cities unlimited

Cities UnlimitedMaking urban regeneration work

Tim Leunig and James Swaffield

Edited by Oliver Marc Hartwich

Policy Exchange is an independent think tank whose mission is to develop and promote new policy ideas which willfoster a free society based on strong communities, personal freedom, limited government, national self-confidence andan enterprise culture. Registered charity no: 1096300.

Policy Exchange is committed to an evidence-based approach to policy development. We work in partnership with aca-demics and other experts and commission major studies involving thorough empirical research of alternative policy out-comes. We believe that the policy experience of other countries offers important lessons for government in the UK. Wealso believe that government has much to learn from business and the voluntary sector.

Trustees

Charles Moore (Chairman of the Board), Theodore Agnew, Richard Briance, Camilla Cavendish, Richard Ehrman,Robin Edwards, George Robinson, Tim Steel, Alice Thomson, Rachel Whetstone.

About the authors

Dr Tim Leunig is lecturer in economichistory at the London School ofEconomics. His research looks at Britainsince 1870, including a number of papersthat look at different aspects of city lifeover the ages. His work has won a numberof international awards, and was recentlyjudged to be ‘outstanding’ by theEconomic and Social Research Council.He has advised Parliament, the Treasuryand the Department for Transport.

James Swaffield is a Research Fellow atPolicy Exchange carrying out research oncities and urban policy. He studiedGeography at Oxford University. Aftergraduating in 2004 he spent a brief periodworking in the public sector before com-pleting an MSc in Cities, Space, andSociety at the London School ofEconomics in September 2006.

Dr Oliver Marc Hartwich is ChiefEconomist at Policy Exchange with respon-sibility for economic competitiveness. Hewas born in 1975 and studied BusinessAdministration and Economics at BochumUniversity (Germany). After graduatingwith a Master¹s Degree, he completed aPhD in Law at the universities of Bochumand Sydney (Australia) while working as aResearcher at the Institute of CommercialLaw of Bonn University (Germany). Havingpublished his award winning thesis withHerbert Utz Verlag (Munich) in March2004, he moved to London to support LordMatthew Oakeshott of Seagrove Bay duringthe process of the Pensions Bill.

The authors have previously workedtogether for Policy Exchange on Cities lim-ited, a report on urban policy in the UKand Success and the City.

2

© Policy Exchange 2007

Published byPolicy Exchange, Clutha House, 10 Storey’s Gate, London SW1P 3AYwwwwww..ppoolliiccyyeexxcchhaannggee..oorrgg..uukk

ISBN: 978-1-906097-28-8

Printed by Heron, Dawson and SawyerDesigned by John Schwartz, [email protected]

Contents

Acknowledgements 4Executive Summary 5

1 Introduction 62 Cities Limited 83 Success and the city 114 Reactions to our work 145 Things will only get worse 176 We cannot accept this 217 What can we do? 228 Expanding London 309 It will not just be London 3710 Economic geography outside of the South East 4111 Setting cities free 4512 What would cities do? 4813 Coping with population decline 5314 Improving governance 5715 Conclusion 60

www.policyexchange.org.uk • 3

Acknowledgements

Policy Exchange would like to thankNatalie Evans, Sam Freedman, PhilippaIngram, Max Nathan and Nick Tiratsoo.

ir

4

Executive Summary

Cities limited, the first of three reports,demonstrated that attempts to regenerateBritish cities over the past ten, twenty oreven fifty years have failed. The gapbetween struggling and average cities, letalone between struggling and affluentcities, has continued to grow. Geographicalinequality is growing. Our second report,Success and the City, examined experienceabroad and the lessons other countries canteach British policymakers. This finalreport, Cities unlimited, uses the evidencethat we have gathered so far about what ispossible and what is not, about what worksand what does not, to offer new policy pro-posals for regenerating Britain’s cities.

Many of the forces that make life toughfor struggling cities will continue. Asdemand for more highly qualified workersgrows, the lower skill levels associated withregeneration towns will make it even hard-er for them to catch up – not least becausetheir brightest and best educated leave forLondon after graduation. Nor is a changeof government likely to continue support-ing regeneration policy. Ministers in thecurrent Labour Cabinet overwhelminglyrepresent inner city areas. A futureCabinet, perhaps more representative ofsuburbs and the wealthy South East, maynot have the same commitment to highlevels of regeneration funding, particularlyif economic circumstances demand asqueeze on public spending.

But if we are honest about the con-straints and realistic about the opportuni-ties then we can make progress. We need toaccept above all that we cannot guaranteeto regenerate every town and every city inBritain that has fallen behind. Just as wecan't buck the market, so we can’t buckeconomic geography either. Places thatenjoyed the conditions for creating wealthin the coal-powered 19th-century often donot do so today. Port cities had an advantage

in an era when exporting manufacturedgoods by sea was a vital source of prosper-ity; today the sea is a barrier to their poten-tial for expansion and they are cut off fromthe main road transport routes. More gen-erally, the economic pull of Europe hasboosted the South East at the expense ofthe North, Wales and Scotland. Luck hasalso played its part: in 1900 London hadfinance and Manchester had cotton.Finance has since prospered and cottoncollapsed, reinforcing geographicalchanges.

There is no realistic prospect that ourregeneration towns and cities can convergewith London and the South East. There is,however, a very real prospect of encourag-ing significant numbers of people to movefrom those towns to London and theSouth East. We know that the capital andits region are economic powerhouses thatcan grow and create new high-skilled,high-wage service sector hubs. At the sametime market mechanisms can be used toinduce some firms to move out of theSouth East.

We propose a significant liberalisationof land use in London and the SouthEast. At present local councils ignore mar-ket signals and zone land for industrialrather than residential use. There are over2,500 hectares of industrial land inLondon alone, and 10,000 hectares inLondon and the South East together. Ifonly half of it were used for housing, itwould create £25 billion in value andallow half a million people to move to anarea that offers much better prospectsthan where they live now. Such a market-led policy would prompt many industrialfirms that are based in London to relocateto where land is cheaper. So some peoplewill leave regeneration cities; some jobswill move to regeneration cities. Both aredesirable outcomes.


We should go further. No one likesurban sprawl per se, but suburbs offer ahigh quality of life to those who live inthem and create the economic “bulk” thatmakes cities both prosperous and vibrant.Increasing the size of London so that ittakes an extra minute’s journey to reach theedge of town would make room for anoth-er million people, increasing their oppor-tunities.

We know that the South East offers thegreatest agglomeration potential, so itmakes sense to think seriously about hav-ing more than one large city in the SouthEast, where London has dominated for solong. We also know that cities based onhighly skilled workers are the most dynam-ic. Oxford and Cambridge are unambigu-ously Britain’s leading research universitiesoutside London and both are well locatedeconomically. We should consider expand-ing both dramatically, just as Liverpool andManchester expanded in the 19th century.Dynamic economies require dynamic eco-nomic geography.

Of course there are successful high-techclusters outside the South East and it isright that they should build on theirstrengths. But in many cases that does notimply huge expansion. Manchester, Leedsand Newcastle, for example, all have decid-ed strengths, but they are not successfulenough to deliver prosperity to neighbour-ing towns, such as Rochdale, Bradford andSunderland, in the way that London is ableto support relatively poor communities inNorth Kent and South Essex.

It is important to recognise the con-straints that face many communities. Theirresidents are certainly aware of them: ourregeneration towns’ share of the popula-tion continues to fall as people get on theirbikes and move to places that offer betterprospects. Liverpool’s population is littleover half its peak level.

There is no doubt that Britain starts at adisadvantage compared with, say,Germany or the Netherlands because its

cities of the industrial revolution are geo-graphically more peripheral than theirs.The relative weakness of local governmenthere has compounded this disadvantage;our studies of cities around the world – inGermany and the Netherlands, in Poland,Canada and Hong Kong – demonstratethat local communities manage theiraffairs better than a distant central govern-ment can ever do.

Devolution has many advantages. Itleads to diversity, and diversity creates evi-dence as to what works and what does not.Anyone who believes in evidence-basedpolicymaking should support large-scaledevolution. We propose that theGovernment should roll up current regen-eration funding streams and allocate themoney to local authorities according to asimple formula based on the inverse oftheir average income levels. Central fund-ing would be available for a handful ofexceptional circumstances, such as decont-aminating highly polluted land.

It would be for local authorities to assessthe opportunities, devise a plan for theirarea and implement it. They would beanswerable not to central government, butto local people. They could spend themoney on traditional regeneration ideas,such as supporting local firms or attractinginward investment. They could use themoney to revitalise town centres or provideinfrastructure links. Or they could pursuea more people-oriented strategy andfinance job clubs or skills training courses.

If a local authority believes that attract-ing entrepreneurs is the right approach,then it would be perfectly legitimate tospend money providing facilities to do so.All local authorities would have the rightto use the money to create a unique sellingpoint – the best parks in Britain, the mosttrees – to get away from being an identikittown. A council could decide to attractparents who care about education byspending its money on schools. It may wellbe the case that doubling teachers’ pay and

Cities unlimited

6

cutting class sizes substantially wouldtransform schools and attract new people.Affluent Manchester workers, for example,might be attracted to buy homes in thecatchment areas of such schools inRochdale. Their demand for local goodsand services would create jobs. Finally, anarea that decided that there was no realisticchance of regeneration could use themoney to help local people to find workelsewhere and to cut taxes for those whoremained.

Many of our regeneration towns areshrinking. Demolition policy is not work-ing: it is expensive and is failing to trans-form local housing markets. We proposethat, wherever possible, we should stopdemolishing properties. Instead, theGovernment should buy up the cheapesthouses on the open market and give themto the next door neighbour in exchangefor an equity stake in the expanded prop-erty. Houses in regeneration towns tend

to be relatively small, so a scheme thatcuts costs for government – it would nothave to bear the costs of demolition orpay for compulsory purchase – andincreases the average size of a house hasobvious appeal.

Finally, if we are to devolve funding tolocal authorities we need to ensure thatthey are accountable to local people. Thatmeans much better scrutiny by the AuditCommission and local media and greaterrights for local people to investigate whatthe town hall is doing. It also means ensur-ing that the voting system makes electionstruly contestable, so that all politicians,whether in office or opposition, know thatvoters can – and will – hold them toaccount at the ballot box.

Realism, mobility and diversity.Freedom, responsibility and accountability.

These are the keys to unlocking thepotential that exists in our towns andcities.


Executive summary

1Introduction

Cities limited set out a critique of urbanregeneration policy over the last tenyears. It did not look at individual pro-grammes, but rather asked the most fun-damental question of all: have the placesthat were the focus of regeneration policybeen regenerated? We did not claim, anddo not claim, that these towns and citieshave not improved since 1997. A decadeof strong economic growth means that itis almost impossible to find any placethat has not improved. Nor did we claimthat the regeneration money that hasbeen spent has achieved nothing. It isalmost impossible to spend billions andbillions, year in and year out, and achievenothing at all. But we did claim – and noone has disputed this – that, far fromcatching up, the places we focused on,towns and cities that experienced waveafter wave of regeneration initiatives,have fallen further behind the nationalaverage. In contrast, towns that were suc-cessful in 1997, and not the subject ofregeneration policies, have pulled furtherahead. The gap between successful andunsuccessful towns and cities haswidened despite a decade of ambitiousregeneration policies. We argued, there-fore, that regeneration policy over the lastdecade has failed.

Success and the city: Learning from inter-national urban policies examined cities andcity regions in five countries. All have dealtwith challenges that were in some wayssimilar to those faced by British cities andcity regions. We looked at the Ruhr area,which faces very similar challenges to thoseof many former industrial regions in

Britain. We looked at Poland, facing up toa complete change in the nature and orien-tation of the economy. We looked atAmsterdam, and the Randstad more gener-ally, an area coping with change andrenewal in the context of a successful econ-omy. And we looked at two classic regener-ation cities, Vancouver and Hong Kong,where strong local government leadershipis taken for granted, and where the locali-ty, not the nation state, determines the pri-orities, makes the decisions and takesresponsibility. These cities showed us thatincentives, power and ability matter. Citiesare more likely to be successful when cityleaders are given responsibility, but respon-sibility alone is not sufficient. As ever, thefoundation of successful policymaking andimplementation requires ability, as well asresponsibility.

Cities Unlimited sets out our vision ofhow Britain should approach regeneration.It takes seriously the lessons of economicgeography, of Britain’s experience and ofexperience elsewhere. It accepts that noone-size-fits-all policy can ever hope tocome to terms with the myriad of differentcircumstances faced by different commu-nities in Britain, particularly if it has beenformulated at the centre and imposed onour towns and cities. The radical policiesthat we set out here – including acceptingsignificant population movements withinBritain and a substantial reduction in thepower of Whitehall and Westminster, cou-pled with corresponding increases in powerfor local councils and, therefore, localaccountability – may not make for com-fortable reading.

8

Many of Britain’s towns and cities havefailed – and been failed by policymakers– for too long. It is better to tell uncom-fortable truths than to continue to claimthat if we carry on as we are then thingswill turn out well. The people who pay

the price of this failure are those born intowns that offer fewer opportunities thanthose enjoyed by people lucky enough tobe born elsewhere. That cannot beacceptable: we need to change ourapproach.


Introduction

2Cities Limited

Cities limited, the first in this series of threepublications on urban policy, examinedthe combined performance of 18 towns,each of which had received significant lev-els of regeneration funding. By aggregatingperformance for a group of places weavoided exogenous effects, such as the clos-ing of a local military base, from dominat-ing our results. The places that we studiedranged from Glasgow in the North toSouthampton in the South, from Merthyr

Tydfil in the West to Hull in the East.Some of the towns were small – Hastingsand Merthyr have fewer than 100,000 peo-ple, while others, such as Sheffield andGlasgow, have more than five times thatnumber.

It would have been possible to have useda different sample of towns. We could haveincluded Grimsby and Middlesbrough,Accrington and Bacup (or just about anyof the 19th-century Lancashire mill

10

1 These towns are respectively

23, 53, 83 and 81 miles from

London

Sheffield Liverpool

Blackburn

Hull

Wigan

Southampton

Coventry

Sunderland

Bradford

Stoke

Leicester

Hastings

Blackpool

Stockton

Warrington

Walsall

Merthyr Tydfil

Glasgow

Figure 1: Geographical location of sample

towns), Portsmouth and Margate. We donot believe that the results would havebeen materially different. Nor do weaccept the criticism made by some that wepicked failing towns and showed that theyhave failed. All our towns received signifi-cant levels of regeneration funding and, assuch, are legitimate cases to include in astudy of the effects of regeneration policy.Those who argue that we simply picked onfailing towns at very least need to explainwhy there are 18 cities for which regenera-tion funding was so ineffective.

In addition, we constructed a smallersample of successful towns. We were care-ful not to include the runaway successes:not everywhere can be the City of London,Chelsea, Oxford or Cambridge. Instead wepicked one affluent London commutertown, Windsor, three more distant townsthat are at least semi-independent ofLondon’s orbit, Milton Keynes,Peterborough and Swindon2 and two inde-pendent, thriving cities, Bristol andEdinburgh.

We judged success by a range of criteria.First, we looked at gross value added(GVA), the local equivalent of grossdomestic income (GDP). Data from theOffice of National Statistics shows thataggregate GVA scores for the 18 citieswhich were subject to regeneration initia-tives fell from 90 per cent of the nationalaverage to 86 per cent between 1997 and2005, continuing an established trend. Incontrast our successful cities’ GVA roseover the same time from 131 per cent ofthe national average to 143 per cent.

Secondly, we used Revenue & Customsdata to look at personal incomes. The storywas fairly similar to that for GVA, but notas pronounced. The personal incomes ofour regeneration group continued to sliprelative to the country as a whole, and par-ticularly relative to our sample of success-ful cities.

We looked at unemployment, which hasfallen across Britain in the last decade, but

remains 40 per cent higher in our regener-ation sample. Since unemployment levelshave remained fairly stable over the last fiveyears, this suggests that they are structural-ly higher in our regeneration towns thanelsewhere in the country. Some, but notall, of this variation in unemployment canbe attributed to skill levels: Leunig andOverman report that individuals withoutGCSE level skills have a one in five chanceof being unemployed if they live inWindsor, but a one in three chance if theylive in Blackburn.2

Our fourth indicator is perhaps themost important: where do people want tolive? This surely is the bottom line. It is theleast technocratic, the least academic, themost outcome focused, because it askswhat are individuals choosing to do withtheir lives. Remember that housing inthese regeneration towns is generallymuch, much cheaper than housing inBritain as a whole. If they really were beingregenerated we would expect people to bemoving to them to take advantage of thisopportunity. Equally, it is costly for thosewho live there already to move out. If theyare owner occupiers they will have to tradedown or increase their mortgage outgo-ings; if they are renters then they will paymore rent. If regeneration were really hap-pening, these people would not leave.3

Far from people moving to these areas,we find that the proportion of the popula-tion living in our regeneration cities hasfallen by about 5 per cent over the lastdecade. These are not places in the processof regeneration and which offer brightfutures to would-be migrants. On the con-trary, these are towns which people are


Cities Limited

2 Leunig T and Overman H,

“Spatial Patterns of

Development and the British

Housing Market”, Oxford Review

of Economic Policy, Spring 2008

3 In contrast social housing ten-

ants are, as is well known,

almost unable to move between

regions and very few do. We can

learn little from looking at the

mobility of social housing ten-

ants. Hills J, Ends and Means –

The future roles of social hous-

ing in England, CASE report 34,

Centre for the Analysis of Social

Exclusion, 2007

“ People on the ground know the reality of success and

failure better than any policymaker or analyst can ever

hope to do ”

leaving, notwithstanding the higher cost ofliving elsewhere. People on the groundknow the reality of success and failure bet-ter than any policymaker or analyst canever hope to do, and they are voting withtheir feet. We do not know how many peo-ple would leave these regeneration townswere they not constrained by differentialhousing costs or the geographical restric-tions imposed by social housing tenancies.

Our results should not have come as asurprise to politicians or the policy com-munity. Although our research wasmethodologically different to other stud-ies, the results were very similar. TheJoseph Rowntree Foundation commis-sioned the University of Sheffield to studythe economic geography of Britain. Theirfocus was on a longer period, 1968 to2005, but their results were similar: thoseareas that were poor in 1968 were poor in2005. Regeneration has not happened.Like us, they found that more successfulareas were drawing further ahead: “Areasalready wealthy have tended to becomedisproportionately wealthier, and we areseeing some evidence of increasing polari-sation. In particular there are now areas insome of our cities where over half of allhouseholds are breadline poor.”4

The Centre for Cities, then at theInstitute for Public Policy Research(IPPR), looked at the last ten years, andcame up with much the same conclusions:many of Britain’s cities are not successes,and are not offering their residents theopportunities that they might legitimatelyexpect after ten years of solid economicgrowth.5 Analysis by Policy Exchange andIPPR comes to the same conclusionbecause the conclusion that regenerationpolicy has failed stems not from ideology,but rather is inescapable to anyone wholooks at the evidence.

Nor can the Government claim igno-rance, for its own commissioned workshows the same thing, time and again. AnOxford University and London School ofEconomics report in 2001 for the SocialExclusion Unit showed that many urbanareas had consistent and persistent prob-lems that showed no evidence of decliningover time.6 They found that every initiallydeprived place outside London recordedbelow average performance between 1995and 1998, the period covered in the report.

More recently the Department forCommunities and Local Government(DCLG) commissioned Oxford BrookesUniversity to assess English cities. Theteam looked at gross value added (GVA)between 1995 and 2002 and found thatrises in GVA in some towns – includingStoke and Blackburn from our sample –were under half the English average.7 Thereport also questioned the effectiveness ofregeneration funding, arguing that it was“open to question whether such policies asphysical urban renaissance and physicalregeneration will pay large dividends interms of improved competitiveness.”8

These results paint a gloomy picture, inwhich regeneration for many cities existsonly in the eyes and speeches of politicians.As we made clear, there is no doubt thatLabour was genuinely committed to regen-eration – these are Labour heartlands afterall, the constituencies of many LabourCabinet ministers. Nor is it the case thatLabour has failed where the Conservativeshad succeeded: the trends that we foundbegan before 1997. We did not identify awell-intentioned but ultimately unsuccess-ful policy change in 1997 that simplyneeds to be reversed. Instead the challengeis harder, and that is why we next lookedaround the world to see how thingsworked there.

Cities unlimited

12

4 Dorling D, Rigby J, Wheeler B,

Ballas D, Thomas B, Fahmy E,

Gordon D and Lupton R,

Poverty, Wealth and Place in

Britain, 1968 to 2005, p xiv,

Joseph Rowntree Foundation,

2007

5 Athey G, Lucci P and Webber

C, Two-Track Cities: The chal-

lenge of sustaining growth and

building opportunity, Discussion

Paper no 11, Centre for

Cities/IPPR, London, July 2007

6 Changing Fortunes:

Geographic patterns of income

deprivation in the late 1990s,

Renewal Research Summary 41,

Social Exclusion Unit /

Department for Transport, Local

Government and the Regions,

London, 2001

7 Department for Communities

and Local Government, State of

the English Cities: The competi-

tive economic performance of

English Cities, London, 2006

8 Ibid, p 224

3Success and the city

Because politicians and civil servants gen-erally work only in their own country,knowledge from other nations does notcross borders at either the political oradministrative level as a matter of course.Domestic policymakers need to make realefforts to learn from other countries’ suc-cesses and failures. International compar-isons are a core part of Policy Exchange’sevidence-based approach to policy.

