centrepiece - cep.lse.ac.ukcep.lse.ac.uk/pubs/download/centrepiece_16_3.pdf · speed rail line....

ISSN 1362-3761

C e n t r e Pi e c eThe Magazine of The Centre for Economic Performance Volume 16 Issue 3 Winter 2011/12

LobbyingExport successPolicy uncertainty

Letters by economistsSchool inspections

High-speed rail

What makes people happy?

What makes people happy? Nearly 20 years ago, the Centre for EconomicPerformance (CEP) organised its firstconference on this question under thetitle ‘Happiness at LSE’. Since then,understanding wellbeing – and whatpractical actions might be taken topromote it – have been a central part ofthe Centre’s research agenda. At thesame time, promoting happiness hasincreasingly been seen in somegovernment circles as a legitimateambition of public policy alongside themore traditional goals of supportingeconomic growth and employment.

In this issue of CentrePiece, CEP’sfounder director Richard Layard outlinesthe development of happiness research at LSE and its interactions with policy:‘the main aim of social science should beto throw light on the conditionsconducive to happiness’, he concludes.Two further articles look at therelationships between people’s workingenvironments and their wellbeing, and

between crime in a local area and the mental health of both victims andnon-victims.

There is much more about wellbeingon our website, including recordings of apublic debate between Lord Layard andhis fellow peer Robert Skidelsky onwhether happiness is the right measureof social progress, and of the two latestevents to celebrate the Centre’s 21stbirthday. Former CEP researcher AndrewOswald gave a fascinating lecture onherd behaviour and the phenomenon of ‘keeping up with the Joneses’; andNobel laureate Daniel Kahnemandiscussed his acclaimed new book,Thinking Fast and Slow.

Elsewhere in this magazine areseveral topical pieces. Henry Overmanquestions the UK government’s decisionto go ahead with HS2, the new high-speed rail line. Iftikhar Hussainsummarises his evidence that the Ofstedinspection system has proved effective inraising educational standards in England.

And Nicholas Bloom argues that weakeconomic growth in Europe and theUnited States is in part due toheightened uncertainty about economicpolicy-making.

CEP researchers are also deeplyinvolved in a new initiative launched inJanuary: the LSE Growth Commission.Co-chairs Tim Besley and CEP’s directorJohn Van Reenen outline its aims thus:‘Even in times of slow growth andprotracted economic turbulence, it isessential to stay focused on the keydrivers of prosperity over the long term.The LSE Growth Commission will usefrontier research to provide a frameworkfor policy-making in the UK to supportsustainable growth.’ How the growthagenda fits with the happiness agendawill be an interesting challenge forresearchers and policy-makers.

Romesh Vaitilingam, [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, DesignRaphael Ltd

Print, Hastings Printing Co. Ltd

© Centre for Economic Performance 2012

Volume 16 Issue 3

(ISSN 1362-3761) All rights reserved.

Centre Piece

Editorial

CentrePiece Winter 2011/12

1

page 8Crime and mental wellbeingFrancesca Cornaglia and Andrew Leighmeasure the impact of crime – and mediareporting of crime – on the mentalwellbeing of both victims and non-victims

page 14Pupils’ progress: howchildren’s perceptionsinfluence their effortsAmine Ouazad and Lionel Page haveconducted an experiment in whichchildren used pocket money to place smallbets on their exam performance

page 16The returns to lobbyingResearch by Mirko Draca and colleagueson Washington’s ‘revolving door’ lobbyistsgives an indication of the value of politicalconnections in the UK

page 24The letters pageAlan Manning questions the value ofresearchers conducting debates abouteconomic policy through the pages ofnewspapers

in brief...

page 2Big ideas: wellbeing and public policyRichard Layard outlines the development of CEP research onwhat makes people happy and how society might best beorganised to promote happiness

page 6Wellbeing in the workplace: the impact of modern managementAlex Bryson and colleagues explore the links betweenemployees’ wellbeing and their working environment, notably thepractices of ‘high involvement management’

page 10School inspections:can we trust Ofsted reports?Iftikhar Hussain assesses the validity of Ofsted’s school ratings,the impact of a ‘fail’ on pupil performance and the extent towhich teachers can ‘game’ the system

page 18HS2: assessing the costs and benefitsHenry Overman considers the arguments for and againstbuilding a new high-speed rail line from London to Birmingham

page 21Policy uncertainty: a new indicatorNicholas Bloom and colleagues have developed a measure ofuncertainty about economic policy-making

page 26Sequential exporting: how firms break into foreign marketsEmanuel Ornelas and colleagues find evidence of ‘sequentialexporting’ – firms experimenting in nearby foreign markets beforeseeking to become big exporters

page 18HS2: assessing thecosts and benefits

page 14Pupils’ progress: howchildren’s perceptionsinfluence their efforts

Contents

page 8 Crime and mental wellbeing


2

In the latest of CEP’s ‘big ideas’ series, Richard Layardoutlines the development of the Centre’s research onwhat makes people happy and how society might best beorganised to promote happiness.

Big ideasWellbeing andpublic policy


3

How can we maximisepeople’s happiness andwhich institutions, policiesand norms will best achieve

this goal? It was to answer this questionthat the subject of economics wasdeveloped – and it was why I became an economist.

Economics has done a great deal toimprove the human lot. But unfortunatelyit has had too narrow a view of thesources of happiness and misery – toofocused on the standard of living. Thisbecame obvious nearly 40 years ago whenRichard Easterlin pointed to the surveyevidence that despite massive rises inliving standards, happiness was notincreasing (Easterlin, 1974). He attributedthis to the fact that people compare theirincome with other people’s incomes andwith their own recent experience.

In 1980, I wrote an article about thepolicy implications if that were true(Layard, 1980). At that time, there wasvirtually no evidence about thedeterminants of happiness. Over the last30 years, all that has changed and there isnow a vigorous infant science ofhappiness. At the same time, we have hadthe continuing experience of ever risingincomes associated with no increase inhappiness in the UK and several othercountries, including Germany and theUnited States.

From these two influences – thescience and the popular experience – hasemerged a major worldwide movement toestablish subjective wellbeing as theaccepted goal of public policy. Pioneers ofthis movement have been the OECD, theFrench president Nicolas Sarkozy and theUK’s prime minister David Cameron.

CEP researchers have also played animportant part. In the 1990s, AndrewOswald (who is now at the University ofWarwick) pioneered the analysis of micro-data (including from the then newlyestablished British Household PanelSurvey) where the outcome of interest washappiness. He made an enormous impactby using international surveys, such asEurobarometer, to show how similar werethe quantitative effects of badexperiences, such as unemployment andbereavement, in different countries(Oswald, 1997).

Much of Andrew Oswald’s work wasdone with David Blanchflower andAndrew Clark, and all three have become

major figures in the field. He alsoorganised CEP’s first conference onhappiness in November 1993, where theparticipants included Daniel Kahneman, apsychologist who a decade later receivedthe Nobel prize for economics, and twoother psychologists, Peter Warr and KamalBirdi from the University of Sheffield. Atthat time, it was rare for economists andpsychologists to work together.

In 2005, I wrote a book on happiness,now in its second edition, in which I triedto juxtapose the philosophical argumentsin favour of wellbeing with the evidenceon its causes – and thus derive someimportant policy implications (Layard,2011). The book has sold over 150,000copies in 20 languages. From it hasfollowed the Centre’s research programmeon wellbeing, designed to push forwardour understanding but also to producepractical action. We can begin with thepractical action.

The first area here has been mentalhealth. By analysing data on a typicalgroup of adults (participants in the 1970British Cohort Survey, when they were 34years old in 2004), it is possible to explaintheir reported levels of happiness andmisery. The results show that mentalhealth eight years earlier explains fourtimes as much of the misery in our societyas does the level of current family income.

Mental illness also has massive economiccosts, putting over one million peopleonto incapacity benefits.

According to another survey, thenational Psychiatric Morbidity Survey, onein six adults would be diagnosed assuffering from clinical depression orcrippling anxiety disorders, but only 25%of these are in treatment, compared withnearly 100% for most physical illnesses.This is shocking – not only because of theavoidable human misery but also because,as one of our studies shows, if cognitivebehavioural therapy (CBT) were madegenerally available, it would pay for itselfthrough savings on incapacity benefits andlost taxes (Layard et al, 2007).

That 2007 paper was written withLSE’s Martin Knapp, the UK’s leadingexpert on the economics of mental health,and David Clark of the Institute ofPsychiatry at King’s College, who is one ofthe world’s leading experts on CBT. Toimprove things, we formed the LSE MentalHealth Policy Group, which in 2006produced The Depression Report. Thisincluded the proposal to train up to10,000 therapists in the UK to delivereffective therapy services in the NHS (CEP, 2006).

The government essentially acceptedour proposals and they are now beingrolled out nationally as the ImprovingAccess to Psychological TherapyProgramme (IAPT) over a six-year period(2008-14). David Clark and I are activelyinvolved as advisers to the programme,and have published two evaluations of it, (Clark et al, 2009; Gyani et al, 2011) – one on the two pilots and one on the first year of roll-out – whichconfirm the soundness of our originalcost-benefit analysis.

In its original form, the programmecovered only adults and not children. Butin 2007, I became a member of the GoodChildhood Enquiry and was co-author ofits final report (Layard and Dunn, 2009).One key chapter was on child mentalhealth, where, with Stephen Scott ofKing’s College, we developed the proposalfor an IAPT programme for children. The government has accepted a version ofour proposal and the programme beginsin 2012.

It would obviously be better to preventmental illness than to have to cure it onceit has developed. Fortunately, there arenow many programmes for school children

The main aim ofsocial scienceshould be to

throw light onthe conditionsconducive to

happiness


4

that attempt to do that and have shown significant results. One of the best known is the Penn ResiliencyProgramme, developed at the Universityof Pennsylvania using the basic ideas that underlie CBT.

With the help of the YoungFoundation and the Local GovernmentAssociation, we found three localauthorities that were interested in piloting the programme in England.Altogether 22 schools participated andthe programme was found to reduce theincidence of teenage depressionsignificantly, especially for those most atrisk (Challen et al, 2011).

