Unleashing the Economic Potential of Agglomeration in African Cities

7/28/2019 Unleashing the Economic Potential of Agglomeration in African Cities

1/62


2/62


3/62

i

United Nations Human Settlements ProgrammeNairobi 2013

UNLEASHING THE ECONOMICPOTENTIAL OF AGGLOMERATIONIN AFRICAN CITIES


4/62

ii

Te Global Urban Economic Dialogue Series

Unleashing the Economic Potential of Agglomeration in African Cities

First published in Nairobi in 2013 by UN-HABIA.

Copyright United Nations Human Settlements Programme 2013

All rights reservedUnited Nations Human Settlements Programme (UN-HABIA)P. O. Box 30030, 00100 Nairobi GPO KENYAel: 254-020-7623120 (Central Office)

www.unhabitat.org

HS Number: HS/126/12EISBN Number(Series): 978-92-1-132027-5ISBN Number:(Volume) 978-92-1-132539-3

Disclaimer

Te designations employed and the presentation of the material in this publication do not imply the

expression of any opinion whatsoever on the part of the Secretariat of the United Nations concerningthe legal status of any country, territory, city or area or of its authorities, or concerning the delimitationof its frontiers of boundaries.

Views expressed in this publication do not necessarily reflect those of the United Nations HumanSettlements Programme, the United Nations, or its Member States.

Excerpts may be reproduced without authorization, on condition that the source is indicated.

Acknowledgements:

Director: Naison Mutizwa-Mangiza

Chief Editor and Manager: Xing Quan Zhang

Principal Author: Ivan urok

English Editor: Roman Rollnick

Design and Layout: Fredrick Maitaria

Assistants: Agnes Ogana, Joy Munene


5/62

iii

FOREWORD

Urbanization is oneof the most powerful,irreversible forces in the

world. It is estimatedthat 93 percent ofthe future urbanpopulation growth willoccur in the cities of

Asia and Africa, andto a lesser extent, Latin

America and the Caribbean.

We live in a new urban era with most ofhumanity now living in towns and cities.

Global poverty is moving into cities, mostlyin developing countries, in a process we call theurbanisation of poverty.

Te worlds slums are growing and growing as

are the global urban populations. Indeed, this isone of the greatest challenges we face in the newmillennium.

Te persistent problems of poverty and slumsare in large part due to weak urban economies.Urban economic development is fundamental toUN-HABIAs mandate. Cities act as enginesof national economic development. Strong urban

economies are essential for poverty reduction andthe provision of adequate housing, infrastructure,education, health, safety, and basic services.

Te Global Urban Economic Dialogue seriespresented here is a platform for all sectors of thesociety to address urban economic developmentand particularly its contribution to addressinghousing issues. Tis work carries many new ideas,solutions and innovative best practices from

some of the worlds leading urban thinkers andpractitioners from international organisations,national governments, local authorities, theprivate sector, and civil society.

Tis series also gives us an interesting insight anddeeper understanding of the wide range of urbaneconomic development and human settlementsdevelopment issues. It will serve UN member States