Success and the city was the result of 50interviews conducted across three conti-nents and five nations. We investigated thesuccess, or otherwise, of Vancouver; theDutch Randstad, particularly Amsterdam;the German Ruhr region; the Polish cities ofWarsaw and Łódź; and finally Hong Kong.

The two most successful cities that wevisited, Vancouver and Hong Kong, werealso the largest. We do not believe that thisis a coincidence. Large cities frequentlyserve as a magnet for high-skilled workers,people capable of driving an economy for-ward, creating and sustaining firms thatoffer good quality, long lasting jobs. Weneed to understand that for cities, econom-ically at least, big is beautiful, an issue wereturn to in the next chapter.

Big can also be politically advantageous.A city such as Vancouver is simply toolarge to fail. Governments can get awaywith neglecting smaller places if they sochoose, but they cannot get away withignoring the largest places. Throughouthistory riots and rebellions have over-whelmingly occurred in the largest cities.The concentration of poverty and derelic-tion are politically volatile, and that is whygovernments take care to ensure that big

cities do not fail too badly. Whatever thelegal or constitutional position, a city’s sizegives it political clout, locally and national-ly. The mayor or council of a large city willhave a large budget. They can afford to hirebetter advisers – both on regeneration andon fighting their corner. The city will havemore representatives in the national parlia-ment than smaller places. With quantity,comes quality: the more members of par-liament a city has, the greater the chancethat one of them is a minister or a heavyhitter of another type. Perhaps it shouldnot be so, but that is the reality.

But we found more to regeneration thansize, or importance: Warsaw is Poland’slargest city yet not obviously successful.What we have found, time and again, isthat the degree of flexibility accorded tolocal areas matters. When areas have to dowhat they are told by bureaucrats hundredsor even thousands of miles away, regenera-tion is stifled. When they are given moreautonomy, regeneration appears easier. Theextreme case of this is Hong Kong, wherethe “one country, two systems” constitu-tional position guarantees Hong Kong adegree of autonomy that we believe to beunique. Hong Kong has faced many chal-lenges over the past two decades, includinguncertainty and fear of change, economicturmoil in the region, and the rise of localeconomic rivals such as Shanghai. Yetthrough all of that Hong Kong’sGovernment was able to act with consider-able autonomy, to be entrepreneurial.

If nowhere else has the same degree ofautonomy or independence as HongKong, nowhere is as centralised as Britain.


Germany gives considerable freedoms notonly to the states – protected under thenation’s federal constitution – but also toits cities. Places in the Ruhr, such as Essenand Bochum have the freedom to experi-ment and to adapt to local conditions.Essen, for example, has used that freedomto devise incentives to attract firms such asThyssenKrupp. A tradition of decen-tralised power structures is also strikinglyevident in the Netherlands and Canada.Thus, Amsterdam can design its owndevelopment and fund it, and Vancouvercan break with national policy on issuessuch as the Olympics.

But freedom does not guarantee success.The Polish cities of Warsaw and Łódź haveconsiderable freedom, perhaps as muchfreedom as Essen and Bochum inGermany. Yet they have had much less suc-cess in regeneration. Ability, as well asresponsibility, is necessary for success. Thatin turn requires a depth of knowledgeabout regeneration and development thatis not easy to create. Freedom, as ever, isnot sufficient: the national state must stepout of the way, but the local area also needsa high level of human capital to be able totake advantage of the freedoms that areoffered. Here Warsaw and Łódź suffer verybadly. Since the Second World War thesecities were part of a highly centralised sys-tem, in which local authorities were not

required, expected, or encouraged to thinkfor themselves. In contrast Germany’s fed-eral system expects and almost requireslocal areas to be different to each other, tolook for fresh ideas, to think. They havethe experience and the expertise thatWarsaw and Łódź lack.

The interplay between freedom and abil-ity is important. While freedom can releasebut not create ability, ability can create free-dom because it makes it easier for those incontrol to let go. Sir Michael Barber, thefounding head of the Prime Minister’sDelivery Unit, recently wrote: “Power is nota zero-sum – enhancing crucial relationshipsbuilds the capacity to get things done.”9

Here Amsterdam stands out. It co-operatedwith other Randstad cities, such as TheHague and Rotterdam, to create a regionalforce that could tackle issues such as region-al transport that no individual city could doon its own. Amsterdam is also notable forengaging with important communityactors, such as housing associations, in orderto ensure that all parties involved in regen-eration are pursuing compatible policies.Both strategies have given Amsterdam moredecision-making and financial power inreality than it appears to have on paper andboth have aided regeneration.

We therefore have two dimensions bywhich to judge a city’s ability to regenerate:whether it has the ability, and whether it

Cities unlimited

14

9 Barber M, “Delivery is all about

‘how?’ not ‘what?’”, Parliamentary

Brief, 11(9) pp18-20, 2008

Headless Chicken Free-range chicken

Warsaw, Łódź Ruhr, Hong Kong

Cages Hens Cockerel

UK Vancouver, Amsterdam

Fre

edo

m

HIGH LOW

Figure 1: Urban Policy Typologies – A Poultry Attempt

LOW HIGH

Ability

has the freedom to use those abilities. Theseare set out in Figure 1, which constructs afour-part matrix, each of which is charac-terised as a type of poultry. Cities andregions with both freedom and ability arerepresented as free-range chickens. Theyhave a good quality of life, and can copewith changes in their conditions as well asis possible, as both Hong Kong and theRuhr areas show. In contrast, ability with-out freedoms makes for cockerel cities. Suchplaces have good ideas and innovative poli-cy solutions, but lack the right to imple-ment them. Nevertheless, their ability tocrow loudly, repeatedly and at length cancreate freedoms, and those with power findit hard to resist. Cockerel cities such asAmsterdam and Vancouver, therefore, havemore freedoms in reality than they appearto have on paper, although those freedomsare conditional and can be reduced or cur-tailed by the centre.

Freedom without ability is a frustratingscenario, because that freedom cannot beused effectively. We characterise this, perhaps

harshly, as the headless chicken quadrant, inwhich an area tries its hardest, but withoutsufficient expertise it can seem to runaround in circles, with achievements com-ing about by chance. We place the Polishcities of Warsaw and Łódź in this quad-rant.

Finally, there are those towns withoutability or freedom, the caged hens of theregeneration world. These towns are entire-ly at the mercy of others. They do not havethe right to make their own policy, nor it isclear that they would be successful werethey to be given those rights. Knowledge oftheir own limitations leads to dependency,so that rather than calling for more deci-sion-making powers, rather than setting outtheir own agenda in the way of Amsterdamand Vancouver, they are content to be thehandmaidens of central government, wait-ing to be told what policies they shouldimplement. All too often Britain’s localcouncils fit into this category, although it iscertainly possible to argue that London isgrowing into cockerel status.


Success and the city

4Reactions to our work

Two ministers, Ian McCartney, MP forMakerfield and until September 2007Minister of State for Trade, Investment andForeign Affairs10 and John Healey, MP forWentworth and Minister of State forDepartment for Communities and LocalGovernment since June 2007, respondedto Cities limited. We mention their reac-tions here because we think they illustratethe sort of thinking that lies behind muchof current urban regeneration policy.

Ian McCartney made his position veryclear in an interview with Wigan Today on9th November 2007.11 He made the boldclaim that “This Government and thecouncil have rebuilt Wigan brick by brickover the past decade,” before going on toadd that “we certainly don’t need peoplebased 300 miles away running down thetown in this fashion.” He stated that theGovernment had invested £130 million inrefurbishing “every one” of Wigan’s coun-cil houses, that schools were being rebuiltand that local town centres were beingrefurbished.

Two aspects of this story stand out.First, McCartney did not challenge eitherthe arguments or the data put forward inCities limited. We did not claim that theGovernment had failed to invest inWigan’s council housing or that town cen-tres had not been refurbished – althoughwe suspect his claim that Wigan has beenrebuilt “brick by brick” to be an exaggera-tion. What we claimed was that despitevery high levels of investment towns suchas Wigan were falling behind both thenational average and the fortunes of suc-cessful towns such as Swindon and

Peterborough. Our claim is not that theGovernment has not spent enough, butrather that its spending has not deliveredthe outcomes we hoped for: good jobs, andplenty of them, so that the people ofWigan are able to enjoy the standard of liv-ing of the average person in Britain.

The second interesting aspect ofMcCartney’s intervention is the responseof people in Wigan. This story was postedon the website of his local paper, allowinglocal people the chance to respond. Sevenpeople did so, and not one supportedMcCartney’s position. A former firefighterwrote: “Perhaps Ian McCartney shouldread the report and then and only thenexplain why the report is wrong instead ofdoing the usual politician’s trick of makingoff-the-cuff statements and expecting peo-ple to believe him because he is an MP.”Mark Conroy described McCartney’s reac-tion as “Typical of New Labour” while R.Johnson wrote: “The report is right,” andan anonymous writer noted that “a lot ofthe younger generation in Wigan have nohope for the future.”

The most detailed comments came fromC. Rowley, “a Wiganer born and bred”who has lived in Hindley for the last 38years, and who wrote: “I was disappointedand concerned having read IanMcCartney’s comments.” Rowley offeredto take McCartney to Hindley’s MarketStreet, “with its charming charismatic graf-fiti, daubed and derelict high street build-ings, an area that has been in decline forthe past thirty years”. Then, he said,McCartney could “explain exactly where ithas been rebuilt brick by brick”. Rowley

16

10 Since Sept 2007

Commissioner General for the

Shanghai World Expo 2010 and

the Prime Minister’s personal

representative on the Socialist

International Committee

11 Bean R, “MP fury at ‘poorer

Wigan’ report”, Wigan Today,

11th November 2007, see

www.wigantoday.net/wigannews

/MP-fury-at-39poorer-

Wigan39.3460842.jp

concludes that were McCartney to visitHindley, “and take a good hard honestlook, on reflection he may then under-stand why Wigan as a whole cannot climbout of the doldrums and attract economicgrowth.”

John Healey, the Minister for LocalGovernment, responded to Cities limitedimmediately on its publication. Althoughacknowledging that “deep-rooted pocketsof deprivation exist”, he said: “We totallyreject these claims.” Again, he did notengage with our findings, but insteaddefended the economic effects of regenera-tion policy by noting that the coalfieldsprogramme had created 16,000 jobs since1996. We do not believe, however, that thebest way to analyse the success or otherwiseof urban regeneration programmes is tolook at outcomes, programme by pro-gramme. We stand by our approach oflooking at the effects of regeneration poli-cy on areas as a whole. After all, for peopleliving in those areas, it is the aggregateposition that matters.

Even when we look at regeneration on aprogramme by programme basis, the suc-cess of initiatives like the coalfields regen-eration programme is difficult to establish.Mr Healey stated that it has generated16,000 jobs since 1996. Clearly for those16,000 people the programme was a suc-cess. It would be wrong to deny that theprogramme had had some successful out-comes, although it is surely the case thatsome of those people, perhaps a substantialproportion, would have got work withoutthe programme, and likely that some whogot work did so at the expense of otherlocal people. But even if we take the16,000 figure as a fair reflection of the pol-icy’s outcome, we need to ask whether16,000 is a large number. This policy cost£379 million, covered 107 different placesand ran for more than ten years. Simplydividing the number of jobs created by thenumber of places covered and the numberof years for which the programme ran tells

us that the coalfields regeneration pro-gramme – hailed by the minister as evi-dence of the success of regeneration – cre-ated 16 jobs per community per year.12 Ifone job every three weeks is success, wecannot imagine the minister’s definition offailure.

There are signs that the Government’sthinking about urban regeneration hasbecome more thoughtful recently. StephenTimms, then the Minister of State forCompetitiveness at the Department forBusiness, Enterprise and RegulatoryReform set out a new way of looking atthese issues in a recent speech to the FabianSociety’s city, business and politics networkseminar.13 He argued that we “need tomove on from this sterile ‘North/Southdivide’ debate”, and look instead atLondon’s exceptionalism. He noted:“London stands out”, it “has been anextraordinarily successful world city overthe past decade”. His conception ofLondon is spot on: “London’s successaccentuates the differences with the UK’sregional economies – but it provides apowerful driver for them too. London isn’ta rival; it’s an ally for regional growth. Thesuccess of London spills over to neighbour-ing regions, which benefit from easy accessto a large market, from diffusion of knowl-edge and best practice.”

The analysis is correct and the speechannounced significant and welcome fund-ing for research into spatial economics. Butthe speech can be read two ways. The min-ister was right that “It makes no sense tojudge the success of the English regionssimply by comparing them with London”.Instead he argued for regional compar-


Reactions to our work

12 We take into account the

time period for which each place

was covered by the programme

13 Timms S, (former Minister of

State for Competitiveness),

Strengthening Regional

Economies, Fabian City,

Business and Politics Network

seminar, Fabian Society,

London, 14th January 2008, see

www.berr.gov.uk/pressroom/Spe

eches/page43638.html

“ Even when we look at regeneration on a programme by

programme basis, the success of initiatives like the coalfields

regeneration programme is difficult to establish”

isons: “So, for example, from 1995 to2004, growth in South Yorkshire wasalmost double that of Düsseldorf inGermany or Lorraine in France. That is thekind of comparison we need to draw, tounderstand properly the effectiveness ofregional policy.” We strongly support theidea of learning about what works andwhat does not work by looking at experi-ence outside the UK. The belief that wecould learn from others underpinned thesecond in this series of publications, Successand the city: learning from internationalurban policies.

But there is a big difference betweenarguing on the one hand as we do that wecan draw qualitative lessons, ex post, fromforeign experience and believing on theother that the ex post construction of a for-eign sample represents a legitimate methodby which to assess British town andregions, or British policymaking. UnlessBritain is bottom of the league table it willalways be possible ex post to find placesthat have done worse than any particularBritish town or region. Finding a town orregion elsewhere in the world that has per-formed worse than a place in Britain does

not mean that the British town has per-formed well, still less that policy has beensuccessful. The minister’s phraseologyallows almost any outcome to be defined asa success, and that cannot be a good foun-dation for evidence-based policymaking. Ifwe are to use the performance of others asa yardstick of our own performance, weneed to specify the comparator towns andregions ex ante not ex post, otherwise wecan simply move the goalposts to whereverour ball happens to end up.

Nevertheless, the analysis in the speechis powerful, and, as we argue in more detailbelow, it is right to see London as unique,and it is right to argue that London’s suc-cess spills over into neighbouring regions.But for people in regions that do notneighbour London this has an obvious andunavoidable implication: if you want toshare in London’s success, you may have tomove to London, or at least to one of itsneighbouring regions. As the old phrasegoes, if you can’t beat ’em, join ’em. Weshall argue that this is the correct conclu-sion, and we hope that the minister’sspeech indicates that this is the way thatgovernment thinking is developing.

Cities unlimited

18


14 World’s Top 10 Economic

Centers, The 2007 List, Standard

& Poor’s, December 2007, New

York

15 Beaven R, Bosworth D,

Lewney R and Wilson R,

Alternative Skills Scenarios to

2020 for the UK Economy: A

report for the Sector Skills

Development Agency, as a con-

tribution to the Leitch Review of

Skills, HM Treasury, 2005 see

www.hm-treas-

ury.gov.uk/media/8/8/alternative_

skills_scenarios_execsummary.p

df

5Things will only getworse

The trajectory that we have outlined so faris not a hopeful one for the many peopleliving in those towns and cities targeted byBritain’s urban regeneration programmes.There are four reasons to think that thingsare likely to get worse, and not better, inthe future.

Economic geographyWe have already set out in Cities limitedhow economic geography has changedsince many of these towns were created inthe 19th century. There is no reason tothink that the changes we have identifiedhave finished, much less been reversed.London’s international status as a premierfinancial centre is unlikely to be usurpedby any other place in Britain.14 The increas-ing role of the service sector will favourthose towns that are already orientedtowards services. Agglomeration effectswill continue to grow, boosting firms locat-ed in big cities, wherever they may be.Firms outside those areas then have achoice: to remain where they are, thusmissing out and risking bankruptcy, or tomove to the thriving cities.

The changing skills mixThe last 50 years have been a period inwhich the value of skills, particularly whitecollar as opposed to manual skills, hasincreased enormously. It used to be thecase that people could leave school at 16and get a job that would give them adecent wage for life. That is increasingly

unlikely these days and will prove to beeven less likely in the future.15 This is a realproblem for our regeneration townsbecause they suffer from a much lessattractive skills mix than London. That isnot an entirely new phenomenon: Londonhas always attracted a large proportion ofthe highly skilled. What is new is the scaleof the income differentials that thisinduces. There was a time when the educa-tional qualifications associated with a towndid not matter that much for its incomelevels, but that time has past.

People in our urban policy town samplewere less likely to be highly qualified, and,as we reported in Cities limited, schoolresults were markedly worse in our regener-ation town sample. Students from theseplaces were 8 per cent less likely to gain fivegood GCSEs than their equivalents fromour prosperous towns. So each town does

0

10

20

30

40

50

Regenerationsample

GB average Successfulsample

less than NVQ2

NVQ4 or above

%

The proportion of working age population qualification levels 2006

Figure 2: Educational attainment

Notes: NVQ2 is five GCSE grades A*-C or equivalent,

NVQ4 is undergraduate degree or equivalent

not start out with an equally well-qualifiedset of 16 year olds every year. We thereforehave a problem: our less successful townshave a lower skills base among the existingworkforce, their students are achievinglower grades in schools, and many of theirbest students leave for London after gettinga degree. In contrast, successful areas startwith a higher skills base among their exist-ing workforce, their students are achievinghigher grades at school, and London in par-ticular attracts many of the best studentsfrom other parts of the country, and indeed,from other parts of the world, after gradua-tion. Given this divergence in human capi-tal it is almost impossible to imagine anykind of overall economic convergence.16

PoliticsFor the last decade British politics has beendominated by ministers who represent poor-er urban areas. Of the current Cabinet – Jack

Straw represents Blackburn; David MilibandSouth Shields, Tyne and Wear; Jacqui SmithRedditch; Harriet Harman Camberwell andPeckham; Hazel Blears Salford; AndrewBurnham Leigh; Hilary Benn Leeds Central;Alan Johnson Hull West and Hessle; EdBalls Normanton, West Yorkshire; andYvette Cooper Pontefract and Castleford.17

Inevitably they are committed to urbanregeneration and it should come as no sur-prise that spending on urban renewal hasincreased in the last few years.

Figure 3 shows, there has recently beena step change in funding, after a decade ofstability. The rise in spending under theLabour Government is comparable withthe rise under the Tories in the late 1980s.But while the Tories were responding to anobvious rise in urban problems caused by asharply changing industrial structure,Labour spent more simply because it choseto. Spending from choice rather thannecessity is much easier to reverse.

Cities unlimited

20

16 Juhn C, Murphy K and Pierce

B, “Wage inequality and the rise

in returns to skill”, Journal of

Political Economy, 101(3): 410-

42, 1993; and Autor D, Katz L

and Kearney S, “Trends in US

Wage Inequality: Revising the

Revisionists”, Review of

Economics and Statistics, 90(2):

300-323, 2008

17 Full list of Cabinet members

as of 24th January 2008, Labour

Party, 30th April 2008 see

www.labour.org.uk/members_of_

the_cabinet

197719781979198019811982 1984198519861987198819891990199119921993199419951996199719981999200020012002200320042005

Per

ann

um fu

ndin

g (£

billio

n, 2

005

real

pric

es) 7

6

5

4

3

2

1

0

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Note: Data based on government spending with explicit urban regeneration objectives and excludes transport, housing and

industrial policies unless part of a regeneration programme. Data include: Urban Programme, Urban Development Corporations

(including Docklands Light Railway and Jubilee Line Extension), Enterprise Zones, Urban Development Grant, Urban

Regeneration Grant, Derelict Land Grant, City Grant, Estate Action, Inner City Task Force, City Action Teams, English

Partnerships, Housing Action Trusts, City Challenge, Coalfields Areas Fund, Objective 1 and URBAN II European Funding, Single

Regeneration Budget Challenge Fund, Manchester Olympic/Commonwealth Bids, Coalfields Regeneration Trust, Coalfields

Enterprise Fund, Regional Development Agencies, New Deal for Communities, Neighbourhood Support Fund, Neighbourhood

Renewal Fund, Housing Market Renewal Pathfinder, Sustainable Communities Plan (without housing costs), Local

Environment/Liveability Fund, Community Empowerment Fund, Thames Gateway, London 2012 Olympics. We gratefully

acknowledge the support of Dr Mark Pennington, Richard Wellings, Prof Peter Roberts Jim Gill and Prof Michael Parkinson in

compiling and reviewing these statistics. The responsibility for any errors and omissions are the authors alone however.

Figure 3 - Annual Urban Regeneration Funding, 1977-2005

Britain will not always have a Cabinetdrawn from poorer inner city areas. It is ofcourse possible that a Cabinet who repre-sent leafy shire towns in Britain’s moreaffluent counties will prove to be as gener-ous towards inner cities, but such an out-come cannot be assumed. To some extentpoliticians will always, and perhaps shouldalways, be in line with their constituencies.It seems most unlikely that levels of urbanregeneration funding are hot topics inWitney or Tatton.