Over 60 schools now use theprogramme, and more and more staff arebeing trained through the mechanism weinitiated. But the programme is quiteshort: it takes 18 hours per pupil. Toachieve a more substantial impact requiresmany more hours.

In the UK, most secondary schoolchildren spend one hour a week for fiveyears on ‘personal, social and healtheducation’ (PSHE), most of whichprobably has no effect. Even the moststructured programme of ‘social andemotional aspects of learning’ (SEAL) insecondary schools has been shown tohave no effect (Humphrey et al, 2010). Toreplace this ineffective method, we have

trawled through all the world’s bestprogrammes and constructed anevidence-based curriculum for 140 hours, which we are hoping, withgovernment backing, to pilot shortly(Layard et al, 2011).

Having a job is a key element ofwellbeing, so we have continued to presstwo of the Centre’s oldest ideas – theapprenticeship guarantee (now an objectof government policy) and a limit to lifeon unemployment benefits. As therecession began, Paul Gregg and Idesigned what we called the JobGuarantee (Gregg and Layard, 2009),which the Labour governmentimplemented as the Future Jobs Fund.Despite favourable evaluations, this hasnow been abolished.

But perhaps our biggest, and leasteffortful, success has been on the nationalmeasurement of wellbeing. In 2008, theOffice for National Statistics decided tohave some work on this andcommissioned me, Robert Metcalfe of theUniversity of Oxford and Paul Dolan, awellbeing economist recently appointedto an LSE chair in social policy, to advisethem (Dolan et al, 2011).

Our proposed questions (see box) arenow being asked of 200,000 people inthe government’s Integrated HouseholdSurvey, and the answers will appearregularly in the country’s official statistics.Through the mediation of the OECD,most advanced countries can be expectedto follow suit.

There have of course been intellectualchallenges to the wellbeing movement. Topromote understanding of our viewpoint,we sponsored a major collaborativeconference on happiness and public policy,the papers from which were published asa special issue of the Journal of PublicEconomics (Besley and Saez, 2008). In it,two colleagues and I investigated the rateat which the marginal utility of incomedeclines as people get richer. We foundthat marginal utility is inverselyproportional to income – an old ideagoing back to the eighteenth centurymathematician Daniel Bernoulli, but neverbefore directly investigated.

More recently, some distinguishedauthors have questioned Richard Easterlin’soriginal hypothesis. But in one paper, wewere able to show that at least in the UK,the United States and West Germany,average happiness has not grown whileaverage real income has shot up. The keyreason is that people mainly value theirincome in relation to the income of others,just as Easterlin first suggested (Layard etal, 2010).

We are now embarking on a majorsystematic study of wellbeing over the lifecourse, with three aims. The first is tobuild a comprehensive model that reallyshows how much different factors matter.The second is to disentangle the true

Evaluation ofmuch of

governmentpolicy could

be undertakenwith happiness

being thecriterion of

benefit

Integrated HouseholdSurvey questions onsubjective wellbeing

I would like to ask you four questions about

your feelings on aspects of your life. There

are no right or wrong answers. For each of

these questions I’d like you to give an

answer on a scale of nought to 10, where

nought is ‘not at all’ and 10 is ‘completely’.

1. Overall, how satisfied are you

with your life nowadays?

2. Overall, to what extent do you

feel that the things you do in your

life are worthwhile?

3. Overall, how happy did you

feel yesterday?

4. Overall, how anxious did you

feel yesterday?


5

Neil Humphrey, Ann Lendrum and Michael

Wigelsworth (2010) Social and Emotional

Aspects of Learning (SEAL) Programme in

Secondary Schools: National Evaluation,

Department for Education

Richard Layard (1980) ‘Human Satisfactions

and Public Policy’, Economic Journal 90(360):

737-50

Richard Layard (2011) Happiness:

Lessons from a New Science, Penguin

(second edition)

Richard Layard, David Clark, Martin Knapp

and Guy Mayraz (2007) ‘Cost-benefit Analysis

of Psychological Therapy’, National Institute

Economic Review 202: 90-98

Richard Layard, John Coleman and Dan Hale

(2011) ‘Using Tested Programmes for

Secondary PSHE’, CEP mimeo

(http://cep.lse.ac.uk/textonly/_new/staff/

layard/pdf/0133_Using_tested_programmes_

for_secondary_PSHE%20-24-03%2011.pdf)

Richard Layard and Judy Dunn (2009) A Good

Childhood: Searching for Values in a

Competitive Age, Penguin

Richard Layard, Guy Mayraz and Stephen

Nickell (2010) ‘Does Relative Income Matter?

Are the Critics Right?’, in International

Differences in Well-Being edited by Ed

Diener, John Helliwell and Daniel Kahneman,

Oxford University Press

Andrew Oswald (1997) ‘Happiness and

Economic Performance’, Economic Journal

107(445): 1815-31

Professor Lord Richard Layard is director of

CEP’s research programme on wellbeing.

Further reading

Tim Besley and Emmanuel Saez (eds) (2008)

‘Happiness and Public Economics’, Journal

of Public Economics special issue

CEP (2006) The Depression Report: A New

Deal for Depression and Anxiety Disorders,

Mental Health Policy Group

Amy Challen, Philip Noden, Anne West and

Stephen Machin (2011) ‘UK Resilience

Programme Evaluation: Final Report’,

Department for Education Research Report

No. 97 (https://www.education.gov.uk/

publications/eOrderingDownload/

DFE-RR097.pdf)

David Clark, Richard Layard, Rachel Smithies,

David Richards, Rupert Suckling and

Benjamin Wright (2009) ‘Improving Access to

Psychological Therapy: Initial Evaluation of

Two UK Demonstration Sites’, Behaviour

Research and Therapy 47(11): 910-20

Paul Dolan, Richard Layard and Robert

Metcalfe (2011) ‘Measuring Subjective

Wellbeing for Public Policy’, CEP Special

Paper No. 23 (http://cep.lse.ac.uk/pubs/

download/special/cepsp23.pdf)

Richard Easterlin (1974) ‘Does Economic

Growth Improve the Human Lot? Some

Empirical Evidence’, in Nations and

Households in Economic Growth: Essays in

Honor of Moses Abramovitz edited by Paul

David and Melvin Reder, Academic Press

Paul Gregg and Richard Layard (2009)

‘A Job Guarantee’, CEP mimeo

(http://cep.lse.ac.uk/textonly/_new/staff/

layard/pdf/001JGProposal-16-03-09.pdf)

Alex Gyani, Roz Shafran, Richard Layard and

David Clark (2011) Enhancing Recovery

Rates in IAPT Services: Lessons from

Analysis of the Year One Data

(http://www.iapt.nhs.uk/silo/files/enhancing-

recovery-rates--iapt-year-one-report.pdf)

causal effects of people’s experiences byproperly controlling for genetic influences.And the third is to use the findings,combined with experimental evidence, toshow how policy evaluation of much ofgovernment policy could be undertakenwith happiness, rather than willingness-to-pay, being the criterion of benefit.

In my view, the wellbeing movement isunstoppable. Happiness is the only goodthat is self-evidently that, a ‘good’ – andwe are coming to know more and moreabout the conditions that make us happyor otherwise. But there is a long way togo and the main aim of social scienceshould be to throw light on the conditionsconducive to happiness and the ways inwhich those conditions can be produced.

The CEP and the LSE are the obviousfocal point for this work. We are proud ofwhat we have done, grateful to thosewho have financed us (especially theEsmée Fairbairn Foundation) and impatientto push further with more completemodels of wellbeing over the individuallife course. London is a great place for allthis, with the fruitful interaction ofeconomists, psychologists and psychiatriststhat we experience in our seminars. I haveno doubt that happiness and wellbeingwill become more and more centralconcepts in our culture – and in laboureconomics, where so many of us began.

Over the past 50 years, wehave had thecontinuing

experience ofever rising

incomesassociated with

no increase in happiness


6

There is a growing body ofevidence that modernpractices of ‘high involvementmanagement’ increase the

profitability of firms. These techniques –which seek to engage employees morefully in their jobs – were first articulatedand advocated by management thinkers inthe early 1980s. But it is only recently that economists have realised their impacton firm performance (Bloom and VanReenen, 2010).

High involvement managementpresumably improves firm performancethrough its impact on employees. Butwhat does it do for their happiness atwork? Modern management practicescould be beneficial for employees’wellbeing – by enriching their jobs, bygiving them more autonomy, by rewardingeffort fairly or by building effective

teamwork – or they could be damaging –by monitoring performance and enforcingtargets in an overbearing way or bydemanding more effort for less pay.

The scant evidence on this question ismainly based on case studies of particularoccupations or self-selected samples ofemployees. Initial studies of highinvolvement management indicated clearproductivity improvements but a barelydiscernible negative impact on employees’wellbeing (Appelbaum et al, 2000). Asecond generation of studies, however,paints a more complicated picture.

For example, high involvementmanagement can enrich people’s work,leading to improvements in jobsatisfaction, but sometimes these come atthe expense of increased absenteeism(Frick and Simmons, 2011). This is becausegreater job demands, which are often

valued by employees, induce stress unlessthey are able to tackle demanding tasks ina fashion that best suits them.

But in many workplaces, job control isnot ceded to employees and, to makematters worse, the social support from

Wellbeing in the workplace:the impact of modern management

High involvementmanagement isassociated with

higher jobsatisfaction, lesstiredness and a

lower probabilityof accidents

How people feel about their jobs is an important part of their overallhappiness yet until now, few studieshave explored the links betweenemployees’ wellbeing and theirworking environment. Alex Brysonand colleagues analyse data fromFinland to assess the impact ofmodern management practices onwellbeing in the workplace.


7

which employees might benefitpsychologically – supportive supervision,for example, or effective unionrepresentation – are also lacking (Woodand Bryson, 2009).