well in their quest for better policies and strategies

to address increasing global challenges in these areas

Joan ClosUnder-Secretary-General of the United Nations,

Executive Director, UN-Habitat


6/62

iv


7/62

v

CONTENTS

FOREWORD III

CONTENTS V

SECTION 1: INTRODUCTION 1

SECTION 2: SETTING THE SCENE POLARISED PERSPECTIVES 3

SECTION 3: THE ROLE OF GEOGRAPHY AND URBANISATION 7

SECTION 4: CAN CITIES HELP TO BUILD MORE INTEGRATED ECONOMIES? 10

SECTION 5: AGGLOMERATION ECONOMIES IN THEORY 13

SECTION 6: DIFFERENT ADVANTAGES OF AGGLOMERATION 16

SECTION 7: DIFFERENT FORMS OF AGGLOMERATION 19

SECTION 8: EMPIRICAL EVIDENCE OF AGGLOMERATION - ANALYTICAL PRINCIPLES 22

SECTION 9: EMPIRICAL EVIDENCE OF AGGLOMERATION - ECONOMETRIC STUDIES 24

SECTION 10: EMPIRICAL EVIDENCE DEVELOPING COUNTRIES 27

SECTION 11: EMPIRICAL EVIDENCE FROM AFRICA 28

SECTION 12: EMPIRICAL EVIDENCE OTHER STUDIES 30

SECTION 13: AGGLOMERATION DISECONOMIES 33

SECTION 14: HOW DO DYNAMIC URBAN ECONOMIES DEVELOP? 35

SECTION 15: THE ECONOMIC TRAJECTORIES OF AFRICAN CITIES 36

SECTION 16: IMPLICATIONS FOR POLICY 41

REFERENCES 47

LIST OF TABLES

TABLE 1:INDICATORS OF EMPLOYMENT GROWTH FOR SELECTED AFRICAN COUNTRIES 7

TABLE 2:ESTIMATES OF AGGLOMERATION ECONOMIES 25

TABLE 3:ANALYSES OF AGGLOMERATION ECONOMIES IN THE SOUTH 27

TABLE 4:PROJECTED GROWTH FOR SELECTED AFRICAN CITIES 40

LIST OF FIGURES

FIGURE 1: ESSENTIAL INTERACTIONS WITHIN CITIES 41


8/62

vi


9/62

1

SECTION 1: INTRODUCTION

Tere is a new mood of optimism about theprospects for Africa, buoyed by regular discoveriesof untapped oil reserves and other minerals,booming commodity exports, big constructionprojects, new shopping malls and other palpablesigns of economic revival. Strong growth over thelast decade coupled with relative resilience duringthe recent global crisis have fuelled perceptions

that the continent has shifted onto a higher planeof development, following several decades ofeconomic decline and political instability (IMF,2011; World Bank, 2011; Economist, 2011).Faced with static populations and saturated marketsin many developed economies, the possibility ofa burgeoning middle class of consumers is alsoenticing investors. Te conventional view amonginternational organisations is that the mainrequirement to sustain this positive trajectory is a

better business climate to attract private enterprise.Reinvesting some of the resources generated fromexploiting the minerals bonanza in moderneconomic infrastructure would also help.

Of course the major problems of poverty andunemployment have not suddenly diminished,nor have the threats of environmental degradationand food shortages posed by global warming.

Africas relatively rapid rate of urbanisation andpredominantly informal urban economies presenta particular set of challenges that are receiving farless attention from governments or internationalbodies (UN Habitat, 2010; CommonwealthSecretariat, 2010; UNCAD, 2011a). Strongpopulation growth in congested cities withinadequate provision of basic infrastructureheightens the risks of increased squalor, hardshipand public health disasters. Te concentrationof people with very low incomes in constrainedspaces also creates the conditions for mounting

frustration, social unrest and exposure to floodingand fire hazards.

Tis juxtaposition highlights the need to connectefforts to bolster African economies with morepositive responses to demographic movements,i.e. to align economic growth and urbanisationagendas. Evidence from around the world suggeststhat linking economic and urban developmentcan generate positive interactions or spillovers thatimprove economic outcomes and human well-

being. Conversely, ignoring the spatial implicationsof economic trends heightens the risk of producingimbalanced, exclusionary and destabilising effects.In the context of climate change, rising energyprices and global pressures to cut carbon emissions,it is also important to incorporate environmentalconsiderations into future plans. Tese concernstend to reinforce the case for concentrated ratherthan dispersed development, including localisedproduction of energy, food, water, building materials

and many other goods and resources.

Tis report seeks to explore what local andnational governments can do to harness theeconomic possibilities of Africas expanding citiesand to maximise the benefits from its abundantnatural resources. It is argued that urbanisation cancontribute to greater economic dynamism andsustainable reductions in poverty if it is planned andmanaged more effectively. Tis should mean that theincrease in population is accompanied by strongeremployment growth, enabling people to enjoy higherliving standards and to lead more useful and fulfillinglives. Tis will require the expansion of productionand not just consumption, improvements in publicservices, and the creation of better functioning urbanenvironments.

Above all, more robust and inclusive growthrequires diversification from Africas primarycommodities. Economic policy needs to build on the

continents competitive advantages by adding valueto its raw materials and minerals. Geography mattersto the prospects for broadening and deepening


10/62

2

UNLEASHING THE ECONOMIC POTENTIAL OF AGGLOMERATION IN AFRICAN CITIES

African economies. Global experience suggests thatcities can help to transform the scale and natureof productive activity and build more integratedeconomies with stronger backward and forwardlinkages. With suitable investment in infrastructure

and institutions, urban environments provide moreefficient logistics, improved access to skills, supportservices and finance capital, stronger connections tolocal and international knowledge and expertise, andmore intensive learning and innovation. Bearing inmind that this is not a quick fix, cities should alsobe planned and managed as places where citizenscan better organise themselves, enhance their ownproductive skills and capabilities, and developpractical ways of constructing better livelihoods that

go beyond bare survival and subsistence activities inthe informal economy.