All opposition parties need to distin-guish themselves from incumbent govern-ments. They need to offer a different set ofpriorities if they are to win over the elec-torate because offering more of the samedoes not create a unique selling point. Adifferent set of priorities offered by anopposition party may well involve lowerlevels of spending on urban regenerationfunding, with inevitable consequences forthe fortunes of these towns.

MoneyThe Government’s finances are in a parlousstate. Asked in a Financial Times survey inJanuary whether the public finances are inreasonable shape 95 per cent of the econo-mists polled who gave unambiguousanswers said No.18 Their comments includ-ed: “The public finances are in very poorshape”; “in an increasingly dire state”; “ashocking state”; “a parlous state”; and an“awful shape”.19 When governments runout of money, something has to give. TheGovernment tried to save money recentlyby phasing in the annual pay award givento the police. This will have no long-termeffect on government spending, since thepolice will receive their full pay rise by theend of the year, but it provides a one-offsaving of £40m this year.20

As the government found out, savingmoney by staging pay awards is politicallycostly: the story dominated the headlines.The same would be true for the salaries of

nurses, doctors and teachers. And as fortrying to cut the number of police, nurses,doctors or teachers, well, that way lies cer-tain electoral disaster.

Regeneration funding, in contrast, isproject and programme based. Individualprojects and programmes come and go, oldones are phased out, and new ones comealong. Sometimes the new ones are betterfunded than the old ones, sometimes theyare not. Importantly, the political cost offunding replacement programmes less gen-erously is not that high. Did anyone out-side the specialist press notice whether theHousing Market Renewal Pathfinders werebetter or worse funded than the HousingInvestment Programme they replaced? Orwhether the Crime Reduction Programmecost the government more or less moneythan the Safer Communities Initiative?Did anyone notice when the CapitalModernisation Fund (Small Retailers)ended? Whether or not these schemeswork is not the point here. The point isinstead the more brutal one of politicalreality: when you run out of money youwant to cut things that the public will notnotice, or at least you can disguise byannouncing a similar programme thatproves on closer inspection to be (much)less well-funded. If the number of nursesemployed goes down by even one that willbe noticed. If the Capital ModernisationFund (Small Retailers) is not replaced, noone notices. That makes the CapitalModernisation Fund (Small Retailers) andmany, many other schemes – includingmuch larger schemes – very vulnerable toany downturn in government finance, evenwhen the Government is genuinely com-mitted to urban regeneration, and even if itbelieves that schemes like this work.

For these four reasons, 1997 to 2008 arelikely to be seen as the high watermark inBritish urban renewal spending. No matterwhich party (or parties) wins the next elec-tion, the nation’s fiscal position means thatthere is little room for manoeuvre, little


Things will only get worse

18 Weale M, NIESR, “Are the

public finances in reasonable

shape? If not, what needs to be

done in 2008?”, Q5,

Economists’ Survey: the results,

Financial Times, January 1st

2008, see www.ft.com/cms/s/0/

6d80bacc-b894-11dc-893b-

0000779fd2ac.html

19 Ibid, Comments from Coynes

J, Capital Economics; Spencer

P, Ernst & Young Item Club;

McCafferty I, CBI Chief

Economic Adviser; Lewis S,

Insinger Beaufort, respectively

20 Parker G, “Brown charts colli-

sion course”, Financial Times,

April 3rd 2008, see

www.ft.com/cms/s/0/7fe6fb66-

0107-11dd-a0c5-

000077b07658.html

room for funding innovative ideas and cer-tainly little room for funding projects thatdo not have obvious, immediate resonancewith large swathes of voters.

This is turn is very bad news for oururban regeneration towns. We have beenquite clear that urban regeneration spend-ing has not worked, in the sense that thesetowns are not regenerated. But we havebeen equally clear that those policies haveraised the standards of living in these townscompared with what they would have beenwithout them. It is, after all, hard to spendbillions of pounds without achieving any-thing. So when those billions dry up – assome of them will – towns that are alreadyslipping gradually further behind the UKaverage will not simply continue to slipbehind at their current rate, but will start toslip behind more rapidly.

This is exactly what happened in themid-1990s, the last time that governmentfinances were heavily constrained by atough fiscal environment. In that periodGVA in our regeneration towns fell relativeto the national average twice as fast as inthe fiscally relaxed late 1990s. The samewas true of personal incomes. Whenmoney is readily available, and regenera-tion funding is high, our regenerationtowns slip gradually backwards. But whenmoney is tight and regeneration funding isunder pressure, they fall much furtherbehind more quickly. Even when the goodtimes return, the last ten years demonstratethat the period of rapid decline is notreversed or even halted. When these townsdecline relative to the national average,government is powerless to reverse thatdecline later.

Cities unlimited

22

6We cannot accept this

Current policies mean that poorer townswill continue to get poorer relative to therest of the country. When the nationaleconomy is doing well, and money forregeneration is plentiful, they will get grad-ually poorer. When the national economyis doing badly, and money for regenerationis hard to come by, they will get poorermuch more quickly.

That is not to say, of course, that theywill get poorer in absolute terms, for theywill not: economic growth reaches all partsof the national economy to some extent.But they will inexorably diverge from therest of the country. Children growing up inthese towns will not have the same experi-ences – and perhaps not even the sameexpectations and aspirations – as thosegrowing up in places that are more typicalof Britain.

We are not talking about divergencefrom a handful of super-rich places – thathas always existed, and always will. We aretalking about growing divergence fromplaces in Britain that are simply average.Adults looking for work will not have thesame opportunities as those who, with thesame skills and determination, are simplylucky enough to be in a different part ofour country. Geographical circumstancemeans that life on benefits, whether inwork, or out of work, will be more com-mon.

That cannot be right, and we should notaccept it. Nor is there any need to accept it,for while there are no easy answers, thereare answers grounded in theory and empir-ical evidence from Britain and abroadwhich point us to solutions that will work.That must be our agenda.


7What can be done?

Accept the reality of economicgeographyA successful regeneration policy has to startby accepting the realities of economicgeography, for it is at least as hard to buckgeography as it is to buck the market.Spatial economics, the study of the eco-nomics of geographical dispersion andagglomeration, advances strong theoreticalarguments, and empirical evidence for theexistence of optimal locations for econom-ic activity and optimal sizes for cities, bothof which vary in time and place. It alsodemonstrates how planners can use pricesignals to guide their decisions so that acountry ends up in a position closer to theoptimal level. A country that uses suchinformation effectively will be richer thanone that does not. We look first at whateconomic geography can teach us aboutoptimal locations, before turning to opti-mal city sizes.

The existence of optimal locations foreconomic activity is at one level very obvi-ous: it is easy to sell sandwiches in a busyarea with lots of workers or tourists look-ing for something to eat, just as it is easy tosell beach mats at the beach. What is per-haps less obvious is that there are optimallocations for firms that at first sight mightnot be expected to have an optimal loca-tion. In fact all firms have an optimal loca-tion. They might want to be near workerswith the relevant skills and aptitudes.Many want to be near their customers,either directly or via good transport links.Others need to be able to export the prod-uct that they make or to import compo-nents. Finally, firms prefer to locate next to

other firms in the same industry. Partly thisis so that they can attract staff from theirrivals, but partly it is because knowledgespillovers within a cluster of firms makethem all more competitive. For all thesereasons many firms prefer to be in or nearmajor centres of population, and to be wellconnected to other centres of population.That gives them access both to staff, tocustomers and to knowledge.

Today Britain trades with Europe morethan ever before. This gives the South Eastan advantage. Aviation has become moreimportant over time, for both goods trans-port and passenger travel, so locations nearairports, particularly airports that offer awide range of destinations, have becomemore attractive, again favouring the SouthEast. Finally, Britain has become a road-based economy, favouring places in themiddle of Britain that are well connectedto motorway networks. Some disadvan-tages of geography would be prohibitivelyexpensive to overcome. It would be possi-ble, for example, for the Government tobuild an airport next to Hull, and subsidiseflights to and from Hull to ensure that it isas well-connected as London. But such apolicy would be economic as well as envi-ronmental folly. We have to accept thatsome locations make more sense than oth-ers and that, although infrastructure can beimproved, there are places that geographi-cal reality dictates will never be well-con-nected.

That does not mean such places aredoomed or that they should be abolished.But it does mean that the only way theywill attract jobs is by offering a lower cost

24

of doing business in compensation fortheir lower connectability. This happens tosome extent already: rents are lower, forexample. But rents make up only a smallproportion of total costs and differences inregional employment rates and regionalwages show us that lower rents in less well-connected places are insufficient to ensurehigh levels of employment. In reality, theonly cost of production that accounts for asufficiently large share of total costs, andwhich can vary place by place, is labourcosts. As a result, the only way towns andcities that are less well connected, andwhich the realities of physical geographydictate will remain less well-connected, cancompete to attract firms is to accept lowerwages.

That is a brutal message, but it is anhonest one. When something is true, noth-ing can be gained by hiding it. The mes-sage for people living in cities that are lesswell-connected is simple and apparentlybleak: if you remain where you are, thenyour chances of being unemployed arehigher and your wages will be lower than ifyou move to a place that is better locatedfor the modern economy. It is worth beingclear about the different impact of thesetwo effects. Skilled workers are not unem-ployed in large numbers anywhere inBritain, for them the issue is one of wages.In places that are economically less well-located they may find that the opportuni-ties are not as well-matched to their skillsas they would be in a more successfulregion, and that wages are generally lower.As economists would say, the skilled labourmarket clears, but it does so at a lowerwage. For unskilled workers the issue isone of unemployment, because unemploy-ment benefits and minimum wages place afloor on wages. As a result we find thatunemployment of those with low or noskills is substantially higher in less-well-connected cities. This is a market that doesnot clear and some people who would gen-uinely like to work remain unemployed.

This matters because it tells us thatthose who are worst affected by poor eco-nomic locations are those with low-skilllevels. Those with higher skills have agreater ability to move because their wagesare sufficient to allow them to move tomore prosperous regions. Those with high-er skills are compensated for their lowerwages by a lower cost of living: housing, inparticular, is much cheaper in our regener-ation cities than either in Britain as awhole, or in our most successful towns. Ifthose with skills can find a reasonable job,their standard of living will be only a littlelower. If it were much lower, then theywould move. But for the low skilled thestory is not the same because unemploy-ment is much more likely in areas that arenot well connected, and those with lowskills find it hardest to move.Unemployment is not only a source ofpoverty; it is also a source of unhappinessover and above the unhappiness that comesfrom poverty. The evidence shows thatthose who want to work do not get used tounemployment, on the contrary, the painof unemployment remains real over time.21

That some areas are innately hard toconnect is not contentious. The difficultyof connecting the middle of the LakeDistrict, or the Highlands of Scotlandexplains why so few people live there: theseplaces, which are outstandingly beautiful,are not good locations for business andnever have been. Those who choose to livethere do so because they offer other attrac-tions.

What may be contentious is to state thatsome of Britain’s existing towns and citiesare poorly located for modern business.This stems from two interlocking factors:that Britain urbanised first and that theoptimal location for business has changedmore in Britain than elsewhere. This com-bination means that our economic geogra-phy is now uniquely poor.

As we noted in Cities limited Britain isexceptional in the extent to which it


What can be done?

21 Layard R, Happiness:

Lessons from a New Science,

Penguin, London, 2005, p 67

urbanised in the 19th century. As early as1841 about half the British populationlived in towns, rising to around four-fifthsby 1914, making Britain by far and awaythe most urbanised country in Europe.22

Britain is therefore unique in the world inhaving the location of so much of its hous-ing stock determined more than a centuryago. This would not matter if the determi-nants of optimal city locations hadremained the same, but they have not. The19th-century British economy was basedon manufacturing, sea-borne exports andcoal, and so the free market built our citieson coalfields near ports. Liverpool was themain port of entry from America, the per-fect place for literally billions of pounds ofraw cotton to arrive from the AmericanSouth every year, ready to feed the mills ofManchester and its environs, which werein turn fed by coal, brought in by canal.Both Liverpool and Manchester grew dra-matically in the 19th century, rising fromsignificant towns of around 100,000 togreat urban metropolises of 600,000 each.They were in the right place, at the righttime, and they grew accordingly.

The factors that meant that they were inthe right place then no longer hold true,because the 21st century does not look likethe 19th century. Firms do not burn theirown coal anymore, and in any case the costof transporting coal has fallen dramaticallyrelative to the value of output. Electricityand gas are moved easily and cheaply acrossthe country. With the exception of a hand-ful of extremely energy-intensive industriessuch as aluminium production, the locationof energy sources is no longer an importantdeterminant of the location of firms.

Nor is sea-shipping as important now asit once was, and even within sea-ports theEuropean-oriented ports of Felixstowe andDover have risen in importance relative tothose of say Liverpool, particularly afterBritain joined the EU. In 1972 Liverpoolwas responsible for more of Britain’s mer-chandise exports than Dover andFelixstowe combined. Twenty years laterthe positions had been dramaticallyreversed, and Dover and Felixstoweaccounted for around ten times as many ofBritain’s exports as Liverpool. Liverpoolwas once the closest place in Britain to theUnited States, today Heathrow has thatposition. It offers more flights, to moreplaces, at more times than all of theNorthern airports combined, and competi-tion pressures mean that fares are usuallycheaper as well.

More generally, the replacement of sea-borne trade with truck-born trade meansthat coastal towns have suffered relative tothose on the main motorway network.Places such as Corby and Daventry wouldhave made little sense as employment cen-tres in the 19th century, yet today theymake excellent economic sense. In con-trast, while places like Hull and Liverpoolwere great economic locations in an era ofsea-based transport, they are much lesseconomically attractive locations fromwhich to do business today.

Finally the North has been unlucky.Liverpool was once a booming city, maderich on the cotton trade. The same wastrue for place like Oldham and Bolton, thecotton capitals of the world. They rivalledLondon, the financial capital of the world.But although London remains one of thegreat financial centres, Liverpool, Oldhamand Bolton are not booming, because thecotton industry proved to have no futurein Britain. It is no fault of Lancashire thatcotton proved to be a less secure basis forlong-term prosperity than finance. TheNorth was simply less lucky than the Southin that its source of economic power disap-

Cities unlimited

26

22 Leunig T and Swaffield J,

Cities limited, p 14, Policy

Exchange, London, 2007

“ When transport costs fall, cities work better, since it

is easier to bring both people and the things that they

need together”

peared for exogenous reasons – competi-tion from much lower wage countries –about which it could do nothing. And itwas doubly unlucky because when the timecame to replace these industries, the Northwas no longer the desirable economic loca-tion it had once been.

There are thus geographical reasons whypeople in the South are generally richerthan people in the North. Southern pro-ductivity is inherently higher than north-ern productivity not because southernersare more quick witted, harder working orintrinsically smarter, but because southern-ers have the advantage of being well-con-nected. Connectivity matters and it raiseswages. This means that the North-Southdivide is not a transient problem thatreflects a short-lived hiatus as manufactur-ing declines and service sector economiestake its place. Nor does it reflect a failure ofgovernment policy to do what is needed.Rather, it is evidence that the North, thepower house of the industrial revolution, isnow a less desirable location for businessfor reasons that are geographical andentirely outside its control or that of cen-tral and local government.

The phrase “North and South” is a con-venient shorthand, but of course there areplaces in the North of England that arelarge enough to have reasonable levels ofmarket potential. Here the successful citiesof Manchester and Leeds stand out, andwe discuss them in greater depth later.Equally, there are places such as Hastingsin the South that are not at all well con-nected, even though they appear to beclose to London and European markets.Again, we look at places like these in moredetail later.

The second main finding of spatial eco-nomics is that city size matters. This hasalways been true, right back through theindustrial revolution, through medievaltimes to the Romans and no doubt before.When people meet each other they learnfrom each other, they get ideas of new

things to do and new ways to do thingsthey already do. Being closer to their cus-tomers, they have a stronger sense of themarket and what customers want. Schoolsand universities, and their students, areaware of what local employers are after,and can think about matching their ownskills and potential interests to thosedemands. In short, employability increas-es, productivity rises and innovation ismore common.

Of course, larger cities have their down-sides: they are more expensive and congest-ed. Initially, as a city grows, congestiondoes not increase and the cost of livingdoes not rise very much; the gain from alarger city outweighs the costs. But laterthese agglomeration economies may ceaseto grow, particularly if a single-industryfirm is dominant. Additional people willnot offer much benefit in knowledgespillovers, but the cost of congestion con-tinues to rise and, if gridlock becomesmore common, to rise rapidly. At thatpoint the city stops growing: people nolonger want to move there because thegains from higher wages are outweighed byhigher costs and greater congestion. Thereis thus an optimal city size: big enough tobenefit from the economies of scale, smallenough that congestion and commutingcosts are not too high.

There are five reasons to believe thatoptimal city sizes have grown in recentyears. First, domestic and internationaltransport costs have fallen. Secondly, serv-ices have become more important. Thirdly,the economy has become more skill inten-sive. Fourthly, information communica-tions technology (ICT) has allowed theproduction process to be split more easily.Fifthly, there has been a big rise in dual-career couples.

When transport costs fall, cities workbetter, since it is easier to bring both peo-ple and the things that they need together.We can see this with a simple example:central London, like central business dis-


What can be done?

tricts everywhere, would collapse if peoplehad to walk to work since there would notbe room for all those who work there tolive within walking distance, nor couldthey readily be supplied with foodstuffs.Big cities need effective transport.

That the service sector has grown in thelast 50 years is well known. What is per-haps less well known is that service sectorfirms gain more from agglomerationeconomies than manufacturing firms. Thisis because manufacturing firms do notseem to benefit from an increase in the sizeof the manufacturing sector per se, butonly from a rise in the number of firms intheir own industry.23 Such benefits areknown as “localisation” economies. A cot-ton manufacturer will see productivity risewhen other cotton manufacturers move tothe town, but not when a firm producingrailway carriages arrives. This leads to sin-gle-industry towns – indeed, cotton townswere even specialised by market segment,with weaving towns in Lancashire’s northand spinning towns in Lancashire’s south,and even within this, fine spinning con-centrated in Bolton, and the spinning ofcoarser yarns in Oldham. There was nogain to fine yarn spinners from being neareither coarse yarn spinners or weavers,indeed since locating in the same placewould have increased congestion, it wouldhave been harmful. As a result, each spe-cialism located in separate, medium-sizetowns. The largest Lancashire towns –Liverpool and Manchester – were as muchmerchant service towns as manufacturingcentres, again suggesting that a concentra-tion on service sector firms leads to largercities.

In contrast there is much evidence thatservice sector firms do gain from the pres-ence of other unrelated service sector firmsin the same area. This seems to reflect inpart a greater reliance on transferable skills.Universities and financial firms bothemploy IT specialists, HR managers, andso on, and insofar as productivity in these

jobs rises with the numbers employedlocally, the university will benefit fromlocating near the financial firm and viceversa. These are termed “urbanisation”economies, and are larger than single-industry “localisation” economies.24

The improvement in skill levels leadsto a change in the nature of work whichraises the level of potential agglomerationeconomies. Once a large proportion ofthe workforce was engaged in workwhose nature was precisely set down bymanagement – such as Fordist produc-tion lines – and for which discretion,knowledge and learning were very limit-ed. Today such work is less common andworkers have more opportunities to showinitiative. Managers and workers can acton workplace discussions and so it makesmore sense than ever to work near otherpeople.

The role of ICT in agglomerationeconomies is much misunderstood. Therewas a time in which people thought that acombination of cheap telephones andbroadband would mean that we could andwould all work from home. Today the costof working from home is far, far cheaperthan the cost of commuting for many peo-ple and yet few people do it. The reason isthat in many jobs agglomerationeconomies are too large to resist. Workersknow, and their firms know, that they aremore productive when at work.25 This isnot because they slack at home, but ratherthat when in work they talk to their col-leagues, formally and informally; theysense the way their firm and industry ismoving, in a way that is simply not possi-ble from home.

In fact, far from leading to dispersion,ICT leads to larger service sector cities. Inthe pre-ICT era firms de facto had tolocate all elements of production togeth-er.26 A car firm, for example, would design,engineer and build the car in the sameplace, and that place would also have headoffice functions such as accounting, per-

Cities unlimited

28

23 Henderson J, “The Sizes and

Types of Cities”, The American

Economic Review, 64(4): 640-56,

1974; Henderson J, Urban

Development: Theory, Fact and

Illusion, Oxford University Press,

1988

24 Rosenthal S and Strange W,

“Evidence on the Nature and

Sources of Agglomeration

Economies”, in Henderson J and

Thisse J-F (Eds) Handbook of

Urban and Regional Economics,

volume 4: Cities and Geography,

Elsevier Science, 2004

25 Boden D and Molotch H,

“The Compulsion of Proximity”,

in Friedland R and Boden D

(Eds), NowHere: Space, Time

and Modernity, University of

California Press, 1994; Ormerod

P, Cook W and Rosewell B, Why

Distance Doesn’t Die –

Agglomeration and its benefits,

GLA Economics, Working Paper

17, June 2006

26 Linden G et al, “Who

Captures Value in a Global

Innovation System? The case of

Apple’s ipod”, Personal

Computing Industry Center, June

2007

sonnel, etc. Since car manufacturing wasthe dominant part of the process, thedesign, engineering, accounting parts werelocated in the place that was optimal formanufacturing. Today ICT means thatthese different elements can be located indifferent places. We find that the non-manufacturing elements are no longerlocated with the manufacturing elements,but are rather located in the same place asothers doing the same type of work. Thenewly launched Nissan Qashqai, for exam-ple, was designed at Nissan’s own designstudios near Paddington, engineered inNissan’s own technical laboratories inCranfield, and then built in Nissan’sassembly plant in Sunderland. Theaccounts are done in Budapest. Nissanchooses to locate its design studio inLondon because London is the centre ofthe British car design industry. Ford has adesign studio there as well. It is a placewhere it is easy to hire talented car design-ers, and where they can rub shoulders withother talented car designers, and becomegreat car designers. The same is true for carengineering in Cranfield, Britain’s centrefor car engineering. The result is that themanufacturing cities are smaller than theywould otherwise have been because theyno longer contain their associated non-manufacturing activities. And the servicesector cities have grown further becausethey attracted not only independent firms,but also the service sector parts of manu-facturing firms that no longer have tolocate alongside their manufacturingwings. The optimal size of service citieswill grow and manufacturing cities willshrink. Furthermore, many of the jobs lostby the manufacturing cities are the higher-skilled, better paid, positions.