In a new study, we analyse nationallyrepresentative data for Finland, a countryrenowned for being an ‘early mover’ inhigh involvement management. It is also acountry whose work practices are locatedwithin a wage structure and industry mixthat, one might imagine, would beconducive to ‘work enriching’ highinvolvement management rather thanbeing geared towards labourintensification.

Our survey asked employees whichpractices they are exposed to in their job,including autonomous teamwork,performance-based incentive pay,systematic training and information-sharing by management. We estimate the effects of these practices on accidentsin the workplace and three measures of wellbeing:

� Sickness absences, both short- and long-term.

� Subjective wellbeing – that is, self-reported job satisfaction, work capacity,state of health and feelings of tiredness.

� Physical discomfort at work, asmeasured by the experience of pain infour different parts of the body: thelumbar region, legs, arms and neck.

In seeking to identify any causal linkagebetween high involvement managementand employees’ wellbeing, the chiefobstacle is non-random exposure to highinvolvement management. It is likely thatthose who do not feel up to highinvolvement will simply avoid it, potentiallybiasing any positive association betweenhigh involvement management andwellbeing.

We are able to overcome this problemwith very rich data on employee absenceand earning patterns that go back eightyears prior to the survey. We are thus ableto account for patterns and incidence ofprior sickness absence, thus ensuring thatour estimates are not simply a product ofemployee selection.

We find that high involvementmanagement is positively associated withvarious aspects of employees’ wellbeing.In particular, it is strongly associated withhigher evaluations of subjective wellbeing,

including higher job satisfaction and fewerfeelings of tiredness at work. It is alsoassociated with a lower probability ofhaving a workplace accident.

But high involvement management isalso associated with having more shortabsence spells. This may be becauseworking in such a system is moredemanding than standard production andbecause multi-skilled employees cover foreach other’s short absences.

So the nature of high involvementmanagement appears to be important foremployees’ wellbeing, but it is not theonly aspect of modern management thatinfluences wellbeing. Another is the wayin which new practices are introduced.What appears to matter here is employeeinvolvement in the process of consultationand negotiation leading to change.

In a separate study of UK data, wefind that organisational change ispositively associated with increased job-related anxiety, but this anxiety is absentwhere employees are covered by a unioninvolved in organisational changes (Brysonet al, 2012). Where employees are notcovered by a union, however, the anxietyeffects of high involvement managementare still apparent, even if they are involvedin consultation over change.

Overall, this body of research indicatesthe potential benefits of high involvementmanagement for firms, but it also showsthat there are circumstances in which itcan have negative effects on employees’wellbeing. In trying to minimise thesecosts, firms should consider not only whatcombination of practices will improveperformance, but also how best tointroduce those practices and monitor theeffects on their workforce.

This article summarises ‘Does High

Involvement Management Improve Worker

Wellbeing?’ by Petri Böckerman, Alex Bryson

and Pekka Ilmakunnas, CEP Discussion Paper

No. 1095 (http://cep.lse.ac.uk/pubs/

download/dp1095.pdf).

Petri Böckerman is at the Labour Institute

for Economic Research, Helsinki. Alex

Bryson of the National Institute of Economic

and Social Research is a visiting research

fellow in CEP’s labour markets programme.

Pekka Ilmakunnas is at Aalto University

School of Economics.

Further reading

Eileen Appelbaum, Thomas Bailey, Peter Berg

and Arne Kalleberg (2000) Manufacturing

Advantage: Why High-performance Work

Systems Pay Off, Cornell University Press

Nicholas Bloom and John Van Reenen (2010)

‘Human Resource Management and

Productivity’, CEP Discussion Paper No. 982

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

dp0982.pdf)

Alex Bryson, Harald Dale-Olsen and Erling

Barth (2012) ‘Do Higher Wages Come at a

Price?’, Journal of Economic Psychology

33(1): 251-63, earlier version available

as ‘How Does Innovation Affect Worker

Wellbeing?’, CEP Discussion Paper No. 953

(http://cep.lse.ac.uk/pubs/download/dp0953.pdf)

Bernd Frick and Rob Simmons (2011)

‘The Hidden Costs of High Performance Work

Practices: Evidence from a Large German

Steel Company’, mimeo

Stephen Wood and Alex Bryson (2009) ‘High

Involvement Management’, in The Evolution

of the Modern Workplace edited by William

Brown, Alex Bryson, John Forth and Keith

Whitfield, Cambridge University Press

Firms shouldmonitor the

effects of newmanagementpractices on

their workforce

In somecircumstances,

high involvementmanagement

can havenegative effectson employees’

wellbeing


8

In 2006, the US Senate Committee on the Judiciary heardevidence from two sources on the economic cost of crime.The director of the Bureau of Crime Statistics told thecommittee that according to surveys, the financial cost ofcrime to victims and their families is $16 billion annually.Immediately afterwards, economist Jens Ludwig told thecommittee that, based on survey respondents’ willingnessto pay to reduce crime in their communities, the cost ofcrime is $694 billion per year.

This 40-fold disparity between the direct costs to victimsand the wider community’s willingness to pay to reducecrime suggests an intriguing notion. What if most of thesocial cost of crime is not endured by victims but by non-victims? What if the financial impact of crime onthose who are killed, assaulted or robbed is just the tip ofthe iceberg for calculating the costs of crime?

The idea that the costs of crime for non-victims may beimportant was noted by Jeremy Bentham as longago as the late eighteenth century. He gave theexample of a man who is robbed on a road,where the ‘primary mischief’ arises from thephysical harm and loss of possessionsoccurring from the robbery, but the crimealso has a ‘secondary mischief’:

‘The report of this robbery circulatesfrom hand to hand, and spreadsitself in the neighbourhood. It finds

its way into the newspapers, and is propagatedover the whole country. Various people, on thisoccasion, call to mind the danger which they andtheir friends, as it appears from this example,stand exposed to in travelling; especially such asmay have occasion to travel the same road.’

What is important about this effect of crime – whichBentham referred to as ‘the alarm’ – is that it affects amuch larger number of people than the direct impact ofcrime. As philosopher Jonathan Wolff pointed out in the2005 Bentham lecture at University College London, evenif the probability of harm is very low, ‘the fear can beever-present for a great number of people, depressingtheir lives’. The cost of crime may therefore be far larger

Crime and mental wellbeing

Crime has clear costs for its victims, but it might also cause considerable harm toother local people who fear being victims in the future. Francesca Cornaglia andAndrew Leigh measure the impact of crime on the mental wellbeing of bothvictims and non-victims, as well as the effects of crime reporting by local media.

in brief...

Increases in crime rates have a negative impact on the mentalwellbeing of local residents


9

than commonly suggested by methods that evaluate onlythe effects on victims and their immediate family.

In a new study, we combine detailed crime statistics withpanel survey data that provide detailed indicators ofmental wellbeing for the same people over a six-yearperiod, 2001-06. The ‘Household, Income and LabourDynamics in Australia’ (HILDA) survey combines measuresof respondents’ tiredness, difficulties with work and socialactivities because of physical or emotional problems, andnervousness and depression into an overall indicator calledthe Mental Component Summary (MCS).

The HILDA data also include information on which peoplehave been crime victims. This makes it possible to examineseparately the effects that changes in crime in a local areahave on changes in the mental wellbeing of residentvictims and non-victims (controlling for changes in localeconomic conditions).

There is a particular analytical challenge involved ininvestigating the effects of being a victim of crime onpeople’s mental wellbeing: the possibility of ‘sorting’ –that certain people who are more likely to be victims ofcrime may at the same time experience lower mentalwellbeing. As far as we know, no previous study has beenable to estimate the effects of different crimes on themental wellbeing of victims, taking account of thepotential selection of vulnerable individuals into incidencesof crime.

Our analysis finds a strong relationship between being avictim of crime and mental wellbeing for both propertycrimes and violent crimes: a seven percentage point fall inthe MCS for these people. Taking account of the potentialselection of vulnerable individuals suggests that sorting isindeed a problem. Nevertheless, we still find aconsiderable impact on mental wellbeing (a twopercentage point fall in the MCS), which is predominantlydriven by being a victim of a violent crime.

Turning to non-victims, we find significant and sizeableeffects of violent crime on the mental wellbeing of non-victims in the local area. Distinguishing between differentcategories of violent crime, it seems that these effects aredriven by incidences of assaults, including sexual assault,and robbery. Thus, these results provide evidence for thehypothesis that by reducing the wellbeing of non-victims,the costs of crime may be substantial.

We also examine the role of local media coverage inenhancing the effect on mental wellbeing of crime rateswithin an area. Extensive coverage of crime incidences inlocal newspapers may exacerbate the effect of criminalactivity on the mental wellbeing of non-victims. To ourknowledge, no previous work has quantified such effectsnor, more importantly, assessed the ‘multiplier’ effect ofarea crime through media coverage on mental wellbeing.

We find that the intensity of reporting of crime increasesits negative effect on mental wellbeing. This suggests thatmedia reporting plays an important role in enhancing thenegative effect of crime on non-victims in the local area.

This article summarises ‘Crime and Mental Wellbeing’ by

Francesca Cornaglia and Andrew Leigh, CEP Discussion Paper

No. 1049 (http://cep.lse.ac.uk/pubs/download/dp1049.pdf).

Francesca Cornaglia is a research associate in CEP’s labour

markets and wellbeing programmes and a lecturer at Queen

Mary, University of London. Andrew Leigh, a former professor

of economics at the Australian National University, is a

member of the Australian House of Representatives.

Local mediacoverage of

criminal activityenhances the

negative effect of crime on

wellbeing


10

Ofsted inspections of schools have been acentral feature of state education in England fornearly 20 years. Research by Iftikhar Hussainexplores the validity of the school ratings thatOfsted produces, the impact of a fail rating onsubsequent pupil performance and the extent towhich teachers can ‘game’ the system.

School inspections:can we trust Ofsted reports?


11

How to ensure that teachers,doctors and other publicsector employees act in thebest interests of the users of

public services is a challenge facinggovernments around the world. Manyhave responded by making increased useof objective performance targets, such aspupil test scores for schools and patientwaiting times for hospitals.