Te report also cautions against oversimplifyingthese issues and exaggerating the economiccontribution of cities per se. Drawing policyrecommendations is difficult without a fullerunderstanding of the processes at work in theirparticular local contexts. Further research is requiredon African urban dynamics, especially the causes

and consequences of concentration. Some of theinternational literature portrays urbanisation as anecessary and sufficient condition for economicprogress, inevitably linked with higher productivity,output and average incomes. Observers conclude

that governments should favour investment in bigcities at the expense of smaller cities and towns.In fact careful consideration of the evidence doesnot justify such bald conclusions. Cities are nota panacea for prosperity and development. Tey

can help to reinforce progress, but there is noautomatic connection between physical proximityand economic growth. A series of other conditionsare also vital, including higher levels of productiveactivity, investment in public infrastructure andhuman capabilities (such as education, health andsocial welfare), more equitable terms of internationaltrade, and stronger local and national governmentinstitutions.

Te next three sections of the report set the contextby outlining the issues, challenges and opportunitiesfor linking urbanisation and development in

Africa. Te following sections review the theoreticalarguments for the economic advantages of cities,the different forms of agglomeration and theinternational evidence relating to the strength ofthese forces. Tis is critical to the case for preferentialpolicies towards cities. Subsequent sections considerthe relevance of these arguments to Africa, including

the obstacles that inhibit agglomeration. Te finalsection discusses some of the main ways in whichgovernments can help to realise the economicpotential of cities.


11/62

3

SECTION 2

SETTING THE SCENE POLARISED PERSPECTIVES

SECTION 2: SETTING THE SCENE .......POLARISED PERSPECTIVES

Te United Nations predicts that Africascurrent urban population of 400 million willdouble over the next 20 years and triple over thenext 40 years (UN Habitat, 2010). Tis will bethe result of natural growth (more births thandeaths) as well as rural-urban migration. By 2050

nearly two-thirds (60%) of Africas population areexpected to be urbanised, up from 40% today.

Africa has been experiencing the worlds fastestrate of urban population growth for decades,although the rate of growth is diminishing.1

Africa also happens to be the worlds poorest andmost precarious continent, raising the spectre ofurbanisation without growth.

Most people are moving to cities in the

hope of finding work and a better life. It isestimated that more two-fifths (43%) of theurban population in sub-Saharan Africa livebelow the poverty line and three-fifths (62%)live in overcrowded slums lacking basicservices and vulnerable to absolute poverty,illness, environmental crises and social disorder(UN Habitat, 2008). Tis is much the highestproportion in the world and the trend is rising.Informal employment already accounts forsome 72% of non-agricultural jobs in sub-Saharan Africa (ILO, 2002) and around 60% ofurban jobs (Kessides, 2006; UN-Habitat, 2008).

Yet informal enterprises operate with minimalcapital and low skills in congested markets(UNCAD, 2011a). Tey therefore generate verylow household incomes, little economic securityand no taxes to pay for better public services. oavoid rising poverty and disaffection, Africa needsto generate productive jobs and livelihoods for the7-10 million young people entering the labour

1 Some doubts about the reliability of these statistics andtheir interpretation have been raised by Potts (2009) andSatterthwaite (2010).

force each year, a disproportionate number ofwhom live in cities because of the youthful profileof migration streams.

Te urbanisation of poverty in Africaundoubtedly presents great challenges of

international significance. Some externalcommentators have portrayed the scenario ascompletely bleak with no hope of improvement.For example, in his book Planet of Slums, MikeDavis gives much attention to Africas cities,

which are said to be growing prodigiouslydespite ruined import-substitution industries,shrunken public sectors, and downwardlymobile middle classes (2005, p.16). Hehighlights extreme conditions of squalor,

degradation, decay and pollution, attributablepartly to globalisation and structural constraintson local actors and institutions. Davis and otherobservers tend to write-off Africa and its citiesas basket-cases, with little sign of hope and nosense that anyone is doing anything about theseproblems (Freund, 2007; Myers, 2011).