Finally, the growth of the dual-careerhousehold, the product of a rising propor-tion of people going to university, meansthat more households than ever seek to livein a city that offers career opportunities topartners who may be in very different pro-

fessions. Large cities are particularly attrac-tive to dual-career couples. Notice thatBritain is still at a very early stage in thisprocess, since the rise in the number ofpeople attending university is a relativelyrecent phenomenon and will take 40 yearsto reach its full effect.

There are therefore a very large range offactors – some of them ongoing – thatmean that the optimal size of cities hasincreased over time, perhaps considerably.

It is worth commenting a little on luckand what economists call “path-dependen-cy” (what everyone else calls the chicken-and-egg syndrome). There was a timewhen Hollywood was just another place.But then it became for a host of reasons theglobal centre for film-making. It is noweasier to make films in Hollywood thanelsewhere because it offers all the ingredi-ents necessary to make films. There isnothing intrinsically favourable aboutHollywood, but once established as thefilm capital of the world, it is hard forother areas to compete with it in film-mak-ing. The same was true a century ago forthe Lancashire cotton industry, then ahighly skilled sector. Lancashire spinnerscould spin cotton like no others. For a cen-tury no one else could compete. But ulti-mately cotton is a sector in which costs areparamount; film-making is a sector inwhich quality is paramount. Cotton couldnot survive in high-wage Lancashire, film-making can and does survive in high-wageCalifornia.

It would be hard today for another areato become the film capital of the Westernworld, just as it would have been hard acentury ago for another area in Britain tochallenge the cotton capital of the world. Awell-meaning politician could not havemoved part of the Lancashire cotton indus-try to Suffolk, because that would havedestroyed the agglomeration economies onwhich Lancashire’s success was founded.The only way in which a Suffolk cottonmill could have competed against those in


What can be done?

Lancashire would have been to pay wagesso low that they compensated for the lossof agglomeration economies, in short,wages so low that no one would have want-ed to work there. That is why it did nothappen.

What was true for the Lancashire cottonindustry a century ago is true for the Cityof London today. That London is Europe’sfinancial centre is partly luck. The originalBritish financiers lived in London ratherthan Bristol or Liverpool, and the agglom-eration economies grew out of that. Theyhave been lucky that Britain has never beenon the losing side of a major war, nor hasour economy or currency collapsed, nor

have governments stifled the City to thepoint of forcing firms to flee. The industryhas had the conditions to succeed and ithas worked hard to do so, helped along theway by growing agglomeration economies.Stephen Timms, the former Minister ofState for Competitiveness, is right: Londonis one of a group of successful global citiesthat can be compared to each other, butwhich stand above other places in thedomestic economy. We cannot, with thebest will in the world, move JP Morgan toBlackburn, or Deutsche Bank toSunderland. We may wish it otherwise, butwe know that it will not happen. Londonis London.

Cities unlimited

30

27 “The History of Shipbuilding

in the North East”, BBC, 2nd

April 2008, see

www.bbc.co.uk/nationonfilm/topi

cs/ship-build-

ing/background.shtml

28 Clark K, “Final Blow for Glass

Factory”, Sunderland Echo, 28th

September 2007, see

www.sunderlandecho.com/news

/Final-blow-for-glass-facto-

ry.3242477.jp

29 1,000 go for Nissan jobs,

Sunderland Echo, 4th February

2008, see


/1000-go-for-Nissan-

jobs.3740559.jp

30 Strike threat at Nissan over

pay row, Sunderland Echo, 22nd

December 2006, see


/Strike-threat-at-Nissan-

over.1943503.jp

31 “Invest in Sunderland”,

Sunderland City Council

Business Invest, 2006, see

www.investinsunderland.co.uk/in

ward-sector-auto.asp

32 Dorling D et al, op cit

Box 1: Sunderland – a case study of the limits of regeneration

Sunderland lies at the mouth of the River Wear in North East England, 13 miles southeast of Newcastle.Between 1850 and 1950, Sunderland was the “largest shipbuilding town in the world”.27 In 1900 ship-building employed over 12,000 men, a third of the town’s adult population, and the city built morethan a quarter of Britain’s ships during the Second World War. But shipbuilding, especially non-mili-tary vessels, proved to have little future anywhere in Britain after 1945, and the last Sunderland ship-yard closed in 1988. Sunderland is also part of the wider Durham coalfield, which declined at a simi-lar time. Vaux Breweries closed in 1999 after more than 100 years of brewing, and commercial glassmaking has recently ceased in Sunderland, after more than a century as a major producer.28

Sunderland has received much regeneration funding, but the most interesting aspect of the cityfrom the point of view of regeneration is the arrival of the Nissan car plant in 1986.

The plant was created on a 300-hectare, greenfield site on the outskirts of Sunderland, sold toNissan at agricultural prices. Opened in 1986, it has grown steadily adding engine assembly in 1990,moving to two-model assembly in 1992, three-model in 2000 and four-model in 2006. Nissanrecently announced that it would move to three-shift production, meaning that the plant would run24 hours a day, seven days a week.

After an investment of £2 billion in the last two decades, it is now Britain’s biggest car plant byboth output and exports, and on any measure is a major success. Once the move to three-shift work-ing is completed, it will directly employ about 5,000 people.29 Application levels are high whenvacancies occur.30

Not only does Nissan provide jobs, but it has also attracted other firms to the area. SunderlandCity Council reports that there are 240 Nissan suppliers in the area, including major automotivefirms such as the French company Valeo and the US firm TRW Automotive. A total of 12,000 peo-ple work in the automotive sector.31

And yet for all this, Sunderland remains poor. With gross value added per person at 16 per centbelow the UK average, it is in the poorer half of our regeneration towns sample. A recent JosephRowntree Foundation study found that the majority of the population was poor, and numbers hadincreased over time.32

The sheer scale of the problems that Sunderland confronts despite a success such as Nissan shouldgive us pause for thought. If direct investment of £2 billion, a plant employing 5,000 directly and afurther 7,000 indirectly, cannot revitalise the city, we need to ask whether it can reach the nationalaverage level of GVA, wages and living standards while remaining its current size.


What can be done?

33 Caulkin G, “Keane turns fire on

the WAGs”, The Times, August 15

20087 see www.timesonline.co.uk

/tol/sport/football/premier_league/a

rticle2260422.ece

34 Key Figures for Education,

Skills and Training, Sunderland

(Local Authority) and Newcastle

upon Tyne (Local Authority),

Neighbourhood Statistics, Office of

National Statistics

35 Thus the recently launched

Qashqai, arguably the most British

Nissan ever, was designed at

Nissan Design Europe in

Paddington, London, with techni-

cal development undertaken at

Cranfield, before being built in

Sunderland

Sunderland suffers from very poor economic geography. It is a long way from most places. It isnot somewhere that outsiders consider a desirable place to live. Roy Keane, Sunderland FootballClub’s manager, commented on the difficulty of attracting good players to the Premiership clubdespite offering equivalent wages to clubs in the South: “I find it surprising that geography seems toplay such a big part … Retire at 35 or 36, you can live wherever you bloody well like – London,Monaco, wherever – and any half-decent footballer will be a multimillionaire anyway. Why is theresuch a big attraction with London? It would be different if it was Chelsea, Arsenal or maybeTottenham, but when they go to a smaller club just because it’s in London, then it’s clearly becauseof the shops.”33

In terms of market potential, we have already noted that being on the coast is disadvantageous.Nearby Newcastle is a mixed blessing: it does increase the market potential, but it is also a rival.Indeed, rather like Bradford to Leeds, Sunderland sits in Newcastle’s shadow. Although Sunderlandis the larger city, Newcastle is the better connected. Newcastle has the M1/A1, the mainline railroute, and the airport. The Tyne and Wear metro is also much more extensive in Newcastle than inSunderland, and although the metro connects Sunderland directly to the airport, it takes an hour tocover the 16 miles. And Newcastle is more affluent, with GVA 11 per cent above that of Sunderland.

Those who want to paint a more optimistic future will point to recent innovations. For example,Sunderland now has direct trains from London. But there are only three a day, a tenth of the num-ber to nearby Newcastle. And they are much slower. There are not that many people who wish totravel to Sunderland from London, and therefore the train stops at many intermediate stations topick up additional passengers. The result is that the 08:04 train from London to Sunderland takesthree quarters of an hour longer than the equivalent train to nearby Newcastle. It is slower to takethe direct train to Sunderland than to take the train to Newcastle, walk to the metro stop, and takethe metro to Sunderland. By that time the Newcastle train will have reached Scotland. In terms ofconnectivity to London, Sunderland is not so much south of Newcastle, as north of Berwick. But interms of reaching Edinburgh, Sunderland is as far south as Darlington. In short, the city is isolated.

Sunderland has a severe problem with its skills base, particularly compared to nearby Newcastle.This is not a problem with the schools: Sunderland and Newcastle schools perform equally at GCSElevel. But local people with skills are more likely to leave Sunderland than Newcastle, and outsiderswith skills are more likely to move to Newcastle than to Sunderland. The result is that Newcastle has75 per cent more people with degree level skills than Sunderland, but Sunderland has 15 per centmore people without any qualifications at all.34

Sunderland sits in Newcastle’s shadow, and Newcastle’s better connectivity and path dependencymakes it hard to imagine any way for Sunderland to escape. Indeed, if anything Sunderland has beenlosing high-skilled private sector employment, as we can see even in Nissan’s case. Initially, NissanTechnical Centre Europe was located in Sunderland, but this moved to Cranfield, Bedfordshire, inthe 1990s, to the heart of the British motor industry research zone. Nissan’s design studios are inLondon, again, close to those of other manufacturers, such as Ford. Some of the middle-skilled whitecollar jobs have left Sunderland as well: in 2003, much of the purchasing department relocated againto Cranfield, and in 2005 parts of finance moved to Budapest. This separation of production fromhigher value added jobs is very much in line with global economic geography trends, and means thata large manufacturing plant may not generate highly paid jobs in its vicinity.35

The evidence strongly suggests that there is next to no chance of people in Sunderland having thesame opportunities as the people of Birmingham or Portsmouth, let alone the same chances as thosein London or Oxford. It is time to stop pretending that there is a bright future for Sunderland andask ourselves instead what we need to do to offer people in Sunderland better prospects.

8Expanding London

As we, and others, have noted, London isan economic super-city and, within Britainat least, it stands in a league of its own.This is both a constraint and an opportu-nity: a constraint in that the presence ofLondon in the South means that theNorth-South (or, for that matter, theWales-South) gap is here to stay. As wehave noted, you cannot move JP Morganto Blackburn, or Deutsche Bank toSunderland.36 If you try to do that, theywill contract, and may leave Britain alto-gether. As a result Blackburn, Sunderland,Merthyr Tydfil and so on will never be asrich as London.

But London also offers Britain a greatopportunity and one that is almost uniquein Europe: the opportunity to expand aglobal city capable of generating largenumbers of high quality jobs. There isevery reason to think that London is cur-rently below its optimal size and if the cap-ital expanded by attracting people fromour regeneration cities, then it could trans-form their lives.

We do not know London’s optimal sizeand it is impossible to work it out ex ante.We showed in the last chapter, however,that there are many reasons to believe thatthe optimal size of service-sector cities hasgrown, perhaps considerably in the last 50years. London itself has grown in thattime, both in terms of the number of peo-ple who live there and the number whowork there. Since 1980, when Docklandsbegan, 14 million sq ft of office space havebeen built in the Canary Wharf Estate, thevacancy rate stands at 0.4 per cent andwork on building another 12 million sq ft

is underway.37 Between 1980 and 1998,the lifespan of the London DocklandsDevelopment Corporation, over 24,000homes were built.38 These houses have notstayed empty, nor have they led to emptyhouses elsewhere. On the contrary,London’s population has grown. The newDocklands flats are not occupied by thosewithout work, nor have its residents“stolen jobs” causing decay elsewhere with-in London. The growth of the financialservices sector in London has been rapidenough to supply these people with work,without harming other parts of London, orthe rest of the country.

What has happened since 1980 can hap-pen again. We know that there is excessdemand for housing in London, in thathouse prices in every part of London arehigher than the national average. Indeed,so great is London’s economic potentialthat land values in the lowest value area inLondon – Redbridge – are higher thaneverywhere else outside the South East.39

That is true only because people from therest of the country (and elsewhere) want tomove to London because it is a job-cre-ation city. That is not to deny that Londonhas some severe employment problems ofits own. Its worklessness rate is surprising-ly high. But those problems predate theDocklands expansion, are not caused bythe arrival of migrants from within Britainor elsewhere, and will not be solved by pre-venting their arrival. London’s unemploy-ment problems are generally localised andrequire tailored local solutions to recon-nect certain groups with the labour mar-ket.40 In general the demand for labour

32

36 Ormerod P, Cook W and

Rosewell B, Why Distance

Doesn’t Die: Agglomeration and

its Benefits, GLA Economics,

Working Paper 17, 2006, see

www.london.gov.uk/gla/publicati

ons/economy.jsp

37 Thomas D, “George

Iacobescu: Transport will drive

growth”, Financial Times, 7th

April 2008, see

http://us.ft.com/ftgateway/super

page.ft?news_id=fto0407200811

13057551&page=2

38 Regeneration Statement,

LDDC, March 1998, see

www.lddc-histo-

ry.org.uk/regenstat/index.html

39 Property Market Report July

2007, Valuation Office Agency,

2007, see www.voa.gov.uk/ pub-

lications/property_market_report

/pmr-jul-07/index.htm

40 Willoughby M, “All Hail to

Tottenham”, On Office, 1st May

2008 see

http://onofficemagazine.com/ind

ex.php?option=com_content&vie

w=article&id=124:all_hail_to_tott

enham&catid=104:property;

Martin R and Morrison P (Eds),

Geographies of Labour Market

Inequality, Routledge, 2003

expands with the supply of labour, andallowing London’s population to grow willlead to an increase of jobs.

We believe therefore that part of thesolution to urban regeneration in areas farfrom London is to allow London to grow,so that some of the people living in thoseareas are able to migrate to London. Thiswill directly improve the lives of those whochose to move – they would not move if itdid not – but as we will see later, it also hasthe potential to improve the lives of thosewho do not choose to move.

We noted that it is impossible to predicthow large London should grow. EvenMichael Heseltine could not have foreseenthe scale and extent to which Docklandswould be a success. But even if we do notknow London’s optimal size, land pricesindicate that London’s optimal size is larg-er, probably far larger, than its current size.Given that London is and will remain rich-er and more productive than anywhere elsein Britain, the Government should bestriving to increase the number of peoplewho have the opportunity to live there.After all, doing so will raise average livingstandards, perhaps substantially.

As we have seen, a city’s optimal size isreached when the additional agglomera-tion gains from a rise in population arefully offset by a rise in congestion costs. Assuch, we can increase the optimal size ofLondon by making it easier for knowledgespillovers to occur and by reducing conges-tion. Both of these require improvementsin transport and that means investments inpublic transport. In the last ten years thenumber of people using the LondonUnderground has increased by 31 per cent,far faster than the increase in the size of thenetwork.41 In short, not only has the sup-ply of transport failed to help London togrow to its optimal size, it has failed evento keep pace with the city’s current needs.

The same is true for overground rail.Although there have been substantialimprovements to many long distance train

lines in the postwar era, London com-muters have been neglected. Despiteadvances in technology in every sector ofthe economy, it is no faster to commute toLondon now than it was after the SecondWorld War, and it is actually slower tocommute now than it was in 1971.42 Thereason is simple: traditional cost-benefitanalysis has included only the gains to thetraveller without considering the widereconomic benefits. That is changing:Department of Transport cost-benefitanalysis now includes a significant elementof agglomeration economies. But what isnot changing is the level of investment.Evidence-based policy requires theGovernment not only to perform the cost-benefit analysis correctly, but also to act onthe results of that analysis.

People do not have to live in London towork in London; they can commute fromthe dormitory towns that ring the city.From an agglomeration point of view itmatters little which option is taken since inboth cases the person works in London.From a practical point of view, both areeminently possible.

There is, in fact, a great deal of landwithin London and surrounding areas thatcould be released for housing. By this wedo not mean building over back gardens,although there will be cases where that isappropriate. Instead we refer to land cur-rently restricted for warehousing and otherlow value uses. The current planning sys-tem requires local councils, to state that aparticular piece of land can be used onlyfor housing, or retail, or industrial use.Local authorities are not required, expect-ed or encouraged to use price signals todetermine to which use land should beassigned. Local authorities appear proud ofthe fact that they prevent land uses chang-ing in response to price signals. The coun-cil for the London Borough of Richmondupon Thames, for example, reports: “Withhigh residential values, there is pressure tochange employment sites to residential,


Expanding London

41 Since 1992 passenger trips

by bus and LU increased by 36

per cent. LU = 1 billion trips per

annum, see www.london.gov.uk/

mayor/transport/facts-and-fig-

ures.jsp

42 Beattie J, “Train Journeys

Slower than in the Age of

Steam”, Evening Standard, 3rd

September, 2007

which the Council has generally resisted.”43

Despite the fact that “Richmond’s unem-ployment rate (by claimant count) is oneof the lowest in London”, despite the factthat “48% of local jobs were taken by in-commuters” many of whom arrive by car,resulting in “problems of traffic intrusionand congestion”, and despite the clear andaccepted shortage of housing in the area,only 0.3 per cent of Richmond’s “employ-ment land” is rezoned for housing eachyear.44 Indeed, the council is not even keento rezone industrial land that causes a nui-sance. A reasonable person might expectthat “Where an existing employment useof a site within a predominantly residentialarea is causing detriment to the amenity ofthat area by reason of noise, vibration,smell, fumes, dust etc” the council wouldbe keen to see it replaced by housing inkeeping with the residential area. But no –the sentence continues that “the Councilwill seek improvements, in order to over-come the nuisance caused to residentialneighbours”. The list of criteria that needto be satisfied before the council will allowthe land to be used for housing is exten-sive.45 Nor are we talking about trivialamounts of land being used for low-valueadded activities. Richmond council notesthat it has around 250,000 sq m of ware-house space alone. Even excluding theassociated car parking and access, this is 25hectares of land that could be used forhousing but which is being used for lowvalue, low employment warehousinginstead.

Richmond is not alone by any means,and such policies have been encouraged bycentral government through its PlanningPolicy Guidance notes. The result is thatland of different classifications is pricedvery differently. Although the exact pricesof different types of land vary by place, it isalmost always the case that land for hous-ing is valued more highly than land forindustrial use, particularly in South EastEngland, and particularly in London. This

is, in effect, a hidden subsidy to industrialemployment in affluent areas, sinceemployers are not required to pay anythinglike the market value of the land that theyuse.

The scale of the subsidy to industrial,warehouse and other commercial uses islarge. In Greenwich, for example, industri-al land is typically valued at £2.1 million ahectare, whereas land for houses is worth£6.5 million, and land for flats ormaisonettes £8 million.46 This is a hiddensubsidy of around £5 million a hectare, or£500 a sq m. With 224 hectares of suchland in Greenwich, the planning systemhas reduced the value of land in Greenwichalone by over £1 billion.47

The subsidy is greater still in innerLondon, reaching £700 a sq m inHackney, and £1,300 a sq m inSouthwark. We find high levels of plan-ning subsidy in Britain’s other propertyhotspots: around £575 a sq m inGuildford, £730 in Oxford, £400 inBournemouth and £950 in Cambridge.The subsidy is much lower in areas that arein receipt of regeneration funding: £150 inStoke, £157 in Merthyr Tydfil, and £168in Middlesbrough. It is truly bizarre that acountry wanting at least some jobs migrateto less prosperous areas has devised a plan-ning system that subsidises land-hungryfirms to remain in more prosperous areas.

