. One ofthe July 7 bombers appeared in a videosaying ‘your democratically electedgovernments continuously perpetuateatrocities against my people and yoursupport of them makes you directlyresponsible, just as I am directlyresponsible for protecting and avengingmy Muslim brothers and sisters’. The use of the words ‘your’ and ‘my’clearly expressed the people with whomhe identified.

The problem is that most British

people know so little about Muslims thatit is very hard to know how widespreadthese feelings are. Our research usesresponses from the Labour Force Survey ofalmost one million individuals to thequestion ‘What do you consider yournational identity to be?’ as well as data onpeople’s views of their rights andresponsibilities from Home OfficeCitizenship Surveys. The answers give littlesupport to the idea of a serious cultureclash within British society.

The British-bornAmong those who were born in Britain,over 90% of all groups of whateverreligion or ethnicity, think of themselves asBritish. In particular, there is no evidencethat Muslims are less likely to think ofthemselves as British than other groups.Our interest in this topic began with theresponses of Muslims, but we came to theconclusion that it was unfair to singlethem out for special attention as they donot stand out in any way.

Table 1 shows the proportions ofBritish-born people of different religions,who think of themselves as British. Ofthose describing themselves as Christian,99.1% report themselves as British. But ofthose describing themselves as Muslim,the proportion is slightly higher at 99.2%,exceeded only by those who are Jewish.Percentages reporting a British identity arelower for Buddhists, Sikhs and Hindus, butare above 95% for all groups. It is hard tolook at these figures and see grounds forconcern. Of course, this does not meanthat the Muslims see themselves as Britishand not Muslim – it is just that they seeno conflict in being both.

Ethnicity has a somewhat larger effecton British identity than religion, as can beseen from Table 2. All non-white ethnicgroups report lower levels of Britishidentity, but this is probably because manyof them are second-generationimmigrants. If we look at young people,those from ethnic minorities whoseparents are British-born report the same


9

Table 1:

Percentage of British-bornpeople reporting British as their national identity,by religion (controlling for ethnicity)

Religion Percentage reporting

British identity

Christian 99.1%

Buddhist 95.7%

Hindu 96.1%

Jewish 99.6%

Muslim 99.2%

Sikh 95.6%

Any other religion 97.0%

No religion 98.8%

Total 99.0%

Contrary to what many people seemto believe, Britain is not riven by a large-scale culture clash

Table 2:

Percentage of British-bornpeople reporting British as their national identity,by ethnicity (controlling for religion)

Ethnicity Percentage reporting

British identity

White 99.1%

Mixed: White/Black Caribbean 97.6%

Mixed: White/Black African 95.3%

Mixed: White/Asian 95.2%

Mixed: Other 91.8%

Indian 94.2%

Pakistani 93.4%

Bangladeshi 94.9%

Other Asian 90.5%

Black Caribbean 93.5%

Black African 94.2%

Other Black 95.3%

Chinese 91.9%

Other 79.4%


10

Figure 1:

Immigrants’ views of their national identity and years since arrival

0 10 20 30 40

0

20%

40%

60%

80%

100%

Slovakia

France

Somalia

Japan

ChinaUSA

Australia Poland

Italy

TurkeyIran

BangaldeshPakistan

Hong KongUganda

IndiaCanada

JamaicaEgypt

Ireland

Cyprus

Malta

Average years since arrival

Per

cen

tage

rep

orti

ng

Bri

tish

id

enti

ty

levels of British identity as the whitepopulation.

There is, however, one group thatstands out as having an extremely low levelof British identity – Catholics fromNorthern Ireland. From our research, itappears that any identity conflict amongBritish-born Muslims is an order ofmagnitude smaller than that amongCatholics from Northern Ireland.

ImmigrantsSo far, we have focused on people born inBritain, but what about immigrants? Thefraction of immigrants who identifythemselves as British varies a lot by countryof birth. But there is a simple explanationfor most of this variation – how longimmigrants have been in the country?

Figure 1 shows that new immigrantsalmost never think of themselves as British,but the longer they remain in the countrythe more likely they are to do so. Thisprocess of assimilation is faster for some

immigrant groups than others, but not inthe way that might be expected. Forexample, Muslims are not less likely to feelBritish than those from other backgrounds,and immigrants from Pakistan andBangladesh assimilate into a British identitymuch faster than the average, while thosefrom Western Europe and the UnitedStates do so more slowly, with Italians andIrish standing out as the groups thatassimilate least into a British identity.

We also find evidence that immigrants

There is no evidence thatMuslims are less likely tothink of themselves as Britishthan other groups

The longerimmigrants remain in Britain, the morelikely they are tothink of themselvesas British

from poorer and less democratic countriesassimilate faster into a British identity. Partbut not all of this can be explained by agreater tendency to take up citizenship.

Rights and responsibilitiesThis last finding might lead one to arguethat whether people think of themselvesas British is not a meaningful indicator ofwhether they feel they belong, nor of theirintegration into British life and values.There is little concern about the fact thatItalians rarely seem to come to think ofthemselves as British because it is felt thatItalians have similar views on the way inwhich society should be run.

So it is conceivable that those born inBritain call themselves British because thatis what their passports say they are butthat they espouse a variety of diversevalues. To examine the values that may liebehind notions of British identity, we alsoconducted an analysis of people’s views onrights and responsibilities. Our findings


11

Immigrants frompoorer and less

democraticcountries

assimilate fasterinto a British

identity

here are very similar to those on nationalidentity: immigrants are very slightly lesslikely to have views on rights andresponsibilities that the popular consensusholds to be ‘desirable’.

But the differences are much smallerthan the differences among the British-born population of different ages and withdifferent levels of education. What’s more,the immigrant groups who emerge ashaving different values from the British-born population are not the ones thathave become the focus of most publicconcern. Muslims, for example, do nothave significantly different values.

These findings strongly suggest that,contrary to what many people seem tobelieve, Britain is not riven by large-scaleculture clash. This is not to deny theexistence of some people who areprepared to use violence to further theiragenda, but our evidence suggests thatthese are a tiny minority.

Our conclusion is supported by

evidence from the 2003 British SocialAttitudes Survey, which asked respondentsto say whether they agreed or disagreedwith the statement ‘Muslims are moreloyal to Muslims than to Britain’. Of thenon-Muslim respondents, only 9%disagreed with a further 25% neitheragreeing nor disagreeing. But among theMuslim respondents (who we mightexpect to be better-informed on thesubject) 45% disagreed, a significantdifference.

The survey also found that 62% ofnon-Muslim respondents thought thatthere was a fairly or very serious conflictbetween Muslims and non-Muslims inBritain, compared with 27% of Muslims. A broader question about conflictsbetween Muslims and non-Muslims in theworld as a whole found 85% of non-Muslims saying they thought that therewas a fairly or very serious conflict,compared with only 67% of Muslims.