Empirical evidence suggests thatorganisations given incentives to performon such measures often ‘game’ thesystem, subverting the intention behindthe target. For example, schools might tryto raise their average test score byexcluding lower ability pupils from thetest. Or when given a target for pupils toattain a certain level in key stage tests,teachers might focus on pupils on the margin of this achievement level, at the expense of both high and lowability pupils.

In England, and in the UK morebroadly, top-down performance targetshave often been complemented byinspection regimes. Examples include theschool inspection system run by Ofsted(the Office for Standards in Education), and the Care Quality Commission for thehealthcare sector. In fact, the UK issomething of a world leader in the area of inspections.

The subjective nature of theperformance evaluation undertaken byinspectors holds the promise of‘measuring what matters’. But inspectionsmay open up a whole new can of worms.In particular, just like targets, inspectionsare top-down and, what’s more, theofficials who lead them may be prone tobiases and prejudices. Left to their owndevices to exercise judgement as they see fit, it is unclear whether inspectors willact in the best interest of pupils andparents. Neither pro- nor anti- inspection

arguments have been backed by hardempirical evidence and this research seeksto remedy this gap.

New evidence on inspection systemsIn the absence of previous empiricalevidence – from the UK or elsewhere – thisstudy uses the case of Ofsted inspectionsof state schools in England to providesome evidence on the effectiveness of suchsubjective performance evaluation systems.The first question addressed is whetherinspection ratings provide any extrainformation on school quality, over andabove what is already available in thepublic sphere.

This ‘validity test’ is implemented asfollows: are inspection ratings correlatedwith underlying school quality measures –constructed from teenage pupils’ surveyreports of teacher practices as well asparental satisfaction – after controlling forstandard observable school characteristics,such as test score rankings and theproportion of pupils eligible for free schoolmeals? If they are, then inspection ratingshave the potential to play an importantrole in providing information for parentsconsidering which school they would liketheir children to attend.

The next question addressed iswhether pupil test scores improvefollowing a fail inspection. This is a thornyempirical problem because it suffers fromthe classic problem of ‘mean reversion’ –the idea that a failed school’s testperformance would have improved(reverted to the mean) even in the absenceof a fail inspection.

Relatedly, whether schools are able togame the system and artificially boostperformance following a fail inspection isalso addressed. The post-fail incentives togame are strong and the prior evidence –from England and elsewhere – suggests

that when teachers are put under pressureto raise pupil test scores, they may wellresort to such strategies.

The nature of Ofsted schoolinspectionsBroadly speaking, Ofsted inspections ofschools have three main objectives: first, tooffer feedback and advice to the head andother teachers; second, to provideinformation to parents to aid theirdecision-making; and third, to identifyschools that suffer from ‘seriousweakness’.

Over the period covered by myresearch (2005/6 to 2008/9), schools weregenerally inspected once during aninspection cycle. An inspection involves anassessment of a school’s performance onacademic and other measured outcomes,

Inspectionsproduce ratings

that can helpparents

distinguishbetween more

and less effective schools


12

this issue, I exploit a design feature of theEnglish primary schools’ testing system tosolve this mean reversion problem. Keystage 2 tests for Year 6 (age 11) pupils areadministered in the second week of Mayin each year. These tests are markedexternally, and results are released toschools and parents in mid-July.

The window between May and Julyallows me to address the issue of meanreversion: schools failed in June are failedafter the test in May but before inspectorsknow the outcome of the tests. Thus theMay test outcome for these schools is notaffected by the subsequent fail; neither doinspectors select them for failure on thebasis of this outcome.

By comparing schools failed early in theacademic year – September, say – withschools failed in June of the same academicyear, I can isolate mean reversion from theeffect of the fail inspection. If there is anyrebound in test scores independent of thefail inspection, we should see it in theschools failed in June. As a check on thisapproach, a comparison of observablecharacteristics prior to the fail rating forschools failed in June (the ‘control’ group)and schools failed in the early part of theacademic year (the ‘treatment’ group)reveals negligible differences between thetwo groups.

The results show that a fail inspectionleads to significant improvements in testscores. Furthermore, the largest gains arefor pupils scoring low on the prior (age 7)

followed by an on-site visit to the school.For the visit, inspectors arrive at the

school at very short notice (no more thana few days), which in theory should enablethem to see the school as it ‘really is’ andlimit disruptive ‘window dressing’ inpreparation for the inspections. During theinspection, the inspectors collectqualitative evidence on performance andpractices at the school. This involvesclassroom observations, in-depthinterviews with the school leadership,examination of pupils’ work as well asdiscussions with pupils and parents.

At the end of this process, the schoolis given an explicit headline inspectionrating – 1 (outstanding), 2 (good), 3(satisfactory) or 4 (unsatisfactory, alsoknown as a fail rating) – and theinspection report is made available toparents and posted on the internet.

Do inspection ratings convey any new informationon quality?Previous research suggests that inspectors’findings are reliable, in that two inspectorsindependently observing the same lessonwill come to very similar judgements aboutthe quality of classroom teaching. Theissue addressed here is whether inspectionratings are also valid, in the sense of beingcorrelated with underlying measures ofschool quality not observed by theinspectors. There is almost no empiricalevidence on this question.

I construct a measure of underlyingschool quality from 14-year-old pupils’survey responses to questions aboutteacher behaviour and practices. These data come from the LongitudinalSurvey of Young People in England, amajor survey supported by theDepartment for Education.

The survey asks six questions on howlikely teachers are to: take action when a

pupil breaks rules; make pupils work totheir full capacity; keep order in class; sethomework; check that any homework thatis set is done; and mark pupils’ work.Further analysis using pupils’ reports ofschool discipline as well as parents’ reports of satisfaction levels yields very similarresults to those reported here for theteacher practices outcome.

A composite pupil-level score ofteacher practices at the school is computedby taking the average of the responses tothese six questions. The validity test is thenundertaken by asking the followingquestion: can inspection ratings helpdistinguish between (or forecast) schoolswith good and poor teacher practices?

The critical issue is whether inspectionratings summarise information aboutunderlying school quality that is notalready available in the public sphere. Myfindings demonstrate that on this measureat least, inspectors appear to be doing areasonable job. Even after controlling forthings like test scores and the socio-economic background of pupils, inspectionratings appear to have substantial power inpredicting underlying quality.

These results suggest that parents whoare looking for a good school ought toplace at least some weight on inspectionratings. The actual weight will depend inpart on how much information parentsalready have about the relevant schools,beyond publicly available information suchas test scores.

The effect of a fail inspectionTurning to the effects of a fail rating on aschool’s future test scores, the data showthat pupils’ performance on key stagetests improves following a fail inspection.But whether this is a consequence of thefail inspection or simply a bounce backafter a year or two’s bad luck is a difficultquestion to answer. To make progress on

Children of low-income

parents benefitmost from school

inspections

The inspectorsseem able toidentify poorlyperformingschools, leading totest score gains


13

key stage 1 test. The gains are largecompared with the effects of other policyinterventions that pupils might experience,such as having higher quality teachers or attending a school with higherattainment levels.

These findings are consistent with theview that the children of low-incomeparents – who are, arguably, the least vocalin holding teachers to account – benefitthe most from inspections. Such evidencemay be especially relevant in the currentpolicy environment, where first, there isheightened concern about raisingstandards for this group of children, andsecond, they are hard to reach using otherpolicy levers.

Evidence on gamingbehaviourFinally, this research also looks into possiblestrategic behaviour by teachers. I find littleevidence to suggest that schools failed bythe inspectors are able to inflate theirpupils’ test performance by gaming thesystem. First, teachers do not exclude lowability pupils from the test-taking pool.

Second, although the evidence onwhether teachers target pupils on themargin of attaining the official proficiencylevel (level 4 on the key stage 2 test) ismixed, I find no evidence to suggest thatsome groups (say, very low or very highability pupils) are adversely affected by thefail inspection.

Third, although test gains fadesomewhat over time, there is evidence tosuggest that for some pupils, the gains lastinto the medium term, even after theyhave left the failed primary school. Thissuggests that teachers inculcate reallearning and not just test-taking skills inresponse to the fail rating.

These findings on strategic behaviourare in stark contrast to a significant bodyof evidence demonstrating dysfunctional

responses to test-based performanceevaluation in other settings. Myinterpretation of these results is that bysubjecting schools to close scrutiny,inspectors may play an important role inlimiting such distortionary activities.

ConclusionsWhat are the broader lessons from thisstudy? The findings are particularlynoteworthy given the indications from pastresearch that subjective assessments maygive rise to various biases. For example,there is evidence to suggest that subjectiveevaluations of workers may lead to‘leniency’ and ‘centrality’ bias in privatefirms (Prendergast, 1999). And evidencefrom the public sector points to staffindulging their preferences when allowedto exercise discretion rather than followingformal rules (Heckman et al, 1996).

Although such biases in schoolinspectors’ behaviour cannot be ruled out,this research demonstrates that theinspection system appears to be effectivealong the following two dimensions: first,inspectors produce ratings that make iteasier to distinguish between more and lesseffective schools; and second, they are ableto identify poorly performing schools,leading to test score gains.

One important feature of the Englishschool inspection system is that the keyoutput produced by the inspectors – aninspection rating and report – is availablefor public consumption on the internet.Consequently, inspectors’ decisions arethemselves subject to scrutiny andoversight. One hypothesis for futureresearch is that this is a key element in driving the positive results found in this study.

This article summarises ‘Subjective

Performance Evaluation in the Public Sector:

Evidence from School Inspections?’

by Iftikhar Hussain, Centre for the Economics

of Education Discussion Paper No. 135

(http://cee.lse.ac.uk/ceedps/ceedp135.pdf).

Iftikhar Hussain is a lecturer in economics

at the University of Sussex and a research

associate in CEP’s programme on education

and skills.