We therefore make an explicit policyproposal. We should move to a systemwhereby landowners have a prima facieright to convert industrial land into resi-dential land at least in areas of above aver-age employment. It is hard to imagine localpeople objecting to warehousing and facto-ries being replaced by houses. This wouldincrease the amount of housing in Londonand the South East. We estimate thatindustrial land currently accounts for2,643 hectares in London,48 and 7,451hectares in the South East.49 Assuming thathalf of that land was used for housing at astandard suburban housing density of 40

Cities unlimited

34

43 “Local Development

Scheme”, London Borough of

Richmond upon Thames, p 20,

April 2007, see

www.richmond.gov.uk/07revised

finallbrldsfinalversion-2.pdf

44 “Unitary Development Plan”,

Chapter 9: Employment and

Economic Activity, London

Borough of Richmond upon

Thames, Section 9.6, March

2005, see

www.cartoplus.co.uk/richmond/t

ext/09_emp_employment.htm

45 Ibid, section 9.41

46 “Residential Building Land

Figures, Property Market

Report”, July 2007, Valuation

Office Agency, 2007, see

www.voa.gov.uk/publications/pr

operty_market_report/pmr-jul-

07/residen-

tial.htm#outer_london1

47 Jobs and the Local Economy,

Greenwich Council, 2007, see

www.greenwich.gov.uk/NR/rdonl

yres/4ABCB213-F0DD-43E1-

9999-3942B275C318/0/4PT2

jobslocaleconomy.pdf

48 URS Corporation Limited,

London Industrial Land Release

Benchmark, Greater London

Authority, Table A1 Industrial

Floorspace Stock in London,

2007, see

www.london.gov.uk/mayor/plann

ing/industrial-land/

49 “ NLUD-Previously

Developed Land Results”,

National Land Use Database,

2006, see www.nlud.org.uk/

document/gose.xls

houses per hectare, this policy should gen-erate more than 200,000 houses inLondon and the South East without build-ing on any greenfield sites or on any gar-dens. It would also lead to an increase inthe value of land of the order of £25 bil-lion.

We said earlier that more housing in theSouth East would have beneficial effectsfor our regeneration towns. It would leadsome firms currently located in Londonand the South East to move to areas whereland was cheaper, that is, to areas that arein need of regeneration. This will be a sus-tainable market-led approach to regenera-tion. One example might be the PostOffice, which continues to sort consider-able volumes of mail in London and theSouth East. Once it was able to sell itsland for housing at a considerable profit,the Post Office would have a greaterincentive than ever before to sell itsLondon sorting offices for housing. Mailwould then be collected in London, takento (say) Leicester by train, sorted inLeicester, and returned to London on thefirst train the following morning foronward delivery. This change would meanfewer Royal Mail jobs in London andmore in Leicester. But since employmentopportunities are more plentiful inLondon than Leicester, those who losetheir Royal Mail jobs in London would bemore likely to find other work than thatthose who gained the Royal Mail jobs inLeicester. As a result, overall employmentincreases, with the increase occurringentirely in Leicester.

Allowing land currently used for com-mercial businesses to be converted into res-idential housing in London and surround-ing areas will yield a considerable numberof new homes. There can be no guarantee,however, that this will be sufficient toexhaust the possibilities of London as thegenerator of high-quality jobs, and thus assomewhere that can offer real opportuni-ties for people from across the country.

It is likely that Britain faces three choic-es. We can constrain development inLondon and the South East, leading tomuch higher house prices in these areas,and to Britain being poorer overall. We donot support that vision of Britain.Alternatively, we can build more denselyon the land that we have set aside for devel-opment, so that many people live inLondon in very small flats. Again we donot find that vision appealing. That leavesthe third option: a significant expansion inthe amount of land allocated to housing.

No one likes urban sprawl. Residents ofWimbledon may well wish that Surbitonhad never been built. To them Surbiton issimply a place they have to pass through toget to the countryside. No doubt the peo-ple of Wandsworth feel the same aboutWimbledon; no doubt the people living inzone one wish that zones two to six hadnever been built. After all, without subur-bia the people of Chelsea would find itmuch easier to get out of London on aFriday evening and return on a Sundayevening. And yet we know that they do notmind urban sprawl that much. If they didthey would not have chosen to live in themiddle of a major city. In fact, Londonwould not be London without its suburbs,and the quality of jobs that are generatedby the agglomeration economies that canonly come from a city of a substantial size.

There is a benefit to the people ofWimbledon from having Surbiton nearby:Surbiton adds more workers who in turnincrease the agglomeration economies whichin turn increase the equilibrium wage rateenjoyed by everyone including those from


Expanding London

“ Allowing land currently used for commercial businesses

to be converted into residential housing in London and

surrounding areas will yield a considerable number of

new homes ”

Wimbledon. The number of people inLondon is also an important reason, in factthe important reason, why London is such avibrant city. The sheer range of activities –art galleries, opera, and so on – could nottake place if it was smaller. Urban sprawlmay not be desirable in itself, but it generatesoutcomes that are valued by many: goodjobs and a wide variety of activities availableto residents and visitors alike.

A very small increase in London’s urbansprawl would generate a great deal of hous-ing. Someone travelling from centralLondon to the countryside will be travel-ling at about 70 miles an hour by the timethey reach the outskirts of town, whetherby train or on the motorway or trunk Aroad. Let us imagine for the sake of argu-ment that we increase the journey time byone minute. One minute at 70 miles anhour is just over a mile. If we expand thesize of London suburbia by one mile, anduse that additional land for housing then,assuming a density equal to an outerLondon borough such as Kingston, we willhave added approximately 400,000 houses,that is, homes for more than one millionpeople.

Would it really be the end of the world ifthose leaving London travel through subur-bia for an extra minute? Indeed, would itreally be the end of the world if those leav-ing London travelled through suburbia foran extra five minutes? It is hard to believethat this would greatly reduce the standardof living of those who currently live inLondon, but if it allowed a million peopleto move from some of Britain’s least suc-cessful places to London, then it wouldtransform their lives. Are those of us wholive in London really so selfish as to refuseto travel for an extra minute to reach thecountryside, knowing that our selfishnesswas destroying the prospects of prosperityfor a million people?

Nor is house building an environmentalproblem. We know that gardens are someof the most bio diverse habitats in Britain:

they are certainly more bio diverse thanBritish agriculture. The number of birdsliving in a hectare of gardens far exceedsthe number living in the equivalentamount of countryside.50 Of course newsuburbs and new towns have to be devisedin sympathy with their surrounding areas.No one wants acres and acres of matchbox-es devoid of schools and shops. It is rightand proper that we design new housing,whether in new towns or in new suburbs,to provide a high quality of living and lowexternalities. But that can be done just aswell by expanding London or any of itssurrounding towns as it can by buildingelsewhere.

The same arguments apply to expand-ing existing towns and to building newtowns. The only areas to which these argu-ments do not apply are small villages.Those who have chosen to live in small vil-lages have chosen to live in small villagesfor a reason: they like small villages.Adding thousands of new homes to a smallvillage changes the nature of the place dra-matically, and although no one has theright to ensure that their area neverchanges planners should be particularlycareful about changing the nature of aplace so fundamentally.

It is widely known that development isvery unpopular with local people as we cansee from the reaction to the Government’sproposed eco-towns. But it need not be so.When land is released from agriculture andassigned to housing its value soars, often bya factor of 1,000. If that value can be cap-tured by the community it can be used topersuade people that development is in theirinterests. In the immortal, and somewhatcapitalist, words of Aneurin Bevan whenexplaining how he overcame consultants’objections to the National Health Service:we can stuff their mouths with gold.

Various methods have been suggested asto how we can do this. Kate Barker hasproposed a planning gain supplement;51

Alan Evans and Oliver Marc Hartwich

Cities unlimited

36

50 Gaston K and Thompson K,

“Evidence for significance of

biodiversity in gardens”, pre-

sented to the Royal Horticultural

Society Conference, “Gardens:

heaven or hell for wildlife”, 2002

51 Barker K, Barker Review of

Land Use Planning, Final Report

– Recommendations, HM

Treasury, London, 2006

have suggested a tariff-based system, asused in Milton Keynes.52 More ambitious-ly, Tim Leunig has suggested an auctionstyle system, learning from our successfulexperience of the 3G mobile telephoneauction.53 All of these schemes have advan-tages and disadvantages, and in reality allshould be piloted before they are rolled outnationwide. But all share a common fea-ture: they give local councils and localcommunities greater financial incentives tosupport development.

Let us take the case of Guildford to givea sense of scale of the incentives that couldbe offered to local people to support devel-opment. Imagine that we were to add10,000 houses to Guildford. This is clearlya substantial number and under the currentsystem would lead to extensive protests.But each house that is built adds approxi-mately £140,000 in value to the land onwhich it sits. Let us imagine that the com-munity captured £100,000 of that value,with the remainder going on buying theland and providing the infrastructure. Thatmeans that Guildford Borough Councilwould be sitting on a total profit of £1 bil-lion, or around £15,000 per resident. Thusthe council could ask local people whetherthey preferred to be given £15,000 a headand accept 10,000 additional houses, orwhether they preferred to keep Guildford atthe same size and make no windfall gain.

As MPs and local councillors acrossBritain know, local residents do not likedevelopment but nor do they like counciltax. Guildford would only have to allow500 new houses a year to abolish counciltax altogether. It seems likely that the peo-ple of Guildford would trade a 1 per centincrease in housing for a 100 per centreduction in council tax. As Bevan showed60 years ago, stuffing people’s mouths withgold is a very effective way of securingagreement. The difference here is that thegold is created by the process, it does notrequire any rise in taxes elsewhere becausethe process is self-financed.

For these reasons we believe it is morethan realistic to imagine that communi-ties in the South of England, includingbut not limited to London, will welcomeadditional housing as long as they are gainsufficiently from the process. Nobodylikes urban sprawl per se, but as houseprices in Wimbledon show, the disutilityof having Surbiton next door is really verysmall indeed. Wimbledon remains adesirable place to live even though it isnow surrounded by a couple of miles ofsuburbia.

The exact extent to which additionalLondon workers should live in London orother parts of the South East is not a sub-ject for this paper: both have the potentialto generate agglomeration economies andthe exact ratio should be dictated primari-ly by market signals. Nevertheless, we notethat if they are to live outside London thenthe quality of transport into the capital hasto improve. People will not commute forthree hours a day. Faster trains increase therange of places that are within commutingdistance of London. This in turn increasesthe number of places that can be expandedwhile still gaining the agglomerationeconomies that London offers. We need tothink seriously about how the railways canbe improved.54 In the postwar era long-dis-tance trains have been improved no end:the train from London to York orManchester is now much faster than it was60 years ago, but commuter trains typical-ly are not faster. That needs addressingbecause faster trains increase the accessibil-ity of the London labour market.Commuter trains are also too expensive formany. It is quite wrong for governmentpolicy to levy franchise payments thattranslate into higher fares for Londoncommuters. The South West trains fran-chise, for example, requires Stagecoach topay over £1 billion to the Government forthe rights to run commuter trains toWaterloo. That £1 billion is a hidden taxon commuters, and amounts to approxi-


Expanding London

52 Evans A and Hartwich O,

Better Homes, Greener Cities,

Policy Exchange, London, 2006

53 Leunig T, In my Back Yard:

unlocking the planning system,

CentreForum, London, 2007

54 The Billion Passenger

Railway, Lessons from the Past:

Prospects for the future,

Association of Train Operating

Companies, 2008, see

www.atoc-comms.org/admin/

userfiles/Billion%20Passenger%

20Railway%20090408.pdf

mately £2 per return journey. In otherwords, if this tax were removed, the priceof an annual season ticket would fall byapproximately £450. If we are seriousabout making London the engine of thewider UK economy, it is bizarre to be tax-ing train journeys into London.

It is important, however, that Londondoes not suck in all of Britain’s skilledworkers, leaving declining towns only withthose who do not have the skills to move.To some extent this is what is happening atthe moment. House prices are such thatthe only people who find it easy to move toLondon are those with high enough earn-ing potential to allow them to live inLondon. The capital attracts a dispropor-tionate number of graduates.

We need to ensure some degree of bal-ance in migration to London. The groupwho are least likely to migrate to Londonare those in social housing, which is highlylocalised. A Wigan resident cannot ask tobe put on the housing waiting list inWandsworth, nor can a Warrington resi-dent ask for social housing in Woking.Social housing is very much for local peo-ple. Given that people in social housingtend to be lower skilled, this means thatthose with low skills find it much harder tomove areas. As John Hills reported, virtu-ally no one in social housing moves for jobrelated reasons.55

We need to change that. If London is toexpand, then a proportion of the new hous-es have to be reserved not simply for socialhousing, but for social housing for peoplewho do not currently live in the area. If weimagine that people from Blackpool moveto an expanding Bromley, then ifBlackpool’s skills mix is not to change itmust lose the skilled and the unskilled inequal measure. That in turn requires thatBromley offers social housing to peoplefrom Blackpool in sufficient quantities.Social housing makes up around 1 in 5 ofBritain’s housing stock. As a rule of thumb,therefore, 1 in 5 net new houses in growingareas needs to be reserved for social housingtenants from areas whose populations arenot increasing.56 In addition, local authori-ties may want to provide additional socialhousing for local residents.

Requiring an expanding Wandsworth,Woking or Witney to offer significant lev-els of social housing for people fromWalsall, Warrington and Wigan will not bepopular, and it will not happen unless localcouncils are required to do so. If we do notrequire expanding areas to offer socialhousing to people from economicallydeclining areas then we will create a spiralof decline in those places. As now, theskilled will leave, leaving the unskilledtrapped in a town that remains poorlylocated, and has an ever lower skills base.

Cities unlimited

38

55 Hills J, Ends and Means –

The future roles of social hous-

ing in England, CASE report 34,

Centre for the Analysis of Social

Exclusion, 2007

56 Ibid, p 43

9It will not just beLondon

At first sight the existence of agglomera-tion economies seems to imply that we willall end up living in London. That is notthe case. Agglomeration economies willcontinue to increase as London – or anyother city – grows, but the congestion costswill also rise. There comes a point, there-fore, when London’s congestion costs arepushing up the cost of living as fast asagglomeration economies are raisingwages. At that point it will have reached itseconomically optimal size.

We know that London has not yetreached that size because high land valuesindicate that people wish to move toLondon. The same is true for the com-muter towns surrounding the capital. Aswe have said, there is a strong economicrationale for expanding both areas.

Nevertheless, at some point, which isimpossible to predict, congestion costs willrise to a level such that people will nolonger want to move to London. Althoughland in all of Inner London, from Chelseato Tower Hamlets, will always be high, atthat point the value of land for housing atthe edges and around London will nolonger exceed the UK average, since peoplewill not, on average, want to move toLondon. The city will have reached itsoptimal size and we should stop expandingit. The same will be true of the Londoncommuter towns, which may reach theiroptimal sizes before or after the capital,and certainly at different times to eachother.

As we, and others, have noted, Londonstands out as a globally successful city,

capable of creating very large numbers ofvery good jobs.57 But as we have just seen,London cannot grow to accommodateeveryone who might like to live there. Thequestion then is whether Britain can createanother London, a city or group of citiescapable of being globally competitive inhigh value-added sectors.

There can be no certain answer to thisquestion, but there are two ways of work-ing out which cities have the greatestpotential to thrive at the global level. Thefirst is the planner’s way. It is to identify thecriteria necessary to be a global city andthen to see which cities are closest to thosecriteria. The second is the economist’s way,which is to see what land prices are tellingus. As we shall see, both methods give thesame results.

Intelligent planners know that thestrongest foundation for cities with globalreach is that they are based on exceptionalskill levels that cannot easily be replicatedelsewhere. When we look to places such asBoston, Massachusetts, and Silicon Valley,in California, we find that they are associ-ated with some of the world’s best univer-sities: Harvard and MIT in Boston, andStanford in the heart of Silicon Valley. It istherefore sensible to ask ourselves whichare Britain’s best universities outsideLondon.

That Oxford and Cambridge are worldclass universities is well known. But it isworth setting out the research gap betweenOxford and Cambridge on the one handand other (non-London) universities insome detail. All university league tables are


57 Sassen S, The Global City:

New York, Tokyo, London,

Princeton University Press, 2001

crude measures and ours is no exception.The measure that we use here captures theresearch intensity of Britain’s universitiesby looking at the ratio of governmentfunding given for research to that given forteaching. In essence, research funding isawarded according to the quantity of highquality research, while teaching funding isgiven on a student per capita basis. As suchthe ratio of one to the other gives the rela-tive research strength of a university.58

Only six institutions receive more forresearch than for teaching: these are ourstrongest research universities. They consistof four London University colleges – theLondon School of Economics and PoliticalScience (LSE), Imperial, University Collegeand the School of Oriental and AfricanStudies (SOAS) – Oxford and Cambridge.Excluding Oxford, Cambridge and theUniversity of London, the average UKinstitution receives only 23 per cent asmuch funding for research as for teaching,about an eighth as much as Oxbridge. Themedian institution receives only 4 per centas much funding for research as for teach-ing. Even the next highest ranked universi-ty, Southampton, receives only a fractionover half as much funding for research rel-ative to teaching as do Oxford and

Cambridge, while universities such asNottingham and Leeds are on a third ofOxbridge levels. There is, in other words, ahuge gulf between Oxford, Cambridge andLondon on the one hand, and the remain-ing UK universities on the other. That isnot to say that universities likeSouthampton, Nottingham and Leeds arepoor universities in any sense, rather it is tostate that in Oxford and Cambridge Britainhas two of the world’s finest universities,and it is likely that these can form the basisof strong, successful, substantial cities.

The alternative way of deducing whichcities should be expanded is that of theeconomist, which involves looking at landvalues. These give a measure of the relativeextent to which people are trying to moveto a place. Outside of London and immedi-ate commuter towns such as St Albans thehighest land values are found in Oxfordand Cambridge. Again, this tells us thatthese cities could grow very effectively.

It is not possible to predict the optimalsize of these cities. The only way to discov-er that is to expand their size and see whathappens to land values. If they remain high,then it is economically rational to expandthe city further. If they do not, then the cityshould not be expanded further.

Cities unlimited

40

58 We exclude institutions

admitting fewer than 100 under-

graduates a year. London

Business School, the Institute of

Cancer Research, the London

School of Hygiene and Tropical

Medicine, the Courtauld Institute

of Art, and the Institute of

Education all have research to

teaching funding ratios exceed-

ing 100 per cent. All are in

London, so their exclusion does

not change the results

59 Ratio is HEFCE research to

HEFCE teaching funding, 2008-

9. Excludes research council

funding, which is awarded by

individual application.

Universities also receive funding

from other sources, such as

charitable foundations, the EU

and business, which are not

included in these figures.

Excludes institutions with fewer

than 100 undergraduate accept-

ances in 2007-8, Higher

Education Funding 2008-09,

BBC 6th March, 2008 see

http://news.bbc.co.uk/1/hi/in_de

pth/629/629/7279454.stm. The

average figure given here

excludes Oxford, Cambridge

and University of London col-

leges. University sizes from

UCAS, 2007 see

www.ucas.com/about_us/stat_s

ervices/stats_online/data_tables/

abushei/abushei2007/

200%

175%

150%

125%

100%

75%

50%

25%

0%London

School of Economics and Political

Science

University of

Cambridge

Imperial College London

Univeristy of Oxford

Univeristy College London

School of Oriental

and African Studies,

University of London

UK University average

Median UK university

Research-intensive undergraduate-accepting universities as judgedby the ratio of HEFCE research to teaching funding, 2008-9

Figure 4: University to teaching ratio59

It is likely that the optimal size could bequite large indeed. At the start of the 19thcentury Liverpool and Manchester werearound the size of Oxford and Cambridgetoday. Over the course of the centuryLiverpool and Manchester both grew five-fold. It is plausible that since 1945 Oxfordand Cambridge would have grown on asimilar scale were it not for very restrictiveplanning policies. Indeed, given that eco-nomic geography tells us that optimal citysizes have probably increased in the 20thcentury, and that optimal sizes are largestfor high-skill, service-sector based cities, itis more than possible that the optimal sizeof each town exceeds one million people,perhaps by a considerable margin.

In many ways both cities are in good loca-tions for expansion. Both are near major air-ports, with Oxford’s proximity to Heathrowedging out Cambridge’s proximity toStansted. Against that, a population of onemillion or more affluent people in globallyconnected businesses in Cambridge would belikely to lead to a rise in the number of long-haul flights from Stansted. Both are also nearLondon, which again adds to the agglomera-tion economies available to all three cities col-lectively, although the journey betweenOxford and Cambridge is somewhat tedious.

The Government wants three millionhouses to be built in Britain by 2020.60 Thebest approach to determining the rightlocations is to follow land price signals thatcan and will change over time, but it isplausible that the optimal economic loca-tion for these houses is a million in andaround London, a million in Oxford and amillion in Cambridge.

Oxford and Cambridge are not citieswith particularly large regeneration prob-lems, but expanding them can create oppor-tunities for people who live in towns andcities that do. As we noted earlier, this doesrequire greater levels of internal migrationthan Britain has been used to in recentyears, but the levels of migration envisagedare far smaller than those of 19th centuryBritain. Given all the improvements intransport and communications in the lastcentury and a half, that cannot be infeasible.


It will not just be London

60 Homes for the Future: more

affordable, more sustainable,

Housing Green Paper,

Department for Communities

and Local Government, 2007,

see www.communities.gov.uk/

housing/housingsupply/

“ Oxford and Cambridge are not cities with particularly

large regeneration problems, but expanding them can

create opportunities for people who live in towns and

cities that do ”

Box 2: Hastings – connecting the connectables

Britain can effectively be split into two parts, which we term “connectable” and “unconnectable”Britain.

By connectable we mean primarily but not exclusively connectable to London. We know thatLondon is not only the richest place in Britain but that it has a proven track record of raising up thecommunities that surround it. Places such as Guildford and Windsor have high skilled populations oftheir own: they would be affluent wherever they were. But there are many areas, particularly east ofLondon, which do not have particularly highly skilled populations and yet are reasonably successful.