In presenting our research findings atvarious universities, we have beensurprised by how many people react bysaying our results are all wrong and thatthey ‘know’ that there is a serious cultureclash. We should be seriously concernedthat they maintain this stance even whenfaced with evidence against it. While theremay not be much of a problem withimmigrants and minorities in Britain notthinking of themselves as British, theremay be a bigger problem in the refusal ofthe indigenous white population to seethese groups as British.

This article summarises ‘Culture Clash or

Culture Club? The Identity and Attitudes of

Immigrants in Britain’ by Alan Manning and

Sanchari Roy, CEP Discussion Paper No. 790

(http://cep.lse.ac.uk/pubs/download/

dp0790.pdf).

Alan Manning is professor of economics at

LSE and director of CEP’s research

programme on labour markets. Sanchari

Roy is a PhD student and research

assistant at LSE.


12

Sharing the fruits of tradein brief...

One of the defining features of globalisation is increasingopenness to trade – the removal of a whole range ofbarriers to the free flow of people, goods, services andcapital. Some of these trade barriers can take the form ofduties and tariffs levied at national borders. Others arecaused by standards – regulations covering labour,environmental issues and health and safety – which differbetween countries and regions. Still others arise fromgeography, for example, inaccessibility or a lack oftransport infrastructure.

As these barriers are dismantled by economicliberalisation, new transport networks and constantinnovation in information and communicationstechnologies, so the impact of trade openness oncountries, firms, regions, communities and economicsectors is coming under closer scrutiny.

One CEP study of the effects of new transportinfrastructure focuses on the US interstate highwaysystem, spanning over 40,000 miles, and mostly

constructed between 1956 and 1975. The highways hadthree goals: to improve the connection between majormetropolitan areas in the United States; to serve USnational defence; and to connect with major routes inCanada and Mexico. As an unintended consequence ofmeeting these goals, the highways crossed many ruralareas, making it possible for researchers to examine theircausal impact on the local economy.

The research by Guy Michaels finds dramatic effects. A country where distances were long, travel was slowand most economic activity was highly localised began tointegrate across its land mass. It soon became apparentthat the new highways had a big impact not onpassenger vehicles but on the large trucks that havebecome the primary mode of cross-county commerce.

The highways increased trucking income and retail sales by 7-10% per capita in the rural counties theycrossed, relative to other rural counties. This suggeststhat highway counties took advantage of the reduction

One of CEP’s core research themes is the impact of tradeopenness on countries, firms, regions, communities andsectors. Two recent studies confirm the gains fromopening up trade – but recognise that addressing theuneven outcomes of globalisation is as big a challenge aspursuing liberalisation in the face of entrenched interests.

Success breeds success, often for the vast majority – but

the fruits of commerce are notalways enjoyed by all


13

in trade barriers to increase their trade with other counties.

More surprising perhaps are the findings on how thehighways affected the demand for skills in rural areas.On average, the highways had no effect on the demandfor high-skilled workers relative to low-skilled workers.But the highways increased the demand for high-skilledworkers compared with low-skilled workers in countieswhere skills were abundant and reduced demand whereskills were scarce.

In other words, by opening new markets, the highwaysdisproportionately benefited high-skilled workers whereskills were abundant and hurt them where skills werescarce. This finding is consistent with the view that tradeincreases the demand for the abundant factor.

Other CEP research on trade liberalisation – by StephenRedding and colleagues – sheds more light on how thereduction in barriers to trade reallocates economicactivity. Here too, the pattern that emerges is one ofsuccess reinforcing success.

At the most basic level, the researchers find that areduction in trade barriers encourages simultaneous jobcreation and job destruction in all industries, but thatgross and net job creation vary with country and industrycharacteristics.

Significantly, there is a net loss of jobs in industries with acomparative disadvantage (those where relative labourand non-labour costs are high), while industries withcomparative advantage enjoy net job creation as joblosses due to exiting firms are exceeded by jobs createdby the entrance and expansion of high-productivity firms.

Likewise with productivity levels and average firmoutput: the gains from liberalisation are greatest inindustries with comparative advantage. Interestingly, it isin these industries that the research finds the highestlevels of 'creative destruction' of firms, which mayexplain why workers in these more dynamic sectorsreport higher levels of job insecurity.

That the overall effect of opening up trade is to increase

aggregate welfare is not in doubt, as both studies makeclear. As average productivity increases, so the price ofgoods is driven down, reinforcing the virtuous circle. Butwelfare gains are not experienced uniformly, and just assome regions or sectors have to grapple with thehandicaps of a low skills base or high input costs, so toodo developed and less developed economies.

Contrary to the anti-globalisers, success breeds successnot just for the few but for the many – and often for thevast majority. But the fruits of commerce are not alwaysenjoyed by all. Addressing those uneven outcomes is asbig a challenge for politicians as pursuing liberalisation inthe face of entrenched interests. Most agree on the needto ‘manage‘ globalisation. That shouldn’t mean reining itback, but helping everyone to jump on board.

This article draws on research described in

‘The Effect of Trade on the Demand for Skill –

Evidence from the Interstate Highway System’ by

Guy Michaels, CEP Discussion Paper No. 772

(http://cep.lse.ac.uk/pubs/download/dp0772.pdf)

and ‘Comparative Advantage and Heterogeneous

Firms’ by Andrew B Bernard, Stephen J Redding

and Peter K Schott, Review of Economic Studies

73(1): 31-66, 2007.

Guy Michaels is a lecturer in economics at LSE

and a research associate in CEP’s labour markets

programme. Andrew Bernard is at Dartmouth

College. Stephen Redding is research director of

CEP’s globalisation programme. Peter Schott is at

Yale University.

Transportinfrastructure

like the US interstate

highway systemincreases trade

Do announcements by theEuropean Central Bank(ECB) affect marketexpectations about thefuture direction of its

monetary policy? And if so, how longdoes it take for these messages to beincorporated into asset prices?

Our findings suggest that when thetone of the press conference differs fromwhat the market expects, there is asizeable and immediate reaction in theprice of three-month Euribor futures –widely traded futures contracts based onshort-term interest rates in the eurozone –and the complete price adjustment usuallytakes no more than an hour. Therefore,central bank communication turns out tobe a crucial and effective instrument forsteering market interest rates.

The fact that market participantsshould react to central bank

announcements is quite surprising foracademic economists, especially after theseminal work of Kydland and Prescott(1977). They were awarded the NobelPrize in 2004 for showing, in a theoreticalframework, that policy-makers face asevere problem of ‘time-inconsistency’ –they tend to renege systematically on theirpast promises – and so it is in the publicinterest to ignore their declarations.