Partly in response to such pessimism, anew group of authors has emerged who aredeliberately less negative and judgemental about

Africas urban poverty (Nuttall and Mbembe,2008; Bremner, 2010; Simone, 2010). Teydraw particular attention to energy, ingenuityand creative spirit of informality. Poorcommunities are not passive victims but ratheractive agents with resilience and imagination tonegotiate the tough environments of Africancities. Tey are capable of adapting to theirphysical and economic constraints and makingthe most of the opportunities available through

experimentation and inventiveness. Yet byfocusing on the positive features of marginalcommunities and the coping strategies evident


12/62

4


in invisible urban practices on the periphery,this literature may also have failed to providean accurate and balanced interpretation of thesituation. It also appears to have missed thegrowing economic dynamism of many African

cities, reflected in major house-building andinfrastructure programmes.

Te economic prospects seem to bebetter than for many years. An unexpectedupturn in many African economies during thelast decade has caused a noticeable change inmood and confidence, with a renewed sense ofoptimism for the future. en years ago Africa

was typically portrayed in very pessimistic terms

following two decades of structural adjustment,government debt crises and collapsing publicinfrastructure. As a result, labour markets andservice delivery in the cities became increasinglyinformal, making workers and householdsmore vulnerable by undermining the quality of

jobs and social protection. Tis was apparent inpublic transport, water, waste removal, housingand land allocation. Te Economist magazinefamously labelled Africa Te Hopeless

Continent in 2000. Yet a decade later thecover of its December 3rd 2011 edition did aremarkable about-turn by describing it as TeHopeful Continent, with a real chance tofollow in the footsteps of Asia:

A profound change has taken hold. Labourproductivity has been rising. It is now growing by,on average, 2.7% a year. rade between Africa andthe rest of the world has increased by 200% since

2000. Inflation dropped from 22% in the 1990s to8% in the past decade. Foreign debts declined by aquarter, budget deficits by two-thirds. In eight of the

past ten years sub-Saharan growth has been fasterthan East Asias Over the past decade six of theworlds ten fastest-growing countries were African

(Economist, 2011).

Recent reports from international organisationssuch as the IMF (2011), World Bank (2011) andglobal consultancies such as McKinsey (2010),

Ernst and Young (2011) and Monitor (2011)have expressed similar up-beat sentiments.Tey almost seem to be competing to up the

ante. Monitor asserts baldy that Africa will beone of the main sites of the next wave of globaleconomic development (2012). Te WorldBank is slightly more tentative: Africa could beon the brink of an economic take-off, much like

China was 30 years ago and India 20 years ago(2011, p.2). Te IMF predicts that four of the

worlds top 10 fastest-growing economies in thenext five years will be from Africa. Africas recentsuccess and positive outlook appear to stem fromstrong global demand for primary commodities(especially oil, gas, metals and minerals such asdiamonds and coal) and agricultural products,coupled with the expansion of domestic consumermarkets as a result of strong demographic growth

and an emerging middle class. For example, thewest coast of Africa has risen rapidly to become amajor supplier of oil to the USA. Africa is believedto contain around 12% of the worlds oil and30% of global mineral reserves. Expenditure oncommercial exploration in Africa has outstrippedglobal increases, rising from $300 million in 2000to more than $2000 million in 2008 (Harrap,2012). Large new deposits are regularly beingidentified, such as the recent discovery of major

oil and gas reserves in anzania, Mozambique,Kenya and Ethiopia (Economist, 7th April 2012).

Historically in other parts of the world andnow in many parts of Asia, the productivity andcost advantages that stem from concentratedpopulations have helped to spur economicdynamism and growth. If they can find theresources and institutional capabilities to do so,investing in infrastructure and other instrumentsof economic and social development shouldhelp African cities to function more efficiently,and thereby secure positive externalities fromurbanisation. Critical public decisions takenover the next few years on where and whattypes of infrastructure are pursued will lock inparticular development trajectories for decadesto come. Tey will either reinforce the positivelink between urbanisation and development,or undermine it and cause urban problems tobecome overwhelming. Careful decisions on

human settlement planning, transport systems(public or private), power generation (high orlow carbon), food production, water, sanitation


13/62

5

SECTION