We think, for example, of the Medway towns. In the 1980s Medway lost two major employers:the Royal Naval dockyard and Metal Box. In the 1980s, therefore, Medway had much in commonwith many industrial towns in the North of England, which were losing their manufacturing bases.But Medway today is a reasonably successful place. It is not the most affluent place in Britain by anymeans but nor is it struggling. The reason for Medway’s revival has little to do with Medway, andeverything to do with London. As anyone who has stood on Chatham station in the morning willknow, commuting is a fact of life for many Medway residents. Proximity to London has keptMedway afloat, and prevents it from entering into a spiral of decline.

Cities unlimited

42

61 We thank the Association of

Train Operating Companies for

supplying us with this data.

62 East Croydon to London is

the most popular journey in

Britain, Gatwick to London is the

seventh. There are more than 14

million journeys a year on these

three routes alone, plus millions

more from intermediate stations.

Medway, then, is an area whose connectedness to London has ensured that it remains economi-cally viable despite severe shocks in the 1980s. There are other places that are poor in southeastEngland that could be connected. In our original sample of urban regeneration towns we wouldpoint to Hastings as an example. But it is not the only one: Thanet towns such as Margate are anoth-er example. These are places that can be regenerated. The way to do so is simple: increase their con-nectedness to London. Given that commuters into London overwhelmingly travel by train, the obvi-ous way to increase the connectedness of places in the South East is to improve the quality of trainlinks between them and London. At the moment the direct train from Hastings to London takes anhour and 51 minutes and stops at between 11 and 13 stations. Given that the distance betweenHastings and London is only 62 miles this means that the train is running at an average speed of 34mph. Were trains to Hastings to run at the same speed as trains to Swindon, the train journey wouldbe cut to 47 minutes, similar to the tube ride from Shepherd’s Bush to Canary Wharf.

But Swindon is lucky in two senses. First, the train line out of Paddington was exceptionally wellengineered. Very few trains out of London manage the speeds of trains to Swindon. Second, trainsto Swindon are fast because Swindon is on the mainline to Bristol, Exeter and Cardiff. No matterhow well we reconfigured the railway system, Hastings will never be on the way to anywhere: it isthe end of the line. This means that the quality of the railway that is viable between London andSwindon is not viable between London and Hastings, and never will be. Being on the coast is a majordisadvantage today.

Realistically therefore, Hastings will never be as well-connected to London as is Swindon. In allprobability, Hastings will never be as rich as Swindon. The challenge is to cascade wealth fromLondon to places that are well-connected around London and then on from those places to Hastings.

There are two obvious candidate intermediate places: Brighton and Dover. Brighton is in manyways a successful city. It is well connected to London, and to Gatwick airport. But it is surprisinglybadly connected to the other towns along the South Coast, in either direction. Not only is there nomotorway link there is no continuous dual carriageway either. In fact, fewer than 7 of the 36 milesbetween Brighton and Hastings are dual carriageway, and even that comes in two parts, and is punc-tuated by roundabouts. It is even worse going east from Hastings: there is no dual carriageway on theroute between Hastings and Dover until you reach the A20 in Folkestone. Given the poor quality ofthe roads, it is not surprising that the RAC predict that it will take an hour to get from Hastings toBrighton, and over an hour and a quarter to get from Hastings to Dover, despite the fact that thesejourneys are only 36 and 46 miles respectively. The rail connections are even worse: it typically takesone hour and five minutes from Brighton and one hour, 40 minutes from Dover, with a change atAshford.

Nor are the connections going north from Hastings any better: there are no dual carriagewaysbetween Hastings and Ashford or between Hastings and Maidstone and only three miles of dual car-riageway between Hastings and Tunbridge Wells; the train lines on these routes, where they exist, arealso slow. Hastings, then, is not connected, but it could be. It is perfectly possible to imagine the dualcarriageway between Dover and Portsmouth, which would connect towns such as Hastings to theChannel Tunnel and the ferries to Europe on the one hand, and begin to create agglomerationeconomies between the otherwise rather isolated towns of the English South Coast on the other.

It would be particularly beneficial for Hastings if Brighton were to grow. Perhaps surprisingly,Brighton already has trains that are fast enough to make the journey from Brighton to London thethird busiest in Britain.61 That line is sufficiently well used (even more so between Gatwick, EastCroydon and London) that increasing the speed of trains on that line is likely to pass cost-benefitanalysis.62 This in turn would allow Brighton to grow and create a South Coast agglomeration area,to the benefit of Hastings and other relatively poor coastal towns, which could then orient themselvestowards Brighton. For Brighton the problems associated with being on the coast are trumped by itsconnectability to London, and it has the potential to create regional spillovers to other towns in thearea.

10Economic geographyoutside of theSouth East

The implications of economic geographyfor the South and particularly the SouthEast are clear. Britain will be unambigu-ously richer if we allow more people to livein London and its hinterland. In addition,Oxford and Cambridge should be theprime cities to see significant but not, byhistorical standards, unprecedented expan-sion.

The lessons for other parts of Britain aremore ambiguous. Nowhere will be as richas London within the foreseeable future. Itis also likely that an expanded Oxford andCambridge would be richer than every-where else in Britain. But that does notmean that everywhere else would be equal.On the contrary, the two important factorsthat we have pinpointed – location andscale – remain powerful indicators of theirpossibilities.

We noted that places with strong marketpotential are likely to be richer than otherareas. Market potential derives from thecharacteristics of the place itself and theplaces to which it is well connected. Scaleis part of a good location. Manchester isstronger for being larger. So is Leeds.Conversely, towns such as Burnley orMerthyr Tydfil suffer from being relativelysmall.

The evidence suggests that the optimalsize of cities which concentrate on the serv-ice sector is larger than for those whichconcentrate on manufacturing. We can seethis at the moment by looking at the bank,Northern Rock. By the standards of North

East England, Newcastle is a large place.But Northern Rock dominates Newcastle’sfinancial services sector, to the extent thatwere Northern Rock to fail, Newcastle’sfinancial services would be all butdestroyed. In contrast, were a similar sizeLondon financial institution to run intotrouble, it would make little difference toLondon’s status as a financial centre, andthe people working in that bank would belikely to find other banking jobs easily.Newcastle, in short, is small enough to bevulnerable to the collapse of a single firm;its economy is not as robust as a muchlarger centre such as London.

Manchester is the third largest metro-politan area in the UK after London andBirmingham . Although income varieswidely in different parts of the city,Manchester as a whole is much richer thannearby towns such as Blackburn, Oldhamand Rochdale – although there are affluentcommuter towns nearby, such asMacclesfield, which cannot be consideredseparate from the Manchester economy.63

Manchester is successful in part because itis large. It is also relatively well connected,with 407 trains during the working day,including a relatively fast and frequentservice to London and Glasgow. The air-port also offers a reasonable range of desti-nations and flights. The same is true ofLeeds. It too has good train links, particu-larly to the South, with over 445 trainsduring the working day, an airport and rea-sonable connections to Manchester. It is


63 See, for example, Inland

Revenue statistics:

www.hmrc.gov.uk/stats/income_

distribution/table3-15-feb08.xls

Cities unlimited

44

richer than its neighbours, large (Bradford)and small (Batley), although as withManchester there are affluent commutertowns nearby (Shipley).64

If we look at the land price data thereseems a strong argument, akin to the casewe made for expanding London, Oxfordand Cambridge, for expanding Manchesterand, to a lesser extent, Leeds. Land for res-idential use in Manchester is currentlyworth around £5.75 million per hectare.Given that industrial land in Manchester isvalued at only a tenth of that amount perhectare, there must be a very good case forreallocating at least some of that land forhousing. But the case for expandingManchester as a whole is dramaticallyweakened by the value of land in townsaround Manchester. Land in Rochdale, forexample, just ten miles from centralManchester, is worth only £1.9 million perhectare, less than the national average.65 Itis hard to see the wider Manchester area asshort of housing. It would be odd toexpand the city dramatically when there isso little demand for housing in nearbytowns.

But size cannot explain everything:Liverpool, Bradford and Sunderland are alllarge; indeed Sunderland is larger thannearby Newcastle. But none of them is rich– quite the reverse. These are cities with aparticular problem: although large, theyare not the hub cities for their areas.Liverpool sits to an extent in Manchester’sshadow, and Bradford and Sunderland areunambiguously in the shadows of Leedsand Newcastle. Manchester, Leeds andNewcastle are clearly the premier cities oftheir areas. They are the wealthiest, bestconnected, and have the highest skills base.It is very hard for nearby rivals to compete.

The same is true in the South East.Places such as Guildford and Romford,Hertford and Watford cannot competewith London. But they do not have tobecause London is a strong enough econo-my to allow them a new role – that of com-

muter towns – complementary to London,not rivals to it. Sadly for Liverpool,Bradford and Sunderland their largerneighbours, Manchester, Leeds andNewcastle, are not as powerful economi-cally as London, which is also much largerrelative to its commuter towns.

The three regional hubs will not lift thesurrounding towns in the way that Londonlifts what would otherwise be relatively lessprosperous areas such as North Kent andSouth Essex. The northern regional hubs arenot as large as London and so have muchless economic muscle. But there are otherreasons as well. Londoners on good wagesmove out of London to the surroundingtowns because they cannot afford the spacethat they would like in the city itself.Housing in London is expensive, whetherwe are talking about flats in the centre or theinner suburbs such as Islington andFulham, or houses in outer suburbs such asWimbledon and Hampstead. Even peoplewith high incomes are forced to leaveLondon and so take their prosperity withthem. They move out along the train lines,to Woking, Reigate and Hemel Hempstead.Those with more modest salaries oftenmove out along the slower train lines,accepting a worse quality service inexchange for cheaper housing. Althoughnot as affluent as those commuting from thethree towns just mentioned, those travellingin each day from North Kent and SouthEssex earn more than they would be able todo locally and as such they increase theprosperity of these areas.

The financial pressure to leave the north-ern regional hubs is much lower. Movingfrom the borough of Camden to Romfordreduces the purchase price of a pre-1919terraced house by £331,000, and movingfrom Camden to Gillingham, in NorthKent’s commuter belt, reduces the price by£386,000. But moving from Manchester toRochdale saves only £178,000, eventhough Rochdale is significantly cheaperthan anywhere else in the region. The gain

64 See, for example Inland

Revenue statistics:

www.hmrc.gov.uk/stats/income_

distribution/table3-15-feb08.xls

65 Valuation Office Agency, op cit

from leaving Leeds or Newcastle is smallerstill. Furthermore, the difference in pricebetween, say, Rochdale and Macclesfield, at£85,000, is much lower than betweenGillingham and Guildford, at £140,000.As a result, people with reasonable jobs inManchester are more likely to be able toafford to live in Manchester orMacclesfield, whereas their counterpartsmight well be priced out of London andGuildford and therefore take their incomesto Gillingham. Gillingham benefits fromthe high price of property elsewhere: itmakes the town attractive to people whomight not otherwise consider it.66

In addition, the transport connectionsto the centre of Manchester, Leeds andNewcastle from surrounding smaller townsare not as good as those into London.Although Londoners grumble about theirtrains, tubes and buses, they are, by UKstandards, incomparable. London is a citydesigned around commuters in a way thatis true of no other city in Britain.

Rather than high skill, high pay workersleaving these regional hubs to live but notwork in their surrounding lower cost areas,the regional hub cities attract the highskilled workers out of the surroundinglower cost areas. Rather than gaining afflu-ent commuters, Liverpool, Bradford andSunderland are more likely to lose theirown local talent, attracted by the quality oflife in the nearby hub. It is very hard to seeconditions in Liverpool and particularlyBradford and Sunderland improving rela-tive to their local rivals, or the nationalaverage, in any conceivable time frame.

The problem for many of our regenera-tion cities is twofold: they are too small tobe an attractive destination in their ownright and they are relatively poorly con-nected to other places that are attractivedestinations. None of these cities is goingto grow dramatically of its own accord, andit is unlikely that any of them are going toattract substantial numbers of highlyskilled workers any time soon.

“Path dependency” or “chicken-and-egg” syndrome is at work: without skilledworkers, firms offering well paid skilledjobs will not move to these towns. Butwithout the jobs, skilled people will notmove there either, resulting in the mainte-nance of the low skilled, low wage, highunemployment status quo.

The Government is committed to dis-persing the Civil Service, as recommendedin the Gershon review,67 and the LiberalDemocrats have even argued for theTreasury to be moved to Liverpool.68

Sometimes dispersion works – theTeachers’ Pension Agency moved toDarlington in 1967. But the Governmenthas been much less successful in relocatinghighly skilled, and in particular, more spe-cialised jobs out of London. We can seethis at the moment with the plans of theOffice of National Statistics (ONS) totransfer the production of statistics fromits London office to Newport in Wales.The National Statistician, Karen Dunnellexpects only 15 per cent of staff to move toNewport.69 One ONS statistician told usthat 15 per cent was an ambitious target,and predicted that the actual figure wouldbe 2 to 3 per. One ONS worker told TheGuardian: “No one wants to go toNewport.”70

The Bank of England has warned MPsthat: “The relocation programme posesserious risks to the maintenance of thequality of macroeconomic data.” It said: “Ifsubstantial numbers of ONS staff areunwilling to relocate, the loss of skilledindividuals could have a severe impact on arange of statistics.”71 The plan also demon-strates why relying on the relocation ofskilled workers is not a sensible way toregenerate depressed areas: they do notmove.

For private sector companies such relo-cation is simply not an option – if they tryit their rivals will poach their staff. Thepublic sector can do it – the ONS cannotgo bankrupt because it has no rivals. But


Economic geography outside of the South East

66 Ibid

67 Gershon P, Releasing

Resources to the Front Line,

Independent Review of Public

Sector Efficiency, HM Treasury,

July 2004, see www.hm-treas-

ury.gov.uk/spending_review/spe

nd_sr04/associated_documents/

spending_sr04_efficiency.cfm

68 The Lib Dem Plan for

Slimming Government, Liberal

Democrats, 2008, see

www.libdems.org.uk/government

/issues/

69 Padgham J, “City Divided

over Prospect for Summer Rate

Rises 5.5% - but what then?”,

The Independent, 10th May

2007, see

www.independent.co.uk/news/b

usiness/news/city-divided-over-

prospect-for-summer-rate-rises-

55--but-what-then-448211.html

70 Seager A, “ONS London Staff

Seek New Jobs rather than go

to Newport”, The Guardian, 2th

January 2007, see

www.guardian.co.uk/business/2

007/jan/26/interestrates

71 “Stats staff 'quitting' over

move”, BBC News, 14/05/07,

see http://news.bbc.co.uk/

1/hi/wales/south_east/6653771.stm

there is a cost – so worried is the Bank ofEngland about the quality of ONS statis-tics in the next few years that it hasincreased the range of non-ONS data thatit uses in its modelling.72 That is a direct, ifhidden, cost to the decision to move theONS. Data from other sources cannot be aperfect substitute for ONS data. The Bankof England, and all other decision-makers,public and private, may well be using lowerquality data in the future. To move theONS offices to Newport without manag-ing to move the staff is a tremendous riskto the economy; it may yet prove to be themost expensive and least successful piece ofurban regeneration ever seen.

This does not mean that there is nofuture for high skilled work outside majorcities: we know that there are many firmsoffering high skilled, highly paid, work inall sorts of places. In many cases the loca-tions are essentially serendipitous. Forexample, the town of St Asaph in NorthWales is home to a cluster of 35 firms pro-ducing military and high-technologyoptics. This grew originally from thePilkington glassworks, but in recent yearsboth Pilkington and other firms in the areahave moved much further up the valuechain. Pilkington Space technology, forexample, produces around half the glassused by the global satellite industry. It isalso an area in which new firms and newtalent are nurtured. Technium OpTIC, aresearch incubation centre, has recentlywon the European Commission’s RegioStarsAward for 2008 for the quality of itsresearch and development.

This example shows that even fairlyremote places in Britain can sustain high-skill sectors. But we need to be careful: itis one thing to sustain an already existingsector, quite another to create it fromscratch. Existing firms in existing loca-tions have only to attract a handful of newstaff each year, often younger people withfewer ties. Relocating firms, in contrast,have to persuade existing staff to move.That is much harder – their partners havejobs, their children are happy at school,and so on. Had the ONS always been inNewport, the Welsh town would probablywork as a location for ONS, just asCheltenham works for GCHQ. It is theact of moving, not the location itself, thatis problematic.73

More generally, although London,Oxford and Cambridge are by far themost research-intensive universities, thereare a number of high quality research uni-versities across Britain. In some cases,such as Southampton, these play animportant role in the local economyalready. In others, such as York, there isthe potential for the university to drive asignificant expansion of the town, in theway that we have suggested could happenin Oxford and Cambridge. It is likely toprove easier to create significant new areasof high skilled work associated withBritain’s top research universities than it isin areas that are currently most in need ofsuch jobs. Migration within a region, aswell as towards the South East, has to bepart of the answer to the challenge ofregeneration.

Cities unlimited

46

72 Giles C, “Welsh transfer a

'risk' to statistics, says Bank”,

Financial Times, 10th May 2007,

see www.ft.com/cms/s/0/

f40355bc-fe92-11db-bdc7-

000b5df10621.html

73 Even so, locating in an area

where partners have difficulty

following their careers creates

two problems. First, the firm or

government department in ques-

tion may have difficulties in

attracting staff to a place that

offers limited opportunities for

their partner, and secondly, if

they do manage to attract the

staff, the skills of the partner

may not be fully used, to the

detriment of the person con-

cerned, and the economy as a

whole

11Setting cities free

Our international research, set out inSuccess and the city, has demonstrated thepower of localism. Places such asVancouver, the Randstad cities of theNetherlands, Hong Kong and the Ruhrcities of western Germany, have all usedtheir autonomy to good effect. That is notto say that by unshackling British cities wewill release a huge tide of civic entrepre-neurship: as the cases of Warsaw and Łódźdemonstrate, freedom is not enough, ittakes expertise as well.

There is much expertise in the Britishregeneration industry, whether in local gov-ernment, regional agencies, central govern-ment or the private sector. But much of it ismisdirected. When talking to peopleinvolved in regeneration at the grassroots, wewere struck by how often they told us of thetime and effort that they put not into regen-eration, but into keeping central govern-ment happy. Central government controlsthe money, and so it is to central governmentthat they look, it is central government thatis their judge, their jury and their hangman.

This is not how it should be. It matterslittle whether central government thinks ascheme has been a success; what matters iswhether local people in the place con-cerned think the scheme has been a suc-cess. There is something of an unholyalliance between centralisers and tech-nocrats in Britain. The centralisers believethat only they have the expertise to deviseplans and only they can evaluate what hasworked. The technocrats believe that localpeople cannot distinguish whether ascheme is successful because of its ownmerits or has just been lucky.

This is wrong. Of course there is a rolefor experts, but they are there to informnot to decide. When people feel disem-powered they stop thinking for themselves.In fact they tend to become reactionaryand antagonised by imposed change. Theyreduce their levels of engagement with civilsociety, and this is worrying on both aneconomic and societal level.74 As we shallsee, there is every reason to think that thismakes regeneration less likely.

We believe that there are four reasonswhy localism is likely to improve regenera-tion outcomes. First, it will lead to diversi-ty, and diversity generates evidence, whichis the cornerstone of good policymaking.Secondly, it will lead to solutions beingmore closely tailored to the wishes andrequirements of local areas. Thirdly, it willallow the generation of new ideas fromwithin the community. Fourthly, there isgood evidence that when people areinvolved in setting their own priorities andfeel that they have a genuine input into theway their area is run, they are happier.Since people’s happiness is the ultimategoal of any society, this is of itself a success-ful outcome.

We concentrate on the first of these fac-tors here, before returning to the others.That devolving power leads to diversity isself-evident, and we can already see thathappening as devolution takes hold and isextended. Labour won the first elections inall three devolved regions – Scotland,Wales (both in coalition with the LiberalDemocrats) and London, where the initial-ly independent mayor, Ken Livingstone,was Labour in all but name.75 The extent to


74 Rodríguez-Pose A and

Storper M, “Better Rules or

Stronger Communities? On the

Social Foundations of

Institutional Change and Its

Economic Effects”, Journal of

Economic Geography, 82(1):

1–25, 2006

75 Northern Ireland has a very

different system, in which British

mainland parties do not mean-

ingfully contest elections. For

that reason we do not consider

Northern Ireland here

which decision-making was different frompriorities set nationally was limited by thesame party being in power everywhere but,even so, variations could be seen: those inpower in all three areas were to the Left ofLabour at Westminster. Since then we haveseen Labour lose ground in all devolvedareas, and lose control of Scotland andLondon altogether.

All those who believe in devolutionshould celebrate the election of non-Labour governments in different parts ofthe UK, not because they are anti-Labourbut because they are pro-diversity. Havingdifferent parties in charge in differentplaces increases the likelihood of differentpolicies being put into effect. Differentpolicies generate evidence about whatworks and what does not, which leads tomore effective government in the long run.

We can see this most clearly in Scotland,both because Labour is no longer part ofthe executive there at all and becauseScotland has greater constitutional powersto change policy than other devolved areas.An example is the plan of the minorityScottish National Party Government toreplace council tax with a local income tax.This has been strongly opposed by theopposition Labour and Conservative par-ties. Wendy Alexander, the ScottishLabour leader, has described it as an“unworkable Scottish jobs tax”.76

Supporters argue that local income taxesexist in many places in the world, includ-ing some American states. It is not for us tosay who is right and who is wrong, becausewe do not, and cannot, know. The onlyway to find out is to try it. If Scotland triesit we will all find out whether it works.