Only 20 years ago, William Greiderpublished his popular book Secrets of theTemple on the US Federal Reserve System.At that time, it was common practice formonetary policy to be surrounded bysecrecy. But nowadays, it is clear thatcentral banks increasingly emphasiseverbal communication with the markets inan attempt to inform them of the likelydirection of future monetary policy. This isvery much the case with the ECB, whichholds all-important news conferences at its


14

What is the minute-by-minute effect of centralbank communication on asset prices? New research by Carlo Rosa and GiovanniVerga finds that the unexpected component ofannouncements by the European Central Bankhas a sizeable impact on interest rate futures.

Wordsmarket reactions when central banks talk

Central bankcommunicationis a crucial and effectiveinstrument for steeringmarket interest rates


15

Governing Council meetings, and it is alsoincreasingly true of the Bank of Englandand the Fed.

But even though central bankers putgreat emphasis on verbal communication,economists have not yet systematicallystudied the relationship between centralbank words and central bank deeds, andespecially not the information content ofcentral bank words elicited by the reactionof financial markets. Our study is anattempt to begin to fill this obvious gap inthe research literature.

We examine the minute-by-minuteeffect of central bank announcements onasset prices. Contrary to what is commonlyfound in the literature of monetaryeconomics, we show that marketparticipants respond to two differentpieces of news rather than just one pieceof news. So in order to correctly describethe conduct of monetary policy, we find

that two dimensions are needed: centralbank actions and, most importantly, centralbank words.

We apply our empirical analysis to theECB’s communication policy because of itsunique institutional feature that on oneday, at two different points in time, theGoverning Council announces first itsmonetary policy decision, and then its likelyfuture monetary policy stance. These eventstake place in sequence. At 12:45 GMT, theECB communicates the new level of itspolicy rate with a press release. Then 45minutes later, at 13:30 GMT, the monthlypress conference starts and the ECB’spresident explains the monetary policydecision to the public, as well as theGoverning Council’s assessment of currenteconomic developments. This speech is veryimportant, especially for traders, because itconveys strong hints about the likely pathof the ECB's future monetary policy.

The use of high-frequency intradaydata rather than daily data to analysecentral bank communication is cruciallyimportant because it allows us fully toexploit the unique institutional feature ofECB monetary policy conduct: we are ableto disentangle the effects of one surprise –the monetary policy shock – from the other– the news shock.

The advantages of using high-frequencydata are best illustrated by means of theexample in Figure 1, where we consider theminute-by-minute movements of the three-month Euribor futures price on a specificGoverning Council meeting day, 6 April2006.

On that occasion, the ECB did notchange its policy rate and this was fullyanticipated by financial market participants.In fact, there is basically no market reactionat 12:45 (indicated in Figure 1 by the firstvertical red line). On the other hand, the

and deeds:


16

1999 2000 2001 2002 2003 2004 2005 2006

-0.6

-0.4

-0.2

0

0.2

0.4

2.96

3.00

3.04

3.08

3.12

7 8 9 10 11 12

Figure 1:

Futures price dynamics on 6 April 2006

Figure 2:

Monetary policy shocks: market reactions to ECB deeds This chart plots the difference between the new policy rate communicated at 12:45, and the one-month

Euribor rate quoted at 10:00. This quantity is normalised by adding the mean equilibrium (liquidity and

risk) spread between the policy rate and the one-month Euribor rate, in the specific case 0.11.

The horizontal axis represents the time of the day while the vertical

axis represents the futures rate. The first red line is at 12:45 GMT,

when the ECB communicates the new level of its policy rate with a

press release. The second red line is at 13:30 GMT, the start of the

monthly press conference.

In Frankfurt, Mr Trichetsays ‘strong vigilance’ –and traders in Londonreact immediately

futures rate fell sharply at around 13:50. It took only twenty minutes: from 3.085 at 13:33, the futures rate fell to2.98 at 13:56.

What explanation could there be forthis quick market response? In his monthlyintroductory statement, the ECB’s presidentJean-Claude Trichet did not use the key

word ‘strong vigilance’, which seems toindicate a strong risk of policy ratespikes in the near future. This was agreat news shock for traders in London,because they had expected a stronghawkish announcement such as theappearance of the signalling word‘vigilance’ on their screens.


17

Everything became clear when ajournalist drew the president's attentionto the matter by asking:

‘Mr Trichet, the markets wereexpecting you to say vigilance inorder to prepare them or prepare foran interest rate rise in May. You didnot say vigilance, was thatdeliberate?’

To which Mr Trichet deftly replied:

‘It is our responsibility to be as clearand transparent as possible withmarket participants, investors andsavers. I would say that the currentsuggestions regarding the highprobability of an increase of rates inour next meeting do not correspondto the present sentiment of theGoverning Council. I would also addthat the sentiment that I see fromtime to time in some remarks ormarket literature concerning theperception that we do not increaserates when we are out of Frankfurt isequally not at all the sentiment ofthe Governing Council. (…) And it istrue, vigilance is not mentioned inthe introductory remarks, as you verywisely remarked.‘ [Emphasis added]

Although the example in the figure isan exceptionally strong case, it serves toillustrate an important and more generalpoint: the ECB is able to move assetprices using words alone, without anyneed for deeds. Obviously to maintain itscredibility, the ECB ultimately has to

deliver what it says it will, that is, tomatch words with policy deeds.

In fact, to complete the story, in May2006, there was no policy rate change bythe ECB. But at the Governing Councilmeeting of June 2006, which was actuallyheld outside Frankfurt (in Madrid), therewas a policy rate increase of 25 basispoints, as Mr Trichet fully signalled to thepublic in the April press conference.

The second main finding of ourresearch is that central bankcommunication can affect asset pricespermanently as well as transitorily. But it took around three years for traders to learn not only how to interpret ECBannouncements but also to believe in them fully.

In this regard, it is revealing to see the changing pattern over time as displayedin Figure 2, which plots the differencebetween what the ECB does and what themarket expects the ECB to do.

Nowadays, market interest ratesrespond only very marginally to ECB deedsbecause the current decision is completelyexpected most of the time and thus alreadypriced in. Indeed, the true reaction takesplace when the president himself makesstatements, especially when their contentdiffers from what the market expects, asshown in Figure 1. Nevertheless, changes inthe ECB policy rate had a significant effectin the first years of the bank’s life when thepublic was still in a learning phase.

We conclude by reiterating that centralbankers’ announcements are a verypowerful tool in driving marketexpectations and even possibly in drivingthe evolution of the real economy.

This article summarises ‘The Impact of

Central Bank Announcements on Asset Prices

in Real Time: Testing the Efficiency of the

Euribor Futures Market’ by Carlo Rosa and

Giovanni Verga, CEP Discussion Paper No. 764

(http://cep.lse.ac.uk/pubs/download/

dp0764.pdf).