Further reading

James Heckman, Jeffrey Smith and

Christopher Taber (1996) ‘What Do

Bureaucrats Do? The Effects of Performance

Standards and Bureaucratic Preferences on

Acceptance into the JTPA Program’, in

Advances in the Study of Entrepreneurship,

Innovation and Growth, Volume 7 edited by

Gary Libecap, JAI Press

Canice Prendergast (1999) ‘The Provision of

Incentives in Firms’, Journal of Economic

Literature 37(1): 7-63

By subjecting schools to closescrutiny, inspectors can play animportant role inlimiting teachers’ability to ‘game’the system


14

The impact of discrimination against particular socialgroups has been widely analysed for its effects on people’saccess to jobs, housing and education. Yet perceptions ofdiscrimination also matter. For example, given that jobsearch takes time and effort, applicants who believe thatpotential employers are likely to discriminate against themmay not apply for those jobs. Similarly, pupils who believethat their efforts will not be rewarded may disengage andthereby reduce their chances of making progress.

How can we measure perceptions of discrimination?Asking individuals directly can provide some information,but it is difficult to infer strong conclusions from suchquestionnaires. Stated perceptions may differ from whatactual behaviour reveals.

Our research brought experimental economics to 1,200pupils in 29 schools across England. Pupils were given anendowment – equivalent to the average weekly pocketmoney in England – and could either keep thisendowment or ‘invest’ it in an exam that would bemarked by either their teacher or an anonymous externalexaminer. If the exam answers were right, pupils coulddouble their weekly pocket money.

The crux of the experiment is that pupils invested theirmoney in the exam only if they believed that their chancesof doubling their investment were large. We asked pupilsa number of questions about the school and theirteachers that allowed us to single out money as the maindriver of their decisions. For example, pupils who believed

Pupils’ progress:how children’s perceptions influence their effortsWhat is the impact of a pupil’s perceptions of how their teacherswill treat them on their motivation, efforts and educationalachievements? To explore this question, Amine Ouazad andLionel Page have conducted an experiment in which schoolchildren could use pocket money to place small bets on theirperformance in an exam.

in brief...

Experimental researchsuggests that boys tendto lower their effortwhen a female teachermarks their exams

that a good relationship with the teacher matters were nomore likely to invest more.

The results of the experiment show that male pupilstended to lower their investment when a female teachermarked their exams. Further analysis confirmed thatfemale teachers in the experiment did tend to awardlower marks to male pupils than external examiners. Somale pupils’ perceptions seem to be roughly in line withfemale teachers’ marking practices. Our results suggestthat male pupils believe that their chances of getting ananswer right are three percentage points lower whenmarked by a female teacher than when marked by anexternal examiner.

Female pupils tended to increase their investment when amale teacher marked their exams. Further analysis,however, showed that male teachers tended to rewardfemale pupils no better than external examiners. Ourresults suggest that female pupils believe that theirchances of getting an answer right are six percentagepoints higher when marked by a male teacher than whenmarked by an external examiner. Overall, female pupils’perceptions were not in line with teachers’ markingpractices: in fact, male teachers tended to reward malepupils more than external examiners.

We also find that ethnicity and socio-economic status (as indicated by eligibility for free school meals) did notplay a role in pupil or teacher behaviour. Gender was the most important driving force behind the results. This is somewhat surprising given the large body ofresearch devoted to teacher biases along ethnic or socio-economic lines.

We also wanted to confront our analytical method withthe traditional method, which directly asks pupils abouttheir perceptions. In our experiments, many pupils saidthey believed that ‘ethnic minorities have equalopportunities at school’ – and pupils who stated itstrongly invested more.

What should we make of these striking results? Our exammainly involved verbal skills, for which there is asubstantial gender gap in favour of girls.Underachievement and dropping out of school are aspecifically male problem, and this experiment sheds newlight on the determinants of this gender gap. Boys oftendisengage in the educational process, and this is likely tobe due in part to their perceptions of their teachers.

In this context, it is notable that, in our experiments, maleteachers induced more investment by boys. There is anunder-representation of male teachers in both primary andsecondary education in England. In primary schools, only15% of teachers are male.

More generally, the experiments reveal that there is nospecific reason why individuals’ perceptions should be inline with actual discriminatory practices. Individuals mayhold the wrong beliefs about the nature of discrimination.And even if they get it roughly right, they mayoverestimate or underestimate the extent of biases.

Furthermore, it is not clear that the discriminatorybehaviour of some individuals – here, male teachersfavouring male pupils – has a direct consequence for thepopulation discriminated against – here, female pupilsexert more effort when assessed by a male teacher.

This could be true in other contexts. The experimentaldesign can be applied to job applications, dating, propertysearches and many other situations. Just by providingmonetary incentives, a researcher can observe perceptionsof discrimination directly from individuals’ behaviourrather than from their stated perceptions. When money isat stake, people start calculating the costs and benefits oftheir actions and reveal what they truly believe rather thanwhat they say they believe.


15

This article summarises ‘Students’

Perceptions of Teacher Biases:

Experimental Economics in Schools’

by Amine Ouazad and Lionel Page,

Centre for the Economics of

Education Discussion Paper No. 133

(http://cee.lse.ac.uk/ceedps/ceedp133.pdf).

Amine Ouazad, an assistant professor at

INSEAD, is a research associate in CEP’s

education and skills programme.

Lionel Page is a research fellow at the

Queensland Institute of Technology.

Girls have better perceptions ofmale teachers – but maleteachers do not reward themmore than female teachers


16

The returns to lobbyingLast autumn’s resignation of Liam Fox, the UK’s defence minister,put lobbying back in the spotlight. Research by Mirko Draca andcolleagues on Washington’s ‘revolving door’ lobbyists gives anindication of the value of political connections in the UK.

in brief...

The resignation of cabinet minister Liam Fox in October2011 was yet another challenge to public confidence inthe UK’s political system. It followed revelations about theexpenses of members of parliament (MPs), controversyover the political power of Rupert Murdoch’s NewsInternational, and severe criticism of former Labourcabinet ministers over the ‘cabs for hire’ lobbying scandal.

But what can we feasibly say about the returns tolobbying activity in the UK? Our study of the value ofpolitical connections in Washington looked at therevenues of US lobbyists who were previously employedby senators. These ‘politically connected’ lobbyists suffereda 24% fall in revenues when the senator that they used towork for left office. This fall in revenue was worth about$177,000 in business and represents the value of aconnection to a sitting legislator.

This figure can be used as something of a benchmark forcalculating the value of being connected to a serving UKcabinet minister. Like senators, cabinet ministers have alot of strategic power in policy-making and theyseem to be the main target of lobbyingactivity in the UK political system.

Applying our US estimates directlysuggests that the median return toa high-level UK connection could bearound £122,000 per year. But it hasto be said that our estimate isspeculative. This is because there is noserious, publicly reported data onlobbying in the UK. We cannotconduct the same type of research herebecause there is no UK equivalent ofthe US Lobbying Disclosure Act.

In practice, the UK lobbying industry is likely to be muchsmaller than the US industry, which is worth $3 billionannually. This is partly because there is less voting alongparty lines in the US Congress and representatives aretherefore more open to influence from lobbyists. In the

Connections to a UKcabinet minister could beworth up to £122,000 a yearfor a working lobbyist


17

‘Revolving Door Lobbyists’ by Jordi Blanes i Vidal,

Mirko Draca and Christian Fons-Rosen is

forthcoming in the American Economic Review.

It was published as CEP Discussion Paper No. 993

in August 2010 (http://cep.lse.ac.uk/pubs/

download/dp0993.pdf).

Jordi Blanes i Vidal is a lecturer in LSE’s

management department and a CEP research

associate. Mirko Draca is a CEP research economist.

Christian Fons-Rosen is an assistant professor at

Universitat Pompeu Fabra. All are part of CEP’s

research programme on productivity and innovation.

Transparency in the UK needs to embracethe US model, where each lobbying

contract is reported on a quarterly basis

UK, cabinet ministers (rather than non-cabinet MPs) arethe focus of lobbying because of their directdecision-making power.

The types of returns to connections seen in the UnitedStates are therefore most likely to be concentrated at thecabinet level. The figure of $177,000 also relates tomiddle ranking lobbyists. At the upper end of the USindustry, ‘superstar’ lobbyists will have practices worthabout $1.5 million per year and a single senate-levelconnection would then be worth $370,000.

The scale of the business dealings of Adam Werritty, LiamFox’s associate, suggests that the return to cabinet-levelaccess in the UK could also be very high. That said,Werritty’s case is unusual because it appears that he wasfunctioning as a lobbyist and as a political adviser at thesame time. This would have been difficult to achieve inthe more transparent US and Canadian systems.

This underscores the need for a UK Lobbying DisclosureAct. Furthermore, transparency in the UK needs to gobeyond a simple ‘register’ of lobbyists – the main currentproposal – and embrace the US model, where eachlobbying contract is reported on a quarterly basis.


18

When governments try to makea decision on big transportprojects, the most desirable

starting point is a cost-benefit analysis,which carefully assesses whether, and byhow much, the benefits to the country willoutweigh the costs. Traditionally, theseanalyses have taken a rather narrow focus,looking at the direct user benefits – thatis, the benefits to people making journeys.

In the case of the HS2 high-speed raillink between London and the North, thesebenefits – faster journeys, less disruption,more capacity – are potentially large.Unfortunately, so are the costs – and boththe costs and benefits are highlyuncertain.

Supporters of the scheme think that

the costs of HS2 are being overstated byfocusing on the total costs and ignoringthe split between revenues and subsidies.They say that fares would reduce the coststo the government from around £30billion to £17 billion.

Opponents of the scheme think thatthe benefits – estimated at around £47billion – are being overstated because theyare partly calculated by turning timesavings on HS2 into money by ‘valuing’the time that people would save. Theproblem with this, they argue, is that thenumber used to value time savings is toohigh because it assumes that people areunproductive on trains.

Some supporters don’t disagree onthis point, but they think that the benefits

are understated because the number ofpassengers will be higher than predicted.They argue that HS2 uses conservativenumbers for passenger growth. There arerailway experts, however, that think thatHS2 has overstated passenger growth byusing projections from more recent years,when the growth of passenger numbershas been high.