From a social science point of view,devolution is like a pilot scheme with con-sent. The poll tax was first tried out inScotland, but that pilot was of little use. Itwas too short in duration for central gov-ernment to be able to work out the effects,and it was unpopular because it wasimposed, irrespective of any merits that itmay have had. In contrast, while localincome tax is unpopular with two ofScotland’s main parties, it is supported bytwo others. If it is implemented it will nothave been imposed from outside and com-pliance is likely to be high. It will be per-fectly possible for others to assess its advan-tages and disadvantages and to learn fromScotland’s experience. Scotland’s decisionto do something differently benefits every-where else.

We saw this at work in the Ruhr and inthe Netherlands. Amsterdam concentrateson diversifying housing tenure and recruit-ing service-based foreign investment whileRotterdam prefers to nurture its culturalsector and maintain manufacturing andlogistics jobs associated with the port. Inpart these choices reflect the fact that theyhave different strengths and weaknesses,but both cities have similarities and canand do learn from each other.

All areas should want every other area tobe given power over regeneration strategybecause the natural variation in policy willgenerate evidence about what works andwhat does not work. That is to the benefitof all.

Our proposal is simple. Every singlenational regeneration programme wouldbe wound down and the money distrib-uted according to need. Nor would werestrict ourselves to targeted regenerationfunding. There would be a strong case forincluding government spending designedto get people back to work, such as theNew Deal funding streams. The Govern -ment should identify items of local expen-diture that are currently financed from thecentre and return them to the lowest level

Cities unlimited

48

76 Gardham M, “SNP's Local

Income Tax Plans Are

Unworkable, Blasts Wendy

Alexander”, The Daily Record,

12th April 2008, see:

www.dailyrecord.co.uk/news/sco

ttish-news/2008/03/12/snp-s-

local-income-tax-plans-are-

unworkable-blasts-wendy-

alexander-86908-20348136/

“ Our proposal is simple. Every single national regeneration

programme would be wound down and the money

distributed according to need ”

of government that is possible. Thus manylocal train lines would be handed to localauthorities. Where they are loss-making,transitional funding relief would be given.

The formula would be straightforward,relying almost exclusively on the averageincome of people in the area, taking intoaccount their ages.77 There would need tobe transitional spending to avoid hugewinners and losers, but the aim would beto limit its duration. It would also beimportant to identify the reasons whysome areas have traditionally receivedhigher levels of funding than their appar-ent needs warrant in order to adjust futurespending.

Thus, it is inevitable that central govern-ment funding to restore former miningareas will continue because disused minesare extremely expensive to remove.Similarly, were Britain to abolish nuclearpower, the cost of decontaminating theland on which Sellafield is built could notreasonably be restricted to the relativelysmall number of people living close by. Insuch cases the company concerned shouldto pay to restore the land when they havefinished using it. It is, of course, hard forcouncils to persuade firms, particularlybankrupt firms, to do this when they leavean area. In future they should instead begiven the right to require that firms post abond with them at the outset to cover thecosts of restoration when they leave.

This payment would be part of thereplacement for payments made duringplanning permission under section 106 ofthe Town and Country Planning Act 1990,and would give councils a greater incentiveto permit development, knowing that thecosts of decontaminating land would be cov-ered in the long run. Without the need topay to clean up derelict land, central govern-ment could reduce the rate of corporationtax. Firms that inflict less damage shouldgain, while those that contaminate land

should lose. In any case, all firms will pay thetrue costs that they impose on society.

Continued funding to remove danger-ous mine workings should be seen as a spe-cial case. All areas will have some contam-inated land, and all will have at least a lit-tle bit of very contaminated land. An areawould only be eligible for direct govern-ment funding if it had an especially largeamount of such land. Former coalminingdistricts would be the obvious beneficiar-ies, though everywhere would have theright to put forward a case.

Areas would not be eligible for additionalfunding simply because they are remote.People are free to choose where they live, andit is almost invariably better for individualsto meet the costs of their own decisionsrather than society as a whole. Remote areasgenerally have more limited access to broad-band, for example, but those living inBrixton have to pay more for their houseinsurance. It is not the job of governments toequalise the cost of every item across thecountry. As countries like Ghana foundwhen they tried this, it is a disaster.78

This new funding stream would nothave strings attached. Local councils wouldnot be answerable to central governmentfor how they spent it; instead they wouldbe accountable to local voters through theballot box. It is called democracy, and asChurchill said, it is the worst system exceptfor every other one that has been tried.79

Funding regeneration in this mannerwould require the abolition of all grants toorganisations such as regional developmentagencies (RDAs). This does not mean,however, that RDAs would be abolished.They have the potential to act as advisersto local councils and to implement poli-cies on their behalf. Local councils ratherthan central government would becometheir customer. They will survive and pros-per if – and only if – local councils believethat they offer value for money.


Setting cities free

77 This will take into account the

fact that a 7 year old having no

income is different to a 37 year

old having no income, and dif-

ferent again to a 77 year old with

no income

78 Rimmer D, Staying Poor:

Ghana’s Political Economy,

1950-1990, Pergamon Press,

1992

79 “Many forms of Government

have been tried and will be tried

in this world of sin and woe. No

one pretends that democracy is

perfect or all-wise. Indeed, it has

been said that democracy is the

worst form of government

except all those other forms that

have been tried from time to

time.” Speech in the House of

Commons, The Official Report,

House of Commons (5th Series),

11th November 1947, vol 444,

cc 206–07

12What would cities do?

We have been very clear that it is not therole of London-based consultants to tellindividual towns and cities what to do.This chapter does not seek to offer a blue-print, and certainly does not claim to offersolutions that will work in every city in thecountry. Instead it seeks to report back onwhat has worked elsewhere, and on inter-esting ideas that people around Britainhave told us that they would like to try ifonly the rules allowed. We do not propose,therefore, that local councils should do allor any of the following, only that these areideas that they should be able to consider.In all cases, local councils would be spend-ing their own money, which they wouldhave complete freedom to use as they sawfit. And they would be completely respon-sible to their electors, who would judgethem on what they have achieved.

The first thing that all councils shoulddo is to make a realistic and hard-headedassessment of their area’s position and itspotential – its strengths, weaknesses,opportunities and threats faced. At themoment, local councils like to portraythemselves as being in a pitiful state, butone that would be transformed if only cen-tral government were to fund a businesspark, high-speed train, new town centre orcultural quarter.

And whatever settlement is received,local authorities have every incentive to tellresidents that it is so inadequate that noth-ing can be expected of them. Councilsbehave like this not because they are dis-honest, but because they are almost entire-ly dependent on central governmentgrants.80 Today many councils have regen-

eration objectives that bear little resem-blance to any possible reality. It would beinvidious of us to name particular towns,but we have been left in no doubt thatcouncil leaderships believe that it is unac-ceptable for them to paint anything otherthan a bright, indeed often absurdlybright, picture of the future. Equally, con-sultants have told us tales of writing regen-eration visions that they know cannot bedelivered. As one remarked, the council isthe customer, and we all know that thecustomer is always right, don’t we?

We believe that giving councils greaterresponsibilities will make them moreresponsible. We can see this already in thecase of the devolved assemblies. The WelshAssembly Government has, for example,created a genuinely high-quality economicresearch advisory panel. It includes not justthe great and good of Wales, such as GarelRhys and Andrew Henley, but also drawson expertise from further afield throughpeople such as the former director of theInstitute of Public Policy Research, GerryHoltham, and ex-monetary policy com-mittee member Stephen Nickell – peoplewho will say what they think needs saying,rather than what an audience wants tohear. Thus Holtham’s 2002 paper for theInstitute of Welsh Affairs “finds a glimmerof hope for Wales amid generally grim tid-ings”.81

The panel have produced interestingreports, and offer realistic advice. Mostrecently, they have asked whether it is real-istic for Wales to try to capture any of thehigh value-added industries currentlybeing attracted to the South East, or

50

80 A good example of this is the

introduction of UK-wide free bus

travel for pensioners. The

Government has assessed how

much it thinks the policy will

cost councils, but councils have

said it will cost more. They may

be right or wrong, but the sys-

tem creates an incentive for

councils to claim this however

generous the settlement is.

“Help hard pressed council tax-

payers, town halls urge

Chancellor”, Local Government

Association, press release, 12th

April 2008, see www.lga.gov.uk/

lga/core/page.do?pageId=329018

81 Holtham G, “Debate: Global

Trends”, Institute of Welsh

Affairs, 2002, see

www.iwa.org.uk/publications/pdf

s/DebateGH1.pdf

“whether it would be more productive totry to identify other activities priced out ofthe South East of England and try toattract them to Wales”.82 These are sensiblequestions, and the willingness of the WelshAssembly Government to ask themdemonstrates that devolving power createsgreater responsibility among policymakers.

Having come up with a realistic assess-ment, councils then need to think abouthow to spend their money. Some of thepotential policies that we set out here willbe very familiar to those in the regenera-tion industry, but others will be morenovel. We do not claim that the followinglist is comprehensive.

Traditional regeneration policiesSupport local industryA large proportion of local regenerationfunding has traditionally been given to firmslocated in areas of high unemployment.That could continue. It would be perfectlylegitimate, for example, for Sunderlandcouncil to decide to spend £8 million of itsregeneration money subsidising Nissan tobuild a paint shop. Equally, it would be legit-imate to use that money for other purposesif the council believes that, without thegrant, Nissan’s cost level would still be suffi-ciently favourable to ensure that it stayedin the area. The Nissan plant means moreto Sunderland than to anywhere else.Sunder land knows more about the plantthan anywhere else. Sunderland can makethe decision better than anyone else. Moregenerally, towns and cities could decide toseek out inward investors, as Łódź did withSiemens and Essen with ThyssenKrupp.There are useful European Union rulesdesigned to prevent wasteful counter-bidsbetween areas for regeneration funds.83

Improve business conditions more generallyAs well as funding specific businesses, itwould be legitimate for councils to spendmore on improving conditions for business

as a whole. The most obvious way to dothis would be by providing business parks.Part of the reason for the North Wales mil-itary optics cluster at St Asaph was the pro-vision of a business park suited to the sec-tor. Similarly, councils could offer adviceservices to existing small and medium-sizefirms. Again, there are laws to prevent suchactivities being too heavily loss making,but councils could still decide that this wasa sensible way to spend their money.Decontaminating and restoring land alsofalls into this category, as does assemblinglarge sites in areas of fragmented land own-ership. All these policies have been imple-mented successfully elsewhere – the Ruhr’sTriple-Z and Hong Kong’s record of suc-cessful land assembly stand out as primeexamples.

Improve infrastructureMany areas of Britain have relatively poorinfrastructure connections – Hastings is sonear to many places and yet so far from allof them. Distance is not the problem,accessibility is. It is not alone in being inthis position, and there are many otherplaces that might wish to improve localinfrastructure. If Worksop wanted a dualcarriageway connection to the M1 and A1,it should have the right to upgrade theA619 or A57. Hull might decide that a fastroad or rail route to York offered a goodreturn for its money. Both cases wouldrequire co-operation between the councilsaffected. Vancouver is using the forthcom-ing 2010 Winter Olympics to gain fund-ing from national and provincial govern-ments for transport improvements such asa high-speed rail link from downtown tothe airport.84

Improving town centresThere is a long history of physical regener-ation in urban economic regeneration,which as we noted in Cities limited reachedits apogee in the Rogers Report, Towards anUrban Renaissance.85 It would be perfectly


What would cities do?

82 Panel Ymgynghorol ar

Ymchwil Economaidd/Economic

Research Advisory Panel, Report

to the Welsh Assembly

Government, 2007, see

http://new.wales.gov.uk/firstmini

ster/publications/economicresea

rch/advisory/1971767/ERAPnov0

7?lang=en, p2, para 2.3

83 It would, however, be illegal

under EU law for anywhere else

in the UK to try to attract Nissan

away from Sunderland by offer-

ing a larger grant. This is very

useful in this context because it

ensures that Nissan cannot

expropriate large amounts of

regeneration money by persuad-

ing towns to enter a bidding war


Success and the city, Policy


85 Urban Task Force, Towards

an Urban Renaissance: The final

report of the Urban Task Force,

Chaired by Lord Rogers of

Riverside, DETR, 1999

legitimate for councils to decide that anattractive town centre would be an effectivecontribution to regeneration. Again, we seeevidence of this elsewhere – Vancouver hasused its zoning plan to ensure that thedowntown area survives intact as a mixed-use area, with heritage areas preservedusing density trading credits.86

Organise local job clubsA city should also be allowed to put itsregeneration money into placement servic-es to get people into work. There is evi-dence that the New Deal has had someeffect and councils should have the right tofund schemes along the same lines. Theyshould have the right to hire state, volun-tary or private companies to do this: theyare answerable to their electorates and notto any particular government’s ideology.

Less traditional regeneration policiesWe stated that councils would have carteblanche in deciding what to do with theregeneration money that they are given.There are any number of policies that acouncil might feel appropriate that wouldbe unlikely to be allowed under the currentsystem. We set out a number of them here.We do not claim either to have listed themall – one of the points of devolving power,after all, is to allow new ideas to be gener-ated – and nor do we claim that all of theseideas should be tried. We simply put themforward as ideas. In each case, these poli-cies are likely to be more attractive in atown that believes that its biggest problemis retaining and attracting a skilled work-force.

Organise national job searchesWe have already said that local councilsshould be allowed to spend their moneytrying to match local workers with localvacancies. But we go further. If councilsbelieve that many of the people in theirarea will have better life chances elsewhere,

they should be allowed to assist in nation-al job searches. They could identify areasthat are either short of labour in general, orof particular types of labour, and invitefirms from those areas to visit them for jobfairs. They could fund visits to otherplaces, so that local people could get a bet-ter sense of what opportunities are avail-able elsewhere. Such places could be with-in easy commuting distance, or furtherafield, that would imply migration.Sunderland could offer free metro passes topeople taking up work or looking for workin Newcastle. Easington could offer peoplehelp in relocating to Eastleigh. None ofthese options would be allowable today,but if these are what local councils believeto be the best answers to the problems thatthey face, then the only constraint on theiractions should be their electorates.

Provide the services and environment thatappeal to high skill workersIt is already seen as legitimate to spendregeneration money on high profile artsspending. The Baltic Centre forContemporary Art in Gateshead, for exam-ple, was designed to make Newcastle andGateshead vibrant places, attractive to thosewith high disposable incomes. But howevermuch the urban intelligentsia may wish itotherwise, high art is a minority pursuiteven among the affluent. It should be equal-ly acceptable to spend money on otheractivities that will attract them – such asgolf. There is no good reason why it shouldbe legitimate to spend regeneration moneyon an opera house, but not on a golf course.Both can attract tourists. Both can make acity a more appealing place to live. Whetherto do either, neither or both should be up tothe council, but it should have the right tomake that decision for itself.

Provide better educationWe selected golf as an example because it iswidely perceived to be a popular activityamong the sort of affluent business people

Cities unlimited

52

86 Leunig T and Swaffield J


Exchange, London 2007

that many of our regeneration towns des-perately need to attract. But there are manyother services that they could offer.Perhaps the most interesting one is highquality education. We have noted thatmany regeneration cities have lower thanaverage exam results and their schools donot act as magnets to outsiders. It would beperfectly legitimate for a council to decideto spend regeneration money on localschools. Canadian cities have successfullyused quality education to attract familiesto relatively remote places, and ManchesterCouncil has recently underwritten six newacademies even though it will not havecontrol over the schools.87

The aim would be twofold. First, itwould serve to attract those who care mostabout education – often the well-educatedwho have benefited from it themselves – tothe city. Such a policy could be quite pow-erful in a town near a more successful citylike Manchester, for example. Many ofManchester’s most affluent workers live tothe South of the city, in Altrincham,Didsbury, Sale and Macclesfield. They donot live in Middleton, Oldham orRochdale, even though these towns arewell within commuting distance. Thequestion these areas might want to askthemselves is not how can we attract firmsto our area, but how can we attract high-skill, high-wage people who work inManchester to live in our area?

Doubling the number of teachers andpaying them significantly more would beexpensive. But it could make an area veryattractive to young families wishing tomove out of city-centre flats. Such peoplebring money with them and at least someof that money stays in the area, where it isspent on local services and supports localjobs. Such a policy would also have a sec-ond effect: improving local education stan-dards. Some of those students will remainin the area longer term and, better quali-fied, they will represent a more attractiveworkforce for potential employers thinking

about where to locate. And if that doesn’twork, if the best schools in the country donot attract skilled workers, and if the besteducated school-leavers do not attractfirms to an area, then the only conclusionwill be stark: this is an area whose otherproblems – most likely intrinsic geography– are so deep-seated as to make it a poorlocation for economic activity. At very leastthat town will have equipped its youngpeople with the skills and qualificationsthey need to get jobs elsewhere. The townmay not have been regenerated, but liveswould have been transformed for the bet-ter – by any sensible measure a successfuloutcome.

Give your city a unique selling pointStoke-on-Trent City Council was recentlypraised in a Communities and LocalGovernment report for the quality of itscivic trees.88 All across Britain parks andstreet trees are under threat. Local authoritybudgets are constrained by limited centralgovernment grants and apparently unlimit-ed central government requirements on theone hand and people’s understandable aver-sion to council tax rises on the other.Discretionary items such as the upkeep ofparks have frequently been victims of harshcuts. Under the policies that we advocate,cities would have a much greater ability tocreate a stronger image for themselves.Stoke might want to build on its reputa-tion for tree care to make itself literally thegreenest city in the region. City parkscould be well funded, and imaginativelylandscaped, city streets could be festoonedwith hanging baskets. We are not sayingthat Stoke – or any other city – should dothis. But we are saying that they should beallowed to do this.89

Cut taxesFinally, local councils should always havethe right to use their money to cut counciltaxes.90 This would emphatically not beallowed under current rules. There are two


What would cities do?

87 See Lim C and Davies C,

Helping Schools Succeed:

Lessons from abroad, Policy

Exchange, 2007, Chapter 3,

Meikle, J, “Manchester under-

writes six city academies”, The

Guardian, 9th January 2008, see

http://www.guardian.co.uk/uk/200

7/jan/09/politics.localgovernment

88 Stoke-on-Trent Council suc-

cessfully attracted external

sponsorship worth over

£100,000 for its tree programme

www.communities.gov.uk/news/

corporate/698826

89 Vancouver has put in place

its EcoDensity green city strate-

gy. Freiburg in Germany is build-

ing itself up around green tech-

nology (see http://vorort.bund.net/

suedlicher-oberrhein/freiburg-

environment-ecology.html).

Boulder in Colorado and New

York’s PlaNYC are both attempts

to make cities more liveable and

green

90 Mid 1990s New York was los-

ing substantial retail activity to

out-of-town developments in

New Jersey, to counteract this

decline and bring retail back to

the city, the city removed VAT on

all clothing under $100. This has

been seen as a very successful

way to keep retail in place and

played a part in the rebirth of the

city

reasons why councils might want to dothis. First, it is possible that lower taxes willattract more people and so raise the eco-nomic potential of an area. Note that sincethose on benefits have council tax paid forthem, this policy would be a way to try toattract those in reasonably well paidemployment or with reasonable retirementincomes, the sort of people who bringmoney to an area and who may go on tostart businesses and regenerate a place.

But there is a second reason why coun-cils might want to use the money to cutcouncil taxes. Some areas are likely toaccept the reality, sooner or later, thatmuch regeneration funding does not passcost-benefit analysis. In those circum-stances, voters will, sooner or later, prefer

tax cuts to spending that achieves little. Ifan area is located in a place that makes solittle sense under current economic condi-tions that decline is inevitable, a councilmay well wish to spend some of the moneyhelping people to leave and some on mak-ing the transition more palatable for thosewho remain. Reducing their bills by£1,000 a year may well be an effective partof the palliative care for those who areunlucky enough to live in towns that are ininexorable population and economicdecline. This will not be a popular optionfor those in the regeneration industry, butit may prove to be very popular with thosewho have seen regeneration vision afterregeneration vision in their area, and won-dered, at the end, what has been achieved.

Cities unlimited

54

13Coping withpopulation decline

If the coal-powered, sea-transport basedIndustrial Revolution led to populationsbeing located in areas that are not ideal fora modern, service sector, road and air-based economy, then allowing significantlevels of migration makes economic sense.Since the British population is not growingdramatically – certainly not by 19th centu-ry standards – this implies that some areaswill experience population decline. In factthis is already happening. Liverpool, forexample, has lost almost half its populationsince its 1931 peak of 856,000. It is notalone in shrinking: half our regenerationtowns have shrunk in absolute terms in thelast ten years; Merthyr Tydfil, Blackpool,Hull and Middlesbrough have lost morethan 5 per cent of their populations.

Population decline is not necessarilydesirable. Migration can break up commu-nities and it is often the case that the mostenterprising migrate, leaving behind acommunity that lacks skills and entrepre-neurs. We can see that by looking at varia-tions in skill levels by area. That there aremore people with high levels of skills inour successful towns than our regenerationtowns is not primarily a reflection of thequality of schooling, but a combination ofthe greater attractiveness of some places toskilled workers and house prices sufficient-ly high in those places to prevent thosewith fewer skills from moving there.