Carlo Rosa is a research fellow at Ente Luigi

Einaudi in Rome and an associate in CEP’s

macroeconomics programme. Giovanni Verga

is a full professor of economics at the

University of Parma.

Further readingFinn Kydland and Edward Prescott (1977),

‘Rules rather than Discretion: The

Inconsistency of Optimal Plans’, Journal

of Political Economy 85(3): 473-91.

Carlo Rosa (2006), ‘Central Bank

Communication: The Case of the European

Central Bank’, PhD thesis at the London

School of Economics.

Carlo Rosa and Giovanni Verga (2007),

‘On the Consistency and Effectiveness of

Central Bank Communication: Evidence from

the ECB’, European Journal of Political

Economy 23(1): 146-75.

13

13:3012.45

14 15 16 17 18

Market interest rates now respond only very

marginally to ECB deeds,more often to ECB words


18

Happiness and the teaching of values

A major purpose of schools must be to helpdevelop good and happy people – especially at a time when growing numbers of children are suffering from emotional disturbance.Richard Layard argues that we need a newcadre of teachers specifically trained to teachvalues and the ways to happiness.

Ever since I first readJeremy Bentham, Ihave been convincedthat the best society isone with the greatesthappiness and

(especially) the least misery. I have becomeeven more convinced of this since modernneuroscience has shown that happinessand misery are objective phenomena andthat what people say about themselvesprovides a great deal of evidence aboutwhat they actually feel.

So what do people say? It is not veryencouraging. Despite huge increases inliving standards, people in Britain, Japanand the United States report themselvesno happier today than people did 50 yearsago. Moreover, a variety of studies ofdepression and anxiety disorders indifferent countries suggest that theseproblems have probably increased inprevalence.

Evidence for Britain certainly indicatesbig rises in emotional disturbance. Forexample, the proportion of 16-year-oldswith serious emotional problems rose from10% in 1986 to 17% in 1999 (Collishawet al, 2004). A study based in the West ofScotland showed similar changes over thatperiod (West and Sweeting, 2003). Andthe latest research by Stephan Collishawand his colleagues, which covers thewhole country and comes right up to2006, shows that things have continuedto worsen.

So what is going on? In the mostcareful study of patterns of happiness,economist John Helliwell has used four

waves of the World Values Survey done in46 countries over the last 20 years. Hetakes as his ‘dependent variable’ theaverage happiness in a country andexplains 80% of the variance by sixvariables, four reflecting ethical values:

■ The perceived trustworthiness ofindividuals.

■ The trustworthiness of government andthe courts.

■ The community involvement ofindividuals.

■ The divorce rate.

Individualism and the fall in trustI want to focus particularly on thetrustworthiness of individuals. Eachsurvey respondent is asked ‘Do you thinkmost other people can be trusted?’ Theproportions saying ‘yes’ vary enormouslyand are highly correlated with the resultsof a Reader’s Digest experiment, whichleft wallets in the streets of differentcountries and then counted the numberof wallets returned to their owners. Trustis very high in Scandinavian countries,and this helps to explain why thesecountries generally come out as thehappiest countries in surveys.

Similar findings about personaldecency emerge from a survey that asked11-15-year-olds ‘Are most of yourclassmates kind and helpful?’ 70% ormore said ‘yes’ in Scandinavian countries,but the figures were considerably lowerfor some other countries: 53% in theUnited States, 46% in Russia and 43% inBritain (UNICEF, 2007).

The sad fact is that in Britain and theUnited States, there has been anextraordinary fall in trust since 1960:from roughly 56% of the populationtrusting most other people to only 30%today. Levels of trust in Britain areparticularly low among young people:only a quarter say ‘yes’ when asked ifthey trust most other people (Park et al,2004). By contrast, there has been no fallin trust in any other European countrysince 1981 when the data were firstcollected. My explanation of this is thehuge growth of individualism in the


19

It should be anexplicit aim of each

school to traincharacter and provide

moral education

United States, which has washed acrossthe Atlantic like a tsunami, hitting Britain first and the rest of Europe muchless, so far.

By individualism, I mean the view thata person’s main goal should be to makethe most of themselves or, more vulgarly,to be as successful as possible comparedwith other people. Put that way, it is ofcourse a zero-sum game, and if that iswhat people value, there is no way oursociety can become happier. To do that,we have to move to a positive-sum gamein which we each care positively aboutthe well-being of others.

Learning the ways tohappinessHow can we do this? I think it requires aneducational revolution in which a centralpurpose of our schools is to teach youngpeople about the main secrets ofhappiness for which we have empiricalevidence. Among these ‘secrets’, I wouldinclude:

■ If you care more about other peoplerelative to yourself, you are more likelyto be happy (Lyubomirsky et al, 2005).

■ If you constantly compare yourself withother people, you are less likely to behappy (Schwartz et al, 2002).

■ Choose goals that stretch you, but areattainable with high probability (Nesse, 2000).

■ Challenge your negative thoughts, and focus on the positive aspects of your character and situation(Seligman, 2002).

These are not exactly novel thoughts. Theyare, if you like, the ‘perennial philosophy’,but rigorously established by modernpsychology. The first principle is aboutcompassion to others; the others areabout compassion towards yourself. Theprinciple of compassion for others cannotof course be based on self-interest – if itis, it will fail. Rather, it has to be realisedthrough deliberate cultivation of theprimitive instinct of empathy, which islatent to a varying degree in each of us.

These principles may seem obvious butthey are not easy to put into practice.

Learning hard things takes an enormouseffort. How can we expect people to learnto be happy without massive amounts ofpractice and repetition? Good parts oftraditional religions (especially Buddhism)involve just such practice, but whatinstitutions do we have today that canplay this basic role?

I believe it can only be done byschools. Parents of course are crucial. Butif we want to change the culture, themain organised institutions we have undersocial control are schools. So what shouldthey do? Though I was once a schoolteacher, I certainly don’t have a completeanswer, but let me first report twosuggestive experiments and then draw outsome general principles.

The Penn Resiliency ProjectThe first example is one of a number ofprogrammes designed to build character,which have been subjected to controlledtrials. The Penn Resiliency Project wasdesigned by Professor Martin Seligman,the founder of ‘positive psychology’. In it,15 11-year-old students spend 18classroom hours on such issues asunderstanding their own emotions andthose of others, and developing concernfor others. They are taught by one teacher,who has been trained in the methodthrough eight hours of online self-studyand 10 days of face-to-face training.

The programme has now beenevaluated in 11 different studies and,except in one school where the trainingwas inadequate, it has reduced the rate ofteenage depression over the next three


20

For every subject –including the teachingof values – we shouldknow from controlled

trials what teachingmethods work best

years by on average one half. It has alsoreduced bad behaviour by one third(Reivich et al, 2005).