In short, there is plenty to argue abouteven if one focuses narrowly on thesedirect user benefits and costs. But asanyone who has been following thedebate will know, the arguments do notjust focus on these narrow issues.Unfortunately, the wider costs andbenefits of HS2 are, if anything, evenmore uncertain.

Public debate continues to rage about thedecision to build a new high-speed rail linefrom London to Birmingham (and possiblybeyond). In the latest in our series on policies of the coalition government, Henry Overmanconsiders some of the arguments for andagainst the scheme – and indicates why it hasbeen so controversial.

HS2:assessing the costs and benefits

Take, for example, the claim by thetransport secretary, writing in theTelegraph to mark the end of the publicconsultation, that high-speed rail is ‘thefast track fix for bridging the North-Southdivide’. This is a clever, but completelymisleading, headline. The article itself ismore tempered – ‘tackling a divide thathas lasted for generations is no easy task’– but it still makes big claims for thepotential impact of high-speed rail.

These claims rely on the assumptionthat reducing journey times (andincreasing capacity) will help firms andworkers in the North to compete moreeffectively for market share in the South(or encourage firms and workers torelocate). But HS2 will also give firms inthe South better access to markets in theNorth. In line with this informal intuition,Nobel laureate Paul Krugman and otherresearchers working on the so-called ‘neweconomic geography’ have shown thatreducing transport costs between ‘core’(the South) and ‘periphery’ (the North)may actually increase disparities, notreduce them. Certainly the evidence onthe direct benefits suggests that theseflow disproportionately to London and theSouth East.

What about the environmentalimpact? In terms of carbon emissions, theimpacts are not large and could well benegative. Most of the journeys on high-speed rail will be additional (or transfersfrom other lines). Extra journeys and fasterjourneys generally require more energy notless (especially if people drive to the trainstation). There will be some savings onlong distance car trips and domestic airtravel – but the latter will be offset if freed

up slots are used for other flights or if HS2 makes getting to Heathrow easierand so generates additional numbers onother routes.

The overall impact depends on what isassumed about how electricity isgenerated. Generous assumptions usingaverage, not marginal, carbon figuresmake the numbers look better. The lasttime I looked at these numbers, HS2 waspredicting a change in average annualemissions in a range from -0.41 to +0.44million tonnes. This is equivalent to just +/-0.3% of current annual transportemissions (and ignores the carbon impactof construction). So the impacts are notlarge and could be negative.

Other environmental impacts arisingfrom the need to bulldoze a straight high-speed train line through somebeautiful countryside are harder to expressin financial terms. These effects clearlyexplain much of the bitter local oppositionfrom people on the route. Actions tomitigate them provide yet anotherillustration of how costly HS2 could prove to be.

For example, the final decision on HS2was delayed while the transport secretarydecided whether to spend an extra £500 million on another tunnel under the Chilterns. To put that figure inperspective, we should note that it is three times the amount allocated to local

Both the costsand benefits of HS2 arepotentially large– but they arealso highlyuncertain

Reducing the costs of transportbetween the South and theNorth may actually increasedisparities not reduce them


19


20

transport projects by the Autumn 2011spending review.

Given all these uncertainties, it is nothard to see why public opinion is sodivided. When the Department forTransport asked people about HS2 lastyear, 47% were in favour of HS2 and 9%against. That sounds like resoundingsupport, until one notes that half of therespondents (50%) agreed that ‘high-speed rail is £30 billion we cannot afford’while only a quarter disagreed with thisstatement. In short, the public are infavour of high-speed rail, providing thatthey do not have to pay for it.

The poll is also interesting on theperceived benefits: 56% of adults agreedthat high-speed rail would be better forthe environment while 63% thought itwould create jobs and growth. As I haveshown, the evidence is fairly weak on theformer claim. And just as with the effectson the North-South divide, the effects onjobs and growth, other than the directimpact of construction, are unknown.(What’s more, construction jobs are part ofthe cost to the economy – a muchmisunderstood point.)

The fact that opinion polls highlightthese two ‘benefits’ suggests that mostpeople don’t know much about thescheme – which turns out to be true: 42%say that they know little or not very muchabout the scheme, while 47% say theyknow nothing.

In short, the costs and benefits of HS2are large and uncertain. I prefer instead tofocus on the opportunity costs: are therethings that we could be doing with £30billion that would yield a higher returnthan ‘£47 billion’? I think the answer isalmost certainly yes, in both the area oftransport – more intra-city schemes, forexample – and more widely.

On the basis of narrow cost-benefitanalysis, this conclusion is backed up bythe Eddington report, published in 2006.Comparing the figures for HS2 with thosefor projects that the Department forTransport had on its books at the time ofEddington suggests that HS2 is, at best, inthe bottom quartile in terms of returns(and indeed, might be closer to being inthe bottom 10%).

One could say that this is irrelevantbecause HS2 has a critical mass that willdeliver wider benefits. But as I haveargued, there is a little evidence tosupport this assertion. If critical mass is

important, then we could considerconcentrating a large amount ofinvestment in particular cities – forexample, Birmingham, London,Manchester and Newcastle. To the best ofmy knowledge, no one has assessed whatsuch a package would look like in termsof the wider impacts.

One final objection to my negativeconclusion might be that ‘we have to haveHS2 because of capacity constraints onthe west coast mainline’. Unfortunately, asthe Eddington report showed, by the timeHS2 is completed, there will be a greatdeal of congestion all over the transportnetwork. Other schemes to tackle thatcongestion are likely to deliver muchbetter returns because these aspects arewell captured by traditional cost-benefitanalysis and, as I have indicated, HS2 doespretty badly on that.

In short, the opportunity costs of HS2are large. To me, this is the fundamentalissue and the reason why I am personallysceptical about the merits of the project.Quite simply, I remain unconvinced thatthis is the best way for the government tospend money. Over the last few years,none of the assertions to the contrary haschanged my mind that this remains thecentral problem with HS2.

Henry Overman is director of the Spatial

Economics Research Centre (SERC), professor

of economic geography at LSE and a research

associate in CEP’s globalisation programme.

Further discussions of HS2 – and

many other policy issues around cities,

regions, transport, housing and the

environment – are available on the SERC blog:

http://spatial-economics.blogspot.com/

Disclosure: Henry Overman sits on the

HS2 analytical challenge panel and has

received funding from the Department for

Transport to look at the wider impacts of

transport investment.

Theopportunity

costs of HS2 are

large: it isunlikely that

this is thebest way

for thegovernment

to spendmoney

The most striking thing aboutthe recent volatility of globalfinancial markets is thatpoliticians are making the

news. The actions of policy-makers andtheir statements about budgets, bailoutsand regulatory reforms are driving thestock market gyrations.

This is not normal. Before the financialcrisis of 2008, stock markets usuallymoved in response to economic news.Strong GDP and employment figureswould send the markets soaring. Poorcorporate earnings would send themarkets crashing.

But today, all eyes are on the policy-makers. Unfortunately, they cannot agree,which is generating massive uncertainty. Infact, according to our new index, whichcharts the evolution of US economic policyuncertainty since 1985, it is now close toits all-time high (see Figure 1). This policyuncertainty is a key factor stalling therecovery and threatening a return torecession.

We construct our index of policyuncertainty by combining three types ofinformation: the frequency of newspaper


21

The damaging impact of economic uncertainty on growth has been reasonably well studied – butwhat happens when there is uncertainty abouteconomic policy-making? Nicholas Bloom and colleagues have developed a measure of this distinct kind of uncertainty, one that showsthe value of restoring stability to current policy actions.

Policy uncertainty:a new indicator

0

100

Bal

ance

d B

ud

get

Act

Firs

t G

ulf

War

Clin

ton

ele

ctio

n

Seco

nd

Gu

lf W

ar

Stim

ulu

s d

ebat

eLe

hm

an a

nd

TA

RP

Ob

ama

elec

tio

n, b

anki

ng

cri

sis

Euro

cri

sis,

201

0 m

id-t

erm

sD

ebt

ceili

ng

dis

pu

te

9/11

Bu

sh e

lect

ion

200

300

2009 2011200720052003200119991997199519931991198919871985

Notes: The index is composed of three sets of measures: monthly news articles containing

‘uncertain’ or ‘uncertainty’, ‘economic’ or ‘economy’, and policy-relevant terms (scaled by the

smoothed number of articles containing ‘today’); the number of tax laws expiring in coming

years; and a composite of quarterly forecasts of government expenditures and one-year

CPI (consumer price index) from the Philadelphia Fed Survey of Forecasters. The data are

available at www.policyuncertainty.com.

Figure 1:

Index of US economic policy uncertainty

US

pol

icy

un

cert

ain

ty i

nd

ex(m

ean

=100

)

articles that reference economicuncertainty and the role of policy; thenumber of federal tax code provisions thatare set to expire in coming years; and theextent of disagreement among economicforecasters about future inflation andfuture government spending on goodsand services.

Our index shows sharp spikes ineconomic policy uncertainty around majorelections, wars and the 9/11 terroristattacks. More recently, it spiked sharplyafter the Lehman bankruptcy inSeptember 2008 and the passage of theTARP (Troubled Asset Relief Program)legislation shortly afterwards. It hasremained high ever since, driven bycontinuing policy uncertainty around the2010 US mid-term elections, the US debtceiling dispute and the crisis of theeurozone.

Of course, policy uncertainty could behigh simply because general economicuncertainty is also high. To test this view,we use Google News listings to constructa broad index of economic uncertaintyand a narrower index focused squarely onpolicy uncertainty.

Comparing these two indices revealsseveral episodes that involve large spikesin economic uncertainty but little or nojump in policy uncertainty (see Figure 2).Examples include the Asian financial crisisof 1997 and several bouts of recessionaryfears in the second half of the 1980s. In short, the data refute the view thateconomic uncertainty necessarily breedspolicy uncertainty.