Many of those with the least skills live insocial housing. The local allocation ofsocial housing makes it very hard for themto move area. That has to change, or thosein social housing will be trapped in com-

munities in which the low skilled arealready overrepresented, with adverseimplications for employment. Clearly noone should be forced to relocate, but weneed to allow those in social housing tomove from areas of population decline toareas of population growth just as those inprivate housing can. If we do that, we willhave gone some way to preventing popula-tion decline in an area becoming synony-mous with a deterioration in its skills base.

Migration may lead to uneven age struc-tures because young people are more likelyto leave declining areas than the old. Thisis certainly the case for internationalmigration: those coming to Britain fromthe accession states of Eastern Europe tendto be young, just as those emigrating to theUnited States, Canada and Australia a cen-tury ago were generally young. But it is,perhaps surprisingly, not true for internalmigration. We can see this by looking atthe age profiles for different towns.Merthyr Tydfil is losing population fasterthan any other town in our sample, but itspopulation distribution is little different tothat of, say, Windsor. Twenty per cent ofMerthyr residents are under 15, comparedwith 19 per cent in Windsor. Similarly, 16per cent of Merthyr’s residents are 60 orover, compared with 15 per cent of thosein Windsor. Merthyr is not a place fromwhich all the young leave, rather thedecline in population is spread reasonablyevenly over the age distribution.

Population decline should also beenseen as an opportunity. There is no partic-ular reason why a city should become less


desirable if its population shrinks.Sometimes a regeneration city enjoys alegacy of past investment in infrastructure.Liverpool, for example, has more railwaystations relative to its population than anyother place in Britain. The amount of openspace per person will increase if populationfalls, even if no additional green space iscreated.

More generally, a declining populationgives opportunities to improve the averagesize and quality of the housing stock forthose who remain or, for that matter, toincrease the amount and quality of publicspace and parkland, as seen in the Ruhrarea in recent years.91 At the moment theGovernment’s standard approach for areasthat have falling populations is to demol-ish housing and rebuild it. Thus theNational Audit Office found that theHousing Market Renewal Programme’sPathfinders areas planned to demolish57,100 properties and build 67,600 newones.92 Only three of the nine Pathfinderareas will see a net decline in their housingstock. We believe that this approach isfundamentally misguided. First, the prin-cipal reason why house prices are low ineach of the micro-areas is indeed likely tobe that they are less desirable than otherareas. But the underlying cause of lowhouse prices is not poor quality housing,but an inability to generate jobs. If a par-ticular area is transformed from undesir-able to desirable, but nothing is done toincrease the economic sustainability of thetown as a whole, then any gains in popu-lation are likely to come at the expense ofother parts of the town. A great deal ofmoney, in other words, can be spent sim-ply shifting the problem from one neigh-bourhood to another nearby. The 2007National Audit Office report on housingrenewal commented: “There is no guaran-tee that intervening in the housing marketin this way will address the causes ratherthan the symptoms of the problems expe-rienced in these neighbourhoods.”93

If an area has structurally low econom-ic potential then the housing stock needsto fall. But in general that should notmean demolition, which has at least fourdrawbacks. First, it involves spending sig-nificant sums of money on destroyingassets that have potential value. Secondly,the prospect of demolition can and doesblight a neighbourhood: people planningfor the long-term will not move to thearea. As such, the threat of demolitionwill increase the transient nature of thatcommunity, as well as leading to moreempty homes. Thirdly, when owner-occu-pied property is demolished in a period ofrising house prices, the lag between aprice being agreed and the owner buyinga new property leaves the owner out ofpocket, by an average of £35,000. Thatcannot be right. Finally, speculators oftenmove in to areas in which demolition ispredicted, bidding up prices and makingdemolition more expensive. Using free-dom of information legislation The Timesfound that the Government had paid aninvestor £450,000 for an end of terracehouse in Manchester, more than twice theprice paid for any other property in thatroad.94

There will be occasions in which dem-olition is the right way forward, butwhere property is structurally soundthese will be exceptions rather than therule. In general, when a house is valued atmore than the cost of construction, thereare too few homes in the area. In suchcircumstances, builders want to buildmore homes, since they can make a prof-it. In contrast, when a house is valued atless than the cost of construction, thereare too many homes in the area. The firstlesson, therefore, is that the number ofproperties should not be reduced in anyarea where the price of a house exceedsthe building cost. Thus recent rises inhouse prices should lead to a reduction inthe number of houses demolished at stateexpense.

Cities unlimited

56




92 Department for Communities

and Local Government: Housing

Market Renewal, National Audit

Office, HC 20 Session 2007-

2008, 9th November 2007 see

www.nao.org.uk/publications/na

o_reports/07-08/070820.pdf

93 Ibid, p 7

94 Ungoed-Thomas J and Waite

R, “Ministers pay £450,000 to

flatten terraced home”, The

Times, 30th August 2007, see

www.timesonline.co.uk/tol/news/

politics/article2558168.ece

When property prices fall below the costof construction there is a case for reducingthe number of homes. If this does not hap-pen, property can become worthless veryquickly because there is no incentive forany individual to demolish his own housein order to make those of his neighboursmore valuable. This is a classic externalityand warrants government intervention.

The plan-led system of identifyinghouses for demolition risks giving specula-tors a free lunch. They can buy up rundown properties safe in the knowledge thatthe Government will buy them out at veryleast for full market value, and usuallymore. The prospect of demolition alsoblights a neighbourhood. We need to moveaway from a plan-led system towards amarket-led system for working out whichhouses are least wanted. Here the price sig-nals are remarkably easy to understand: thecheapest houses (taking size into account)are the least desirable, and those are theones that Government should be takingout of supply.

It would be possible for theGovernment simply to demolish the hous-es (although not the flats) that it buys inthis way, creating small green spaces scat-tered across the city as happens in theRuhr.95 But in general demolition is notthe right way forward because demolishinga house destroys something of value. Weneed to find a way of reducing the numberof houses without destroying the housesthemselves. The obvious answer is tomerge houses laterally so that the peoplewho remain in the city that is witnessing afall in population end up living in biggerhouses, at no cost to society.

Let us imagine a row of terraced houses,of the sort that are widespread in manyregeneration towns. One comes up for sale,at a low price. At this point the stateshould step in, and consider buying it. Itwould approach the two people who liveon either side of it, and ask them whetherthey would like it. Let us assume for sim-

plicity that one says yes, and one says no.At that point the Government will buy theproperty, and sell it to the neighbour.Given that regeneration communities aregenerally poorer than others, it is unlikelythat the neighbour will have the capital topay for any proportion of the house out-right. Instead the state would own a pro-portion of the new, larger house, rather likea mortgage charge. No rent or interestwould be payable, but when the new larg-er house is sold, government would gain atleast some of its money back. The new res-ident would be responsible for all the alter-ations necessary to merge the two houses.Of course, many of the new, larger proper-ties would be rather odd in layout anddesign. At least initially, many would retaintwo staircases, two electricity and gas sup-plies, and so on. But for many poor fami-lies, living in relatively cramped condi-tions, being given a larger if oddly designedproperty for the cost of creating an internaldoorway or two would be a very attractiveprospect. Over time, when the electricsneeded renewing, the gas boiler replacingor the kitchen refurbishing, the housewould gain a new, more unified structure.

Apart from a payment in the form of ashare of ownership, there would be onlyone condition for taking over a neighbour-ing property in this way: that the formerlyseparate house must be merged with thenew house. It must not be let separately orsold as a separate dwelling. The LandRegistry deeds would be changed to reflectthis. The planning system requires that ahouseowner gets planning permission tosubdivide a property and that would notusually be given unless local economic


Coping with population decline

Leunig T and Swaffield J,

Success in the city, Policy


“ We need to move away from a plan-led system

towards a market-led system for working out which

houses are least wanted ”

circumstances had changed dramatically.In those cases the State would have theright to all of its investment back.

In general the Government should notbe selling property acquired in this way tolandlords. In those circumstances wewould expect the new (larger) property tobe re-let to a larger number of people,which would defeat the purpose of reduc-ing the supply of housing. If landlordswanted to buy the house next door, thenthey could do so simply by outbidding theGovernment.

There will be times when neither neigh-bour is interested in having a larger prop-erty. In those cases the State would not buythe house concerned, but instead allow itto be sold in the normal way for a marketprice. The State would wait for the nexthouse to become available and againapproach the neighbours to see whetherthey would be interested in a larger prop-erty. So long as a reasonable number can beacquired like this, the downward spiral ofhouse prices can be avoided.

Sometimes both neighbours would beinterested in a larger house. In that caseeither the house being bought by the Statecan be shared between them or it can beauctioned. As time went on theGovernment would learn the usual propor-tion of the new house that it would be

offered, and would decline to buy houseswhen neighbours offered lower propor-tions, as well as when the neighboursdeclined to buy the property altogether. Itis impossible to know ex ante the propor-tion of the larger house that theGovernment would be offered in general,or how much the value of the house wouldincrease as a result of being larger. We can-not say, therefore, how financially effectivethe scheme would be. That said, we wouldnot expect it to be profitable: if it wereprofitable, it would be happening alreadyon its own, without government interven-tion.

Although not profitable, the scheme isattractive at many levels. First, it avoids thecosts of demolition and, when the house issold, the Government gains a proportionof its value. Secondly, it avoids compulsorypurchase and all the upset it causes to thosepeople who, for perfectly legitimate rea-sons, do not wish to move. Thirdly, itavoids all the problems of blight that occurwhen an area is marked for wholesale dem-olition. Finally, some people will get to livein bigger houses at no cost to society andonly trivial costs to themselves.Regeneration cities are not known for largehouses, so an increase in the size of somehouses is likely to be welcome and willimprove local living standards.

Cities unlimited

58

14Improving governance

We have proposed giving large sums ofmoney to local authorities, with no stringsattached. The risk is that it will not bespent effectively. The fact that central con-trol over regeneration funding has faileddoes not guarantee that local control willsucceed. In short, there is a danger thatBritain’s “caged hens” will become “head-less chickens”, unable or unwilling toimplement policies that work. We need toensure that local authorities have theincentive and the expertise to use themoney wisely.

Currently, central government tends toprovide the incentives to follow particularpolicies. It sets targets, and measures theextent to which local authorities meetthem, whether they are for school perform-ance or recycling. That is not the modelthat we are advocating, indeed we think itstifles local initiative and local choice.

Instead we want local authorities to beresponsible to local people. That is easy tosay, but self-evidently not so easy toachieve. It is outside the scope of this paperto offer proposals for radically restructur-ing councils but nevertheless there aresome obvious areas that can be improved.

Local authorities are elected bodies andare ultimately responsible to local peoplethrough the ballot box. But outside thedevolved areas, accountability is limited bytheir lack of power. Giving more power tolocal government will improve the qualityof local governance in four ways. First, itwill increase local accountability becausepeople will know who to praise and who toblame. At the moment local authoritiescan always excuse poor services or high

taxes by blaming central governmentbecause it is so dominant. Very often –though not always – they are right, but theimportant point is that voters cannot tellwhen they are right and when they arepassing the buck. Giving them more powermeans that the excuses will no longer wash,and voters can better judge the quality oflocal leadership.

Secondly, many councillors are awarethat electoral success depends on claimingresponsibility for good things that wouldhave happened anyway, and blaming cen-tral government for anything that goeswrong. Once they know that they will bejudged according to their genuine per-formance, they will have a greater incentiveto get the fundamentals right instead ofwasting time on spin.

Thirdly, giving local authorities morepower increases the attractiveness of acareer in local government either as acouncillor or as a council officer. No longerwill your primary responsibility be to copewith a deluge of instructions fromWhitehall, but rather you will have thepower, and the resources, to shape yourcommunity. It is likely that this willincrease the range and quality of peoplewilling to serve in local government at alllevels. That in turn will improve perform-ance.

Fourthly, giving local authorities morepower will increase the range of policiesthat are tried, generating evidence aboutwhat does and does not work. This evi-dence can be used both by social scientistsand those interested in evaluating policy,and by voters. If the voters of Blackburn


see that Burnley is doing manifestly betterthan their town, they can ask why this is. Itmay be that Burnley is lucky enough to behome to the next James Dyson, but itmight be that its policies are better, andthat the party in power – or in opposition– in Blackburn can and will learn fromBurnley.

Two of the greatest incentives for electedpoliticians are the prospect of gaining officeand the fear of losing it. For that reason itis hard to believe that a city in which aparty – any party – will never realisticallylose control is good for politicians’ incen-tives. Although there are examples of “one-party states” at local level that are effective,in general being guaranteed power comewhat may is not good for effective gover-nance. This is as true in Britain as in one-party states in sub-Saharan Africa. Andone-party councils are all too common inareas needing regeneration.

It is not the purpose of this paper to sug-gest wholesale reforms to the structure oflocal governance, but we do think thatBritain should be open to significant levelsof experimentation. Here we note withapproval the Scottish Assembly’s decisionto adopt the single transferable vote systemof proportional representation for localgovernment. It is not that we believe thatthis system is better than others, but ratherwe believe that trying this system will allowus to see whether it is more effective. It ispossible – and indeed likely – that it willcurtail one-party rule in many places. If so,it has the potential to improve the qualityof governance. But it is equally possiblethat it will lead to permanent coalitionsbetween the same parties, election afterelection, which would do little to increasethe discipline placed on politicians by theballot box. It is too early to tell whether thenew voting system will increase theaccountability of councils to their voters,but it is an experiment that should be sup-ported and whose effects should be studiedhard.

Let us imagine then, that local govern-ment has considerable powers that theycan use to better or worse effect. Let us alsosuppose that the voting system presents theruling party with a realistic chance of los-ing office, and offers at least one otherparty a realistic chance of gaining office.What we now need is a system that allowslocal people to make well-informed deci-sions whether to re-elect an incumbent ornot.

Here Britain starts with two tremendousadvantages: the Audit Commission, andthe local press. The Audit Commissionalready reports on the performance ofalmost every aspect of local government. Itdoes so thoroughly, objectively and apolit-ically. And yet virtually no one knows howthe Audit Commission ranks the councilon any measure.

There are two reasons for this. First,there is a perception that it is pointless toknow. Many local councils are one-partystates, so knowing that they are badly rundoes not allow you to change anything. Weneed a voting system that enhances con-testability. Secondly, the AuditCommission is not currently set up toexplain their findings to a wide audience.

This can and should be changed. Wewant to see the Audit Commissionenhance the democratic process by spend-ing much more time explaining its find-ings. It should do so in public, with formalcommittee hearings in every council inBritain, in which councillors of all partiescan grill those who wrote the report. Whatis the evidence for this conclusion? What isthe basis of that conclusion? What is it thatothers, who you say have done better thanus, have done?

But this should not be limited to formalhearings in council chambers. The AuditCommission should also make itself avail-able to explain its findings, in person, toeach and every political party representedon the council. Councillors will not all beexperts on regeneration, and they need the

Cities unlimited

60

opportunity to ask for things to beexplained to them in private, in the waythat ministers are briefed by civil servants.The same is true for council officers, forwhom neutral Audit Commission briefin-gs that compare their performance withtheir peers can be a very effective way ofdisseminating best practice.

We should not stop there. Britain has aremarkably vibrant local press, capable ofengaging with serious local issues. TheAudit Commission needs to brief the localpress, again in private and in person, onwhat their findings mean. We need semi-nars in which local journalists can exploreexactly what the Audit Commission is say-ing, for better and for worse.

Even then, there are more groups for theAudit Commission to engage with. Theyshould be willing to discuss their findingswith any community group that wants tofind out more. There are a myriad of localgroups: the local rotary club, the church,synagogue and mosque, a school PTA, atenants’ association, a local residents’group, the WI. It doesn’t matter who theyare, the Audit Commission needs toengage with any part of civil society thatwants to hold local government toaccount. And not only that, the AuditCommission must actively seek out civilsociety across Britain and engage withthem, presenting information in a way thatallows them in turn to hold local govern-ment to account.

Here the local media have an importantrole. Britain is lucky to have exceptionallyhigh quality local media. Most local news-papers in Britain make real efforts to coverlocal stories properly and to understandthe issues at stake. Unlike national newspa-pers, they rarely have strong a priori polit-ical affiliations. They are thus well placed

to disseminate well-presented informationfrom the Audit Commission and othersources to their readers.

The same is true of BBC local radio andof both local BBC and independent televi-sion news programmes. In all cases theseprogrammes are protected to some extentfrom competition and have formal man-dates to cover local news. They have thestaff and the budgets to make a good job ofunderstanding the issues and explainingthem to people in the area.

Finally we should also make it easier forindividuals to find out more about what isgoing on and whether regenerationschemes are delivering good value formoney, and achieving their aims. To thatend, the power of freedom of informationlegislation needs to be bolstered so thatinterested individuals can find out, in astraightforward and timely manner, why acouncil decided on a particular approachor policy, why they awarded a contract to aparticular company or organisation, andthe results achieved as a result of themoney spent. Although both the AuditCommission and the local media haveimportant roles to play, sometimes there isno substitute for someone local and deter-mined, who knows what is going on andcan sense that a project is poorly conceivedor poorly delivered.

Of course, none of this will preventpoliticians of every hue trying to spinevery finding in the way that suits them.But the more people who know more ofthe truth, the less politicians can get awaywith partial representations of the truth,or worse. If we are to hold local govern-ment to account, we need citizens to bewell-informed, and able to act on thatinformation.


Improving governance

15Conclusion

We have set out a new agenda for regenera-tion policy. We do not claim that, if actedon, every town or every region can beregenerated. Instead we argue for a muchmore hard-headed analysis of the potentialof towns that are struggling and an accept-ance that at least some of them have littleprospect of offering their residents the stan-dard of living to which they aspire. Townsthat were once in the right place at the righttime, may now well be in the wrong placeat the wrong time, severely restricting theirability to attract jobs. No one is suggestingthat residents should be forced to move,but we do argue that they should be toldthe reality of the position: regeneration, inthe sense of convergence, will not happen,because it is not possible.96

Furthermore, we do not believe that it isacceptable for planning policies to con-strain internal migration as tightly as theydo. At the moment restrictions on housebuilding in the South East mean that it isvery difficult for those without very wellpaid jobs to move to the area. Those withlower levels of skills are trapped outsideBritain’s most prosperous region, particu-larly if they live in council housing. Thatcannot be right and we should not acceptit. We have argued that there are manyways to make greater levels of house build-

ing acceptable to people in the South East,all of which revolve around providing localauthorities with greater self-funding finan-cial incentives to support development.

Not every struggling town has a bleakfuture. The critical issue is market poten-tial. Some areas that are struggling couldbe well connected to successful areas. Suchplaces have much greater economic poten-tial than those places that are intrinsicallymore isolated. Coastal cities, whether largelike Liverpool and Hull, or small likeScunthorpe and Blackpool are most vul-nerable. All have accessibility problems in aroad-based era, because they are almostalways at the end of the line. Our largestcoastal cities are generally large for a reasonthat has disappeared: ocean-going ship-ping, and ship building. They have lostmuch of their raison d’etre, and it is hardto imagine them prospering at their cur-rent sizes. Sunderland demonstrates justhow hard it is to regenerate such a city.

We argue that all towns and cities,whether we believe them connectable ornot, should have the right to determinetheir own policies. We believe that theGovernment should roll up almost all cur-rent regeneration funding, and allocate itaccording to need. Areas would then beable to spend the money on traditionalregeneration schemes if they believe thatthose regeneration schemes will delivervalue for money. Or they would be allowedto spend the money on less conventionalregeneration schemes, including deliberate-ly setting out to attract highly skilled work-ers from elsewhere, and facilitating localpeople’s search for work in other, better

62

96 Given the geographical com-

position of trade in the global

economy which is determined

by embedded spatial and skill

based wage inequality, see

Krugman’s conundrum: “The

elusive link between trade and

wage inequality”, The

Economist, Economic Focus,

p92, 19th April 2008

“ We believe that the Government should roll up

almost all current regeneration funding, and allocate

it according to need”

placed, locations. And, if they really believethat there are few prospects for their area,whatever policies are followed, they shouldbe able to spend the money on palliativemeasures, improving the quality of life forthose people who, for whatever reason,cannot or will not move. Such people mustnot be forgotten.

That increase in freedom needs to comewith a big increase in accountability. Localpeople must be able to judge the success orotherwise of their council, and must beable to act on their judgement. That meansthat councils must be assessed regularlyand accurately, and that the results of suchassessment must be explained to local peo-ple in a way that simply does not happenat the moment. It is no good an assessorknowing whether a council is doing well orbadly, we need local people to know theanswer as well.

Crucially, we need local people to beable to act on what they find. We needlocal government to be importantenough that people bother to find outwho runs the council, and how well theyare doing. And we need an electoral sys-tem that makes each and every councilgenuinely contestable between at leasttwo parties, or groups of parties. Noparty should ever be able to take the elec-torate for granted. When local govern-ment has power, and when elections canchange how an area is run, then we canexpect local people to vote and to holdpoliticians to account.

A genuine and honest assessment ofconnectability and potential; the freedomto make and implement policy; and thediscipline of real accountability to real peo-ple: these are the basis for regenerationpolicies that will work.


Conclusion

cities unlimited

Documents

urban policy

policy experience

policy development

cities limited

new policy ideas

london sw1p

expanding london

london school ofeconomics