I believe this programme is an objectlesson in how our educational systemshould develop. It has always amazed mehow little the scientific method is appliedin the classroom, except perhaps in theteaching of reading and number. In thelast three years, I learned a lot about theresults of psychological therapy, all ofwhich are based on controlled trials – andI have become even more amazed at thescarcity of proper evaluation in education.

If something as sensitive aspsychological therapy can be scientificallyevaluated, surely the same should apply toclassroom education? For every subject,we should know from controlled trialswhat teaching methods work best. Andthis applies as much to the teaching ofvalues as to anything else.

The other lesson that emerges fromthe programme is the importance ofdetailed and systematic training of theteachers. As has been found withpsychological therapy, the effectiveness ofa given teacher will vary enormouslyaccording to how well they have beentrained. We no longer need to rely on theinspired amateur.

Of course, controlled trials are onething; delivery on a large scale is another.Having read these findings, it seems to meworthwhile to introduce the programmeon a large scale in Britain. Compared withsmall-scale controlled trials, when half thechildren are excluded, large-scale deliveryto every child in a community should have

bigger effects per child because each childtaking the programme would interact withother children who had also taken it.

If this applied to all the children in acity, it should be possible to modify thewhole youth culture of that city. I wastherefore thrilled that the very imaginativelocal authority of South Tyneside hasdecided to implement the programme inall its schools, while Manchester andHemel Hempstead are using it in a dozenfurther schools. The Department forEducation and Skills (DfES) are paying fora full-scale evaluation.

If the programme is as successful aswe expect, we hope it can contributesignificantly to the first year curriculum for Personal, Social and Health Education(PSHE) in secondary schools, and can then be followed by programmes witholder children.

Values schoolsThe second experiment began in WestKidlington primary school in Oxfordshireand has now been copied in at least adozen state primary schools that callthemselves ‘values schools’. The basic aimis to give the children control of theiremotions by familiarity with uplifting ideas and the practice of silent reflection(Farrer, 2000)

The school has adopted 22 key valuewords – words like ‘honesty’, ‘hope’ and‘respect’ – each taking a turn as the ‘wordof the month’, with which teaching anddiscussion continually connect. Silentreflection is practised daily at the end ofschool assembly, and at the beginning of

most classes. The key question children areasked is ‘What am I like when I most likemyself?’ Ancillary questions are ‘Whatmakes me happy?’ and ‘How can I makeother people happy?’

Four principles of reformWith these two practical examples inmind, let me offer four principles thatshould govern the role of schools incharacter-building.

First, it should be an explicit aim ofeach school to train character and providemoral education. Teachers should standfor clear values and, when asked aboutmoral questions, should make clear whatthey believe. It is not enough to treatmoral issues as interesting topics fordebate. It is more important to train upthe emotions that support moral actionthan the intellectual skills involved, thoughof course both matter.

Second, each secondary school shouldhave specialists in PSHE (or, as the subjectmight more appealingly be called, ‘lifeskills’). And while the whole school shouldreflect the values taught in PSHE, theremust be full-time professional leadership.It is no good having it taught mainly bypart-timers without specialist training,though they can certainly help. This is oneof the most difficult subjects to teach, andmost people who teach it should havetaken it as a specialism in their training.Without a cadre of specialist teachersacting as standard bearers of themovement, there is no chance of theeducational revolution we need.

Third, there is no chance of success


21

unless the movement is grounded inscience. We live in a scientific age, and,although pockets of fundamentalismremain, only science can and shouldpersuade the young about the routes to ahappy society. Now for the first time wehave in positive psychology a science thatprovides the underpinning for moralityand personal liberation.

Fourth, the curriculum shouldinclude: managing your feelings; lovingand serving others; appreciating beauty;love, sex and parenting; work and money;a critical approach to media; politicalparticipation; and moral philosophy. TheDfES has been active in curriculumdevelopment in this area, and the SEALprogramme (social and emotional aspectsof learning) is well-grounded, withexcellent materials at primary school level.

But what I am proposing is somethingmore campaigning and more high profile.We need a government commitment toproducing a major specialism in this area,with a serious teacher trainingprogramme. Moreover, PSHE shouldcontinue up to age 18, and includeprojects that are graded by the school. Itshould also feature in the head teacher’sreport for every university applicant.

Do we need it and can we succeed?Clearly I am talking about a movement ofmoral reform. Do we need it, and can itsucceed?

Let me remind you where we started. Ithink we have a pretty good societycompared with most that have existed.

But we have not become happier in thepast 50 years. We have made enormousprogress in the mastery of nature butnone in the mastery of ourselves – and ifwe want to make further progress in well-being, it has to come mainly from thelatter. If we think we cannot afford thetime because of the threat of globalcompetition, we had better take anintroductory course in economics.

On top of that general case, there isthe specific worry these days about youngpeople. Concerns about their behaviourhas led to the Prime Minister’s ‘respect’programme, which so far has been morerepressive than preventive. What I amsuggesting would contribute significantlyto prevention – and there is seriousscientific evidence that young people arebecoming more disturbed.

So what are our chances of success?Some people would say they are weak.Some programmes in schools focused onparticular problems like drugs have hadlittle success though there is someevidence that wider programmes based onthe whole ethos of the school are moreeffective (Bonell et al, 2007; Weare andGray, 2003).

But I am talking about somethingbigger than a programme; I am talkingabout the reversal of a major culturaltrend towards increased consumerism,increased inter-personal competition andincreased interest in celebrity and money.

Many people assume that culturaltrends go on in the same direction forever.But that is not my reading of Englishhistory. I see something more like cycles:

increased Puritanism in the sixteenthcentury giving way to increased licencefrom around 1660; then from around1830, increased Puritanism again and thegrowth of Victorian values; and in thetwentieth century, again increased licence.So why not in the twenty-first century ashift away from ‘anything goes’ – or atleast a shift towards a morecompassionate society?

But how plausible is the educationsystem as the main catalyst of change?Well, it has been so in the past.Universities played a major role in the


22

We need to create aprofession of fully trained

PSHE teachers, givingevidence-based teaching

that changes lives

Puritan revolution of the sixteenth andseventeenth centuries, and Thomas Arnoldof Rugby did more than anyone to convertthe middle classes to Victorian values.Today, more people spend more of theirlife in the education system than everbefore – so it is the natural catalyst forchange. 50% of young people say thattheir main ambition is to be happy: it’s the most commonly stated ambition andvery sensible too (Park et al, 2004). Let’shelp them.