So why is policy uncertainty so highnow? To identify the drivers of policyuncertainty, we drill into the Google

Figure 2:

Economic policy uncertainty and overalleconomic uncertainty

0

100

1987

sto

ck m

arke

t cr

ash

Firs

t G

ulf

War

Dis

solu

tio

n o

f U

SSR

Clin

ton

ele

ctio

n

Asi

an f

inan

cial

cri

sis

Ru

ssia

n f

inan

cial

cri

sis/

LTC

M

9/111000

2009 2011200720052003200119991997199519931991198919871985

■ Policy-related news-measured economic uncertainty■ Overall news-measured economic uncertainty


22

Notes: The overall news-based economic uncertainty index is composed of the monthly

number of news articles containing ‘uncertain’ or ‘uncertainty’ as well as ‘economic’ or ‘economy’

(scaled by the smoothed number containing ‘today’).

Nor

mal

ised

nu

mb

er o

f n

ews

arti

cles

Recent policyuncertainty hasemerged fromthe US debtceiling disputeand the eurozone crisis

PolicyUncertainty

ambi

guity

am

biva

lenc

e

an

xiet

y

be

wilderm

ent concern confusion dilemm

a disquiet distrust doubtfulness

hes

itanc

y

i

ncer

titud

e

in

deci

sion

irr

esolu

tion misgiving mistrust oscillation perplexity puzzlem

ent quandary

What makes policy uncertainty soharmful? When businesses are uncertainabout taxes, healthcare costs andregulatory initiatives, they adopt acautious stance. Because it is costly tomake a hiring or investment mistake,many businesses naturally wait for calmertimes to expand. If too many businesseswait to expand, the recovery never takesoff. Weak investments in capital goods,product development and worker trainingalso undermine longer-run growth.

How much near-term improvementcould we expect from a stable, certainty-enhancing policy regime? We usetechniques developed by ChristopherSims, one of the two 2011 Nobellaureates in economics, to estimate theeffects of economic policy uncertainty.

The results for the United Statessuggest that restoring 2006 (pre-crisis)

News listings and quantify the mix offactors at work. Several factors underliethe high levels of US policy uncertainty in2010 and 2011, but monetary and taxissues predominate.

One clear example involves the Bush-era income tax cuts originally set toexpire at the end of 2010. Democrats andRepublicans adopted opposing positionsabout whether to reverse these tax cuts. Rather than resolve the uncertaintyin advance, Congress waited till the final hour before deciding to extend thetax cuts.

Other examples include recent movesin the US Senate to increase tariffs onChinese imports, which threaten to set offa trade war. And in Europe, the continuingdebates over potential bailouts forcountries and banks contribute to aclimate of policy uncertainty.

This article summarises ‘Measuring

Economic Policy Uncertainty’ by

Scott R Baker, Nicholas Bloom and

Steven J Davis (www.policyuncertainty.com).

Scott R Baker is at Stanford University.

Nicholas Bloom, also at Stanford University,

is a research associate in CEP’s productivity

and innovation programme. Steven J Davis

is at the University of Chicago’s Booth

School of Business.

2%

1%

0

-1%

-2%

-3%

-4%

-5%

6%

24 26 28 30 32 34 382220181614121086420

■ Average estimate of change in production/employment in response to increased policy uncertainty■ One standard-error bands

0

-1

-2

-3

24 26 28 30 32 34 382220181614121086420


23

Figure 3:

US production and employment after a policyuncertainty shock

Ind

ust

rial

pro

du

ctio

n i

mp

act

(per

cen

tage

dev

iati

on)

Em

plo

ymen

t im

pac

t(m

illi

ons)

Months elapsed since the policy uncertainty shock

Restoring pre-crisis levels

of policyuncertainty couldboost productionand employment

significantly

levels of policy uncertainty could increaseindustrial production by 4% andemployment by 2.5 million jobs over 18 months (see Figure 3). That would not be enough to create a boomingeconomy, but it would be a big step in the right direction.


24

On 7 September 2011, a letter from 20 eminenteconomists was published by the Financial Times urgingthe government to scrap the 50% marginal tax rate onannual incomes over £150,000 per year introduced inApril 2010. The days that followed saw a flurry of lettersboth agreeing and disagreeing with this view. I myselfcontributed to it by having a letter published in the FT thefollowing day.

I do believe it is important for social scientists to come outof their ivory towers and seek to influence debate aboutimportant issues of public policy. And because the press isone place where these debates are played out – and theletters pages the most readily available way to get one’sideas some publicity – it is perhaps natural to findeconomists writing letters to the newspapers.

But at the same time I have a lot of reservations aboutwhether debates conducted through letter pages are really

productive. That might seem oddcoming from someone who hastwice in recent years written

letters to the papers.But rather like mychildren, I am

inclined to say‘they started it’ and

that my letters wereresponses to others. I also

believe that it is consistent bothto pen a letter in response to

others and to thinkwe might all havebeen better off notwriting them at all.

Let me explainthe problem.

There is an incentive to write ‘group’ letters to thenewspapers as they get attention. The 50p letter to the FTwas reported in most other UK newspapers, thesubsequent letters expressing disagreement were not.

So if you want to influence policy in some way, thenthere is an incentive for making a pre-emptive strike.And you don’t have to worry too much aboutthe content of the letter: the authority of theseletters is not really in their content but thesignatories to them. This is probably just as wellas the list of signatories is often much longerthan the body of the letter.

Although it always helps to have someone high profileon board (DeAnne Julius played that role in the 50pletter), most of the signatories are probably unknowneven to the highly educated readership of the FT. Sorryto disappoint those of my colleagues who think beinga famous economist is the same as being famous.

The best-known example of a group letter is the group of364 economists who wrote to The Times in 1981expressing their opposition to the policies of the Thatchergovernment – as if having one economist opposed foralmost every day of a (non-leap) year made the content ofthe letter more compelling.

Given that the debate conducted on the letters pagesoften seems to come down to ‘my list is longer than yourlist’, it is not surprising that the general public end upwith a rather low opinion of economists. In the case ofthe 50p letter, it was less than a week before the FTpublished a letter asking ‘can any economist demonstratetheir benefit… without another disputing it?’ What startsoff as a pre-emptive strike by one side ends up in mutuallyassured destruction for economists on all sides of thepolicy debate.

The letters pageLast autumn, the Financial Times published a series of letters fromeconomists about the pros and cons of scrapping the 50p tax ratefor high earners. Alan Manning questions the value ofresearchers conducting such debates about economic policythrough the pages of newspapers.

in brief...

Academic debates conductedthrough newspaper letters pagesare rarely productive

Some of this reaction is a bit silly. Almost all contentiouspolicy issues are complex with arguments for and against,rarely with all evidence pointing in one direction. Themaximum length of a letter to the FT is not conducive to anuanced discussion of the issues. But the way economistsproceed also asks for such a reaction.

Newspapers will continue to want to publish such letters,so the only way to improve matters is for some degree ofprofessional self-regulation. I would like public debate tobe based on the issues not on appeals to authority – solet’s see an end to group letters. Reducing the number ofsignatories would also free up column inches for thecontent of the letters. And let’s make sure that we providethe workings behind our reasoning – not just ourconclusions but why we have arrived at those conclusions.

My main issue with the initial letter on the 50p tax ratewas that it simply consisted of an assertion, and it madeno attempt to provide any serious evidence for its claims.It set a bad example for how serious economists go abouttheir business.

What about the substance of the issue: should the UKretain or abolish the 50p tax rate on the top 1% ofincome earners? Here, I am not going to answer thatquestion because the appropriate way for a professionaleconomist to proceed is to wait until there is some

evidence on its impact. I think it is fairly clear what themain issues to look at will be: does the 50p tax rate act asa sizeable disincentive to work; does it encourage highearners to live elsewhere; and does it mean that highearners find ways to evade or avoid paying higher taxes?

We do have existing, though sometimes conflicting,evidence on related issues in other countries and othertimes. But we are going to have to wait for someevidence on the impact of the 50p tax rate and we shouldnot pass judgment until that is in – which, according torecent newspaper reports, may be soon. The evidencemay well be ambiguous but that is the best we will have.

Do I expect economists to adopt my self-denyingordinance with regard to group letters? I doubt it – self-regulation rarely works. So I fear I will find myselfcontinuing to scan the letters pages of the FT ready toturn to my laptop to tap out another nail in the coffin thatcontains the reputation of economists.


25

Alan Manning is professor of economics at

LSE and director of CEP’s research

programme on communities.

Economists have anobligation to provide serious

evidence for their claims

EXPORT


26

How do firms break into foreignmarkets? Trade theory tends toemphasise the substantial ‘sunk

costs’ that they have to incur to startexporting. The implication is that only themost efficient firms can afford to export.

Yet recent empiricalresearch drawing oncustoms data from severalcountries has uncoveredpatterns of foreign entrythat seem difficult toreconcile with high sunkcosts. Many domestic firms

enter foreign markets every year; theyoften start selling small quantities to asingle neighbouring country; and almosthalf of them stop exporting within a year.At the same time, new exporters thatsurvive the first year tend to expandexports to their initial markets and alsomove into other markets.

If entry is so costly, how can weexplain so many firms starting exportactivity with so few initial sales and suchlow survival rates? And what could explainthe seemingly sequential entry pattern ofthe surviving exporters?

Many countries are looking to their exportsectors as a source of future growth, but how dodomestic companies make a success of sellingtheir output abroad? Research by EmanuelOrnelas and colleagues finds evidence of‘sequential exporting’ – firms experimenting innearby foreign markets before seeking tobecome big exporters.

Sequential exporting:how firms break into foreign markets


27

and distant countries for domestic firms.When a nearby country lowers its

trade barriers, it attracts exports frompreviously purely domestic firms. As thesenew exporters learn about their ability toserve foreign markets, some fail and giveup exporting, whereas others are verysuccessful and decide to expand into otherforeign destinations. As a result, tradeliberalisation in the nearby countrypromotes entry not only there but also indistant non-liberalising countries, albeitwith a lag.

Similarly, the reduction of tradebarriers in a distant country, by raising thevalue of profitably exporting there, alsoenhances the value of exportexperimentation in nearby markets,spurring entry into the latter even in theshort run. Once some of the entrantsrealise a high export potential from theirexperience in the neighbours’ markets,they move on to the market of theliberalising distant country.