I have no doubt that new institutionswill also develop for adults. In California,the psychologist Paul Ekman hassuggested chains of ‘compassion gyms’,where you train your mind in compassion,just as you train your body in the physicalgym. I also hope that the churches will domore to help people train their minds inthe mental disciplines that we know canlead to serenity and compassion.

But from a public policy perspective,we must start with schools. This is a goodmoment. People are worried about youngpeople from many angles. We have goodtools with which to help them. The keyneed is to create a profession of PSHEteachers, who give evidence-basedteaching that changes lives, and that goeson to 18.


23

Further reading

Chris Bonell, Adam Fletcher and Jim

McCambridge (2007), ‘Improving Schools

Ethos May Reduce Substance Misuse and

Teenage Pregnancy’, British Medical Journal

334: 614-17.

Stephan Collishaw, Barbara Maughan, Robert

Goodman and Andrew Pickles (2004), ‘Time

Trends in Adolescent Mental Health’, Journal

of Child Psychology and Psychiatry 45(8):

1350-62.

Frances Farrer (2000), A Quiet Revolution:

Encouraging and Sharing Positive Values

with Children, Rider.

Sonja Lyubomirsky, Laura King and Ed

Diener (2005), ‘The Benefits of Frequent

Positive Affect’, Psychological Bulletin 131:

803-55.

Randolph Nesse (2000), ‘The Evolution of

Hope and Despair’, Journal of Social Issues

66: 429-69.

Alison Park, Miranda Phillips and Mark

Johnson (2004), Young People in Britain. The

Attitudes and Experiences of 12-19 Year

Olds, DfES Research Report No. 564.

Karen Reivich, Jane Gillham, Andrew Shatte

and Martin Seligman (2005), A Resilience

Initiative and Depression Prevention

Program for Youth and their Parents, Penn

Resiliency Project.

Barry Schwartz, Andrew Ward, John

Monterosso, Sonja Lyubomirsky, Katherine

White and Darrin Lehman (2002),

‘Maximizing versus Satisficing: Happiness is a

Matter of Choice’, Journal of Personality and

Social Psychology 83: 1178-97.

Martin Seligman (2002), Authentic

Happiness, Free Press.

UNICEF (2007), Child Poverty in Perspective:

An Overview of Child Well-being in Rich

Countries, Innocenti Report Card 7.

Katherine Weare and Gay Gray (2003), What

Works in Developing Children’s Emotional

and Social Competence and Wellbeing?, DfES

Research Report No. 456.

Patrick West and Helen Sweeting (2003),

‘Fifteen, Female and Stressed: Changing

Patterns of Psychological Distress over Time’,

Journal of Child Psychology and Psychiatry,

44(3): 399-411.

Professor Lord Richard Layard is director of

CEP’s research programme on well-being.

His book Happiness – Lessons from a New

Science was published in 2005.

This article is an edited version of the 2007

Ashby Lecture, which he delivered at the

University of Cambridge on 2 May.

The government shouldcommit to producing a

major specialism in thisarea within the

postgraduate certificateof education


24

Faith primary schools:better schools or better pupils?

Government policy in many countries now favours policiesbased on incentives, governance, increased choice andcompetition. In England, this idea has come to be linked tothe expansion of the faith school sector because itsymbolises choice and diversity in the education system, itembodies the kind of practice in admissions andgovernance that policy-makers wish to promote, and –crucially – it offers, so many people claim, highereducational standards.

Arguments for expansion of the faith sector are usuallysupported by evidence of higher average levels ofachievement in religiously affiliated schools compared withthe secular sector. But the fact that we observe higherattainment in the faith sector has little bearing on theadvantages offered by moving a ‘typical’ secular schoolpupil into the faith sector. This is because pupils currentlyattending faith schools are not ‘typical’.

Pupils studying at religiously affiliated schools differ frompupils in secular schools along several dimensions, many ofwhich – such as family background – are correlated withtheir academic achievement. These differences arise in partbecause the parents and pupils who choose faith schoolsmay have different preferences and attitudes towardseducation; and in part because faith schools may haveoperated some forms of ‘covert’ selection in theiradmissions procedures.

in brief...

Average levels of pupil achievement in religiouslyaffiliated primary schools seem to be higher than in the secular sector. Research by Stephen Gibbons andOlmo Silva examines whether faith schools really arebetter or if their pupils are just different.

Religiousaffiliation has little impact on aprimary school’seffectiveness atteaching coresubjects

In our research, we investigate whether faith schools reallyraise pupil attainments more than other schools, orwhether they simply enrol pupils with characteristicsconducive to faster educational progress. We also try tounderstand whether any beneficial impact of attending afaith school comes from its religious affiliation or fromspecific governance arrangements.

To answer these questions, we consider pupils at the endof their primary schooling in England, when they havereached the age of 11. We make use of a large census thatincludes information on pupils’ past and currentachievements, school type and characteristics, place ofresidence (postcode) and schools attended.

Using this information, we compare the test results of 11year old pupils who attend faith primary schools with thoseof pupils in secular primary schools with similar abilities (asmeasured by their achievement at the age of 7) and whoseem to have similar preferences and family background –in particular because they live in the same street or blockof housing and because they go on to attend the samesecondary school.

In addition, we compare the attainments at the age of 11for secular and faith primary school pupils who exhibitsimilar levels of commitment to religious schooling throughtheir choice of secondary school.

The highlights of our research results are that:

■ Faith primary schools only offer a very small advantageover secular schools in terms of test scores at the age of11 in mathematics and English. Moving a ‘typical’secular school pupil into the faith sector would push himor her up the test-based pupil rankings by less than onepercentile.

■ Any benefit of attending a faith primary school is linkedto the more autonomous governance arrangements thatcharacterise ‘voluntary aided’ schools (such as controlover admissions procedures). Pupils in religiouslyaffiliated schools that broadly fall under the control ofthe local education authority – that is, ‘voluntarycontrolled’ schools – do not progress faster than pupilsin secular primary schools.

■ All of the apparent advantage of faith school education

can be explained by unobserved differences betweenpupils who apply and are admitted to faith schools andthose who do not. Pupils who do not attend a faithprimary school up to age 11 but attend a faithsecondary school thereafter perform just as well at age11 as those who attended a faith primary school butthen attend a secular secondary school.

On the basis of this evidence, it seems clear that whetheror not a primary school is religiously affiliated has littlebearing on its effectiveness in educating children in corecurriculum subjects.


25

Any benefit ofattending a faithprimary school is linked to theirmore autonomousgovernancearrangements

This article summarises ‘Faith Primary

Schools: Better Schools or Better Pupils’ by

Stephen Gibbons and Olmo Silva, Discussion

Paper No. 72 from the Centre for the

Economics of Education (CEE) at CEP

(http://cee.lse.ac.uk/cee%20dps/ceedp72.pdf).

Stephen Gibbons and Olmo Silva are both

CEE researchers and active members of

CEP’s wider research programme on

education and skills.


26

more academically orientated, generaleducation.

Specifically, we consider theconsequences of a reform that affectedaccess to grammar schools in NorthernIreland, when England is used as thecomparison group. The two regions of theUK differ in that the grammar schoolsystem has been retained in NorthernIreland whereas it was graduallydismantled in England in the 1960s and 1970s.

The hallmark of the grammar schoolsystem is that children are selected on thebasis of measured ability at the age of 10or 11 whereas in the comprehensivesystem, children of different abilities areeducated in the same schools. The‘selective aspect’ of the education systemis currently under review by the newadministration in Northern Ireland.

Although the education systems ofEngland and Northern Ireland are alsodifferent in other respects (for example,the schools in the latter are mostlysegregated by religion), there areimportant similarities. The two regionshave broadly the same curriculum andthey have the same examinations forstudents at the ages of 16 (GCSEs) and 18(A-levels).

The important considerations for ourresearch are that the examinations arecomparable across the two regions andthat the reform only occurred in one ofthem. That reform consisted of wideningaccess to the more academic track withinNorthern Ireland at the time of the ‘openenrolment’ reform in the late 1980s. Thisis the only differential change thathappened across the two regions.

Our research shows that the reformenabled a very significant increase in thenumber of Northern Irish pupils who couldattend the more academic track (grammarschools) at the end of primary school,between the pre-reform birth cohort

What are the overall effects on educationalattainment of widening access to the moreacademic track? Research by Eric Maurin andSandra McNally investigates using the ‘naturalexperiment’ of the grammar school system inNorthern Ireland, which has survived long afterits dismantlement in England.

It is difficult to know whetherwidening access to schools thatprovide a more academicallyorientated, general educationmakes a difference to average

educational achievement. Although therehas been a shift in this direction in manyOECD countries, reforms have beendifficult to evaluate because they are often accompanied by other importantchanges to the educational system orbecause they have been introduced at thesame time everywhere so there are nocomparison groups.

But the consequences of such reformare deeply controversial and very much acurrent policy issue. In particular,opponents argue that increasing access tothe more academic track harms the qualityof education for everyone withoutimproving the prospects of those able toattend the academic track.

Our research makes use of a unique‘natural experiment’ to identify the neteffects on overall educational outcomes ofwidening access to schools that provide a

Widening access togrammar schools:the educational impact in Northern Ireland

Widening access to the more academic trackcan generate positive net effects

(children born in 1978) and the post-reform birth cohort (those born in 1979).By comparing educational outcomes inNorthern Ireland and England before andafter the reform, we can identify the effectof widening access to the academic trackon overall educational attainment.

Using administrative data before andafter the reform, we find that the openenrolment reform of 1989 (which affectedthe 1979 birth cohort) had a clear impactin Northern Ireland relative to England. A15 percentage point increase in thenumber of pupils enabled to attendgrammar school (at the age of 11) wasaccompanied by shifts of similar magnitudein the number achieving five or moreGCSEs at A*-C and one or more A-level.This suggests a strong causal effect ofexpanding the more academic track onoverall educational achievement.

Just before the reform, there was achange affecting admissions in a qualitativeway. Up to 1988, girls and boys wereassessed in different categories so that thesame percentage of entrants to theadmission test would obtain a given grade(determining whether or not they could beadmitted to grammar school). Following ahigh court ruling in June 1988, thispractice was discontinued and from thenon, girls and boys were assessed together(affecting grammar school intakes in 1989,the 1978 birth cohort).

This change was to the advantage ofgirls since they outperformed boys on theverbal reasoning tests that were the basisof selection. The one-year gap betweenthis qualitative change to admissions andthe open enrolment reform generatedsignificant upward and downward shifts inthe relative proportion of girls enabled toattend grammar school across the cohortsborn between 1977 and 1980.

We find that these shifts have beenfollowed by parallel shifts in girls’subsequent relative outcomes at the agesof 16 and 18. This confirms theconsiderable effect of grammar schoolentry on educational outcomes using adifferent source of identification to thatused for comparing outcomes over time

between England and Northern Ireland.We also replicate this latter analysis forboys and girls separately and confirm ourearlier results.

Thus, whether we compare girls andboys within Northern Ireland or makecomparisons by gender between NorthernIreland and England, it is clear thatgrammar school reforms have a strongimpact on educational outcomes.Furthermore, the design of the educationalsystem – in this case, the mechanism ofentry into grammar schools – hasconsequences for gender differences ineducational outcomes.

As well as considering the overall effectof expanding the academic track oneducational outcomes, we are able to usethe same experiment to consider whetherthe selective system is a contributory factorto observed inequalities between socio-economic groups with regard to latereducational outcomes. Specifically, we cananalyse the effect of the reform accordingto whether children are eligible for freeschool meals, which roughly correspondsto families in the bottom quarter of theincome distribution.

We find that there were big differences

before and after the reform between theprobabilities of lower-income groupsentering grammar school and achievinggood educational outcomes at the ages of16 and 18. The reform had an equalimpact on children with and without freeschool meals in terms of entry to grammarschool and educational achievement at theages of 16 and 18.

Hence, we conclude that grammarschool attendance had no less effect onrelatively disadvantaged pupils than it hadon more advantaged pupils. Therefore, thebarriers that make it difficult for childreneligible for free school meals to entergrammar schools in the first place (such aslower test scores at the age of 11 becauseof lower parental resources) have a long-term effect on inequality due (in part) tothe lower probability of children on freeschool meals entering grammar school.

Although this research cannot beinterpreted as evaluating the overall effectsof a comprehensive or selective (‘tracked’)system of education, it is an example ofwhere widening access to the moreacademic track has generated positive neteffects. It illustrates the high price thatpupils pay for being excluded from theacademic track, even when they are someway down the ability distribution withintheir birth cohort.

The study also provides clear evidence that selection into the moreacademic track really has a causal impact –the improvement in educational outcomes is not simply an artefact of the selection process.


27

This article summarises ‘Educational Effects

of Widening Access to the Academic Track:

A Natural Experiment’ by Eric Maurin and

Sandra McNally, a forthcoming Discussion

Paper from the Centre for the Economics of

Education (CEE) at CEP.

Eric Maurin is research director at the École

des Hautes Études en Sciences Sociales in

Paris and professor at the Paris School of

Economics. Sandra McNally is a CEP

research fellow and CEE’s deputy director.

When more pupils wereable to attend NorthernIrish grammar schools,overall educationalattainment increased

Grammar school attendancehad no less effect on

relatively disadvantagedpupils than on more

advantaged pupils


28

in brief...

The economy is probably the most successful legacy ofthe Blair years. Ironically, New Labour’s economic policieshave been set by his heir, Chancellor Gordon Brown.

Britain has enjoyed 15 years of continuous growthcombined with low inflation. The labour market hasabsorbed a large number of new entrants, especially fromthe wave of migration from Eastern Europe, yetunemployment has remained at historically low levels.Even on the Achilles’ heel of productivity, Britain hasnarrowed the gap with her major competitors and keptup with the American productivity miracle.

So why does Labour have trouble converting theseeconomic gains into the political currency of popularity?Leaving Iraq aside, forgetfulness, fiscal policy and fairnessare the main reasons.

First, the public appears now to take economic prosperityfor granted. People seem more likely to give credit forsuccess to the Thatcher reforms, to the Bank of England,to being outside the euro or to cheap Chinese importsthan to the government. With the exception ofglobalisation, however, thesewere policy choices of thegovernment. Independence ofthe Bank was a bold andsuccessful early move. Overall,Labour has accepted theimportance of competitivemarkets and labour marketflexibility for economic success.

Second, Labour hassignificantly raised tax as a share of national income andspent the money on publicservices. ‘Tax and spend’ is

exactly what socialist governments are supposed to do,of course, but unlike previous Labour governments,neither Blair nor Brown have boasted about it. Nor havethey overspent in the early years of power only to beforced by circumstances to cut back and increase taxes inlater years (as happened most notoriously in 1976 whenthe then Labour Chancellor, Denis Healey, had to turn tothe International Monetary Fund).

The government’s problem is that the public expectsgreater improvements in hospitals, schools and policingfrom their tax pounds than they have seen. It alsoremains to be seen if the increased complexity of the taxsystem – coupled with more labour market regulation –could undermine long-term growth.

Third, unlike the 1980s, the long growth period has notbeen accompanied by rapidly growing inequality. Therehas been a panoply of redistributive policies such as theNational Minimum Wage and tax credits for the lowpaid, which have reduced poverty, especially for workingfamilies. But although inequality in the bottom half ofthe income distribution has narrowed, inequality at the

top has continued to widen.On the left, this has led topressures for more aggressiveredistribution.

Blair leaves behind aneconomy in better shape thanany previous Labour leader. Butwill the voters give hissuccessor, the man who hasoverseen the currentprosperity, the benefit of thedoubt when the next recessioncomes? At the moment itlooks unlikely.

John Van Reenen is director of CEP and

professor of economics at LSE.

Blair’s economic legacy

Brown has overseen the currentprosperity – the question is howvoters will treat him when thenext recession comes

Blair leaves behind an economy in better

shape than any previousLabour leader

Publications

For further information on CEP publications please contact:The Publications Unit


Houghton Street, London WC2A 2AE

Telephone +44 (0) 207 955 7673

Email [email protected]

RESPECTAlan ManningCEP Discussion Paper No. 793May 2007

UNCERTAINTY AND THE DYNAMICS OF R&DNick BloomCEP Discussion Paper No. 792May 2007

ENTRY AND ASYMMETRIC LOBBYING:WHY GOVERNMENTS PICK LOSERSRichard Baldwin and Frédéric Robert-NicoudCEP Discussion Paper No. 791May 2007

CULTURE CLASH OR CULTURE CLUB? THE IDENTITY AND ATTITUDES OF IMMIGRANTS IN BRITAINAlan Manning and Sanchari RoyCEP Discussion Paper No. 790April 2007

DOES THE UNDERGROUND ECONOMY HOLD BACK FINANCIALDEEPENING? EVIDENCE FROM THEITALIAN CREDIT MARKETGiorgio Gobbi and Roberta ZizzaCEP Discussion Paper No. 789April 2007

AMERICANS DO I.T. BETTER: US MULTINATIONALS AND THEPRODUCTIVITY MIRACLENick Bloom, Raffaella Sadun and John Van ReenenCEP Discussion Paper No. 788April 2007

THE EFFECT OF MARITAL BREAKUP ON THE INCOME DISTRIBUTIONOF WOMEN WITH CHILDRENElizabeth Ananat and Guy MichaelsCEP Discussion Paper No. 787April 2007

SEIGNIORAGEWillem BuiterCEP Discussion Paper No. 786April 2007

PRODUCTIVITY GROWTH, KNOWLEDGE FLOWS AND SPILLOVERSGustavo Crespi, Chiara Criscuolo,Jonathan Haskel and Matthew SlaughterCEP Discussion Paper No. 785April 2007

WHY HAS THE BRITISH NATIONALMINIMUM WAGE HAD LITTLE OR NO IMPACT ON EMPLOYMENT?David MetcalfCEP Discussion Paper No. 781April 2007

THE MARGINAL UTILITY OF INCOMERichard Layard, Guy Mayraz andStephen NickellCEP Discussion Paper No. 784March 2007

INFORMATION TECHNOLOGY,ORGANISATIONAL CHANGE ANDPRODUCTIVITY GROWTH: EVIDENCEFROM UK FIRMSGustavo Crespi, Chiara Criscuolo andJonathan HaskelCEP Discussion Paper No. 783March 2007

INFLATION PREMIUM AND OIL PRICE VOLATILITYPaul Castillo, Carlos Montoro andVicente TuestaCEP Discussion Paper No. 782March 2007

Recent CEP Discussion Papers available free from:http://cep.lse.ac.uk/pubs

TEMPORARY AGENCY WORKERS AND WORKPLACEPERFORMANCE IN THE PRIVATE SECTORAlex BrysonMHRL Discussion Paper No. 3May 2007

LEEWAY FOR THE LOYAL: A MODEL OF EMPLOYEE DISCRETIONFrancis GreenMHRL Discussion Paper No. 2April 2007

THE THEORY AND PRACTICE OF PAY SETTING Alex Bryson and John ForthMHRL Discussion Paper No. 1December 2006

Manpower Human Resources Lab Discussion Papersavailable free from:http://cep.lse.ac.uk/pubs/opsp.asp?prog=MHRLDP

CentrePiece is the magazine of the Centre for

Economic Performance at the London School of

Economics. Articles in this issue reflect the

opinions of the authors, not of the Centre.

Requests for permission to reproduce the

articles should be sent to the Editor at the

address below.

Editorial and Subscriptions Office


London School of Economics

Houghton Street

London WC2A 2AE

© Centre for Economic Performance 2007

Volume 12 Issue 1

(ISSN 1362-3761) All rights reserved.

Centre Piece

The back issues

CentrePiece - eprints.lse.ac.ukeprints.lse.ac.uk/47043/1/CentrePiece_12_1.pdf · The Magazine of...

Documents

Transcript of CentrePiece - eprints.lse.ac.ukeprints.lse.ac.uk/47043/1/CentrePiece_12_1.pdf · The Magazine of...