Our research conjectures that a centralforce behind this behaviour is theuncertainty that firms face about theirown ability to export profitably. Becausethe drivers of export success are differentfor each firm (they are ‘firm-specific’), theycan only be uncovered when a firmactually starts exporting. But because theyare uncertain and entry is costly, newexporters tend to start small to avoidadding negative variable profits to thepotentially lost sunk costs.

At the same time, because what drivesexport success for an individual firm tendsto have ‘global scope’ beyond the firstmarket it enters, substantial entry costslead new exporters typically to enter asingle destination first and then to developtheir export potential. If their performanceis good in the first market, firms willgradually expand there and pursue‘sequential exporting’ to otherdestinations.

Researchers in international businesshave long recognised that exportprofitability is uncertain and firm-specific.For example, an early study shows howthe distinct knowledge and competenciesassociated with export success (which aretypically related to product adaptation,marketing and distribution) are onlyacquired by firms once they start their foreign operations (Johanson andVahlne, 1977).

Previous research has also illustratedhow a company with global scope canapply knowledge from its initial foreignoperations to new export destinations.Analysing firms in four emerging exportsectors in Argentina, a recent studydocuments how such export-specificknowledge can be used when accessingdifferent foreign markets (Artopoulos et al, 2011).

Similar reasoning applies to firm-specific characteristics of demand. Forexample, trade facilitation agencies, suchas SITPRO in the UK, stress the importanceof uncovering foreign demand for would-be exporters, indicating that the keyuncertainty is about persistentcomponents of demand, some of whichmay be present in multiple countries.

Developing our conjecture theoreticallyprovides a number of novel implicationsfor the dynamic behaviour patterns ofexporting firms, which we test empirically.Using firm-level data on all Argentinemanufacturing exports between 2002 and

2007, we find strong evidence that firms’first foreign destination plays a crucial rolein explaining future patterns of foreignentry. It is in that first market where firmslearn the most.

Specifically, as long as a firm continuesto be an exporter, its growth on entry (atboth the intensive margin – sales in themarket – and the extensive margin – thenumber of markets served) is significantlyhigher in its first foreign market than inmarkets it enters subsequently. Theoutcome is similar for exit: a firm is morelikely to stop right after entering its firstforeign destination than it is to leavemarkets entered subsequently.

But if ‘export experimentation’ isindeed key, the differential effect of thefirst market should not apply universally toall exporters. For example, if the firm wereto start exporting again after a break,there would no longer be a fundamentaluncertainty to be uncovered.

Similarly, if a firm starts exporting byserving multiple markets, it must bebecause it is relatively confident about itsexport success – so on average the role ofself-discovery should not be aspronounced for such firms as it is forsingle-market entrants. Uncertainty aboutexport profitability should also be lessmarked for producers of homogeneousgoods, for which global reference pricesare available.

In turn, our theory about sequentialexporting suggests that we should observerapid first-market export growth, earlyentry into additional markets and frequentearly first-market exit primarily amongfirst-time, single-market exporters ofdifferentiated products. This is indeedwhat we find empirically.

Hence, while firm-specific uncertaintyis but one possible force shaping firms'export strategies, our evidence indicatesthat it plays an unequivocal role inexplaining sequential exporting. Noticethat our mechanism does not deny thepossibility of a firm’s productivity (andother characteristics) also shaping itsexport behaviour. Even if a firm becomesmore efficient, which will raise the appealof exporting, uncertainty about potentialprofitability in foreign markets could stilltrigger sequential exporting.

The policy implications of a process ofsequential exporting driven by self-discovery are far-reaching. Consider theimpact of trade liberalisation in nearby

Uncertaintyabout their

potentialprofitability

in foreignmarkets leads

firms to pursue‘sequentialexporting’

EXPORT


28

Thus, our findings suggest theexistence of a ‘trade externality’: lowertrade barriers in a country induce the entryof foreign firms into other markets. Forexample, trade liberalisation in a distantbut large country A (say, China) caninduce firms from country B (say, the UK)to start exporting to nearby country C (say,Germany). This possibility could provide anovel motive for international coordinationof trade policies, one that strengthens therationale for institutions like the WorldTrade Organization (WTO).

If the trade externality is stronger atthe regional level, this possibility could also help to explain the pattern of freetrade agreements throughout the world. Indeed, the impact of trade agreementscould be very distinct from what existingstudies indicate.

For example, a regional tradeagreement can boost exportexperimentation by lowering the costs ofaccessing the markets of bloc partners. Asa result of more experimentation, a greaternumber of domestic firms would eventuallyfind it profitable to export to countriesoutside the bloc. This would generate‘trade creation’ that is very different fromthe concept that economists oftenemphasise: in addition to promoting intra-bloc trade, a regional trading bloc can alsostimulate exports to non-members.

If the agreement were of themultilateral type, tracking down its effectsbecomes even trickier. Indeed, third-country and lagged effects of tradeliberalisation may help toexplain the difficulty inidentifying significant tradeeffects of multilateralliberalisation, thuscorroborating well-entrenched beliefs that theWTO (and its predecessor,the General Agreementon Tariffs and Trade)have been crucial inpromoting world trade.

This article summarises ‘Sequential

Exporting’ by Facundo Albornoz, Hector

Calvo Pardo, Gregory Corcos and Emanuel

Ornelas, CEP Discussion Paper No. 974

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

dp0974.pdf).

Facundo Albornoz is at the University of

Birmingham. Hector Calvo-Pardo is at the

University of Southampton. Gregory Corcos

is at the Norwegian School of Economics and

Business Administration. Emanuel Ornelas

is director of CEP’s research programme on

globalisation.

Further reading

Alejandro Artopoulos, Daniel Friel and Juan

Carlos Hallak (2011) ‘Lifting the Domestic

Veil: The Challenges of Exporting

Differentiated Goods Across the Development

Divide’, National Bureau of Economic

Research Working Paper No. 16947

Jan Johanson and Jan-Erik Vahlne (1977)

‘The Internationalization Process of the

Firm – A Model of Knowledge Development

and Increasing Foreign Market

Commitments’, Journal of International

Business Studies 8: 23-32

The process ofsequentialexporting

suggests thatlower trade

barriers in onecountry can

induce the entryof foreign firms

into othermarkets

Trade liberalisation in distant butlarge country A (say, China) mayinduce firms from country B (say,the UK) to start exporting tonearby country C (say, Germany)

PUBLICATIONSCEP Discussion Papers are available as electronic copies free to download from the Centre’s website:http://cep.lse.ac.uk/_new/publications

THE DETERMINANTS OF INTRAFIRMTRADE: EVIDENCE FROM FRENCH FIRMSGregory Corcos, Delphine Irac,Giordano Mion and Thierry VerdierCEP Discussion Paper No. 1119January 2012

INSTITUTIONS AND EXPORT DYNAMICSLuis Araujo, Giordano Mion andEmanuel OrnelasCEP Discussion Paper No. 1118January 2012

PREFERENTIAL TRADE AGREEMENTS AND THE LABOUR MARKETEmanuel OrnelasCEP Discussion Paper No. 1117January 2012

THE PROVISION OF RELATIVEPERFORMANCE FEEDBACKINFORMATION: AN EXPERIMENTALANALYSIS OF PERFORMANCE AND HAPPINESSGhazala Azmat and Nagore IriberriCEP Discussion Paper No. 1116January 2012

FISCAL MULTIPLIERS OVER THE BUSINESS CYCLEPascal MichaillatCEP Discussion Paper No. 1115January 2012

GRAVITY REDUX: MEASURING INTERNATIONAL TRADECOSTS WITH PANEL DATADennis NovyCEP Discussion Paper No. 1114January 2012

THE CAUSAL EFFECTS OF AN INDUSTRIAL POLICYChiara Criscuolo, Ralf Martin, HenryOverman and John Van ReenenCEP Discussion Paper No. 1113January 2012

PAYING FOR PERFORMANCE: INCENTIVE PAY SCHEMES ANDEMPLOYEES’ FINANCIAL PARTICIPATIONAlex Bryson, Richard Freeman, ClaudioLucifora, Michele Pellizzari andVirginie Perotin CEP Discussion Paper No. 1112January 2012

LEFT, RIGHT, LEFT: INCOME AND POLITICAL DYNAMICS INTRANSITION ECONOMIESMichael Carter and John MorrowCEP Discussion Paper No. 1111January 2012

CEP POLICYANALYSISAPPRENTICESHIP POLICY IN ENGLAND:INCREASING SKILLS VERSUS BOOSTINGYOUNG PEOPLE’S JOB PROSPECTSHilary SteedmanCEP Policy Analysis No. 13December 2011http://cep.lse.ac.uk/pubs/download/pa013.pdf

For further information on CEP publications and events please contact:

The Publications Unit, Centre for Economic Performance, Houghton Street, London WC2A 2AE

Telephone +44 (0) 207 955 7673

Fax +44 (0) 207 955 7595

Email [email protected]

We accept Visa, Access and Mastercard for payment by post, fax and phone.

Overseas rates on request.

CEP EVENTS

LIONEL ROBBINS MEMORIALLECTURES 2012CLIMATE CHANGE AND THE NEW INDUSTRIAL REVOLUTIONSpeaker: Professor Lord Nicholas Stern, London School of Economics

Dates: Monday 21, Tuesday 22 and Wednesday 23 February 2012

Time: 6.30pm

Venue: Old Theatre, Ground Floor, Old Building, London School of Economics, Houghton Street, London WC2A 2AE

This event is free and open to all with no ticket required. Entry is on a first come, first served basis. For any queries email [email protected] or call +44 (0) 207 955 6043.

http://cep.lse.ac.uk/_new/events/event.asp?id=140

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

© Centre for Economic Performance 2012

Volume 16 Issue 3

(ISSN 1362-3761) All rights reserved.

Centre Piece

The back issues

Follow CEP on Twitter @CEP_LSE

centrepiece - cep.lse.ac.ukcep.lse.ac.uk/pubs/download/centrepiece_16_3.pdf · speed rail line....

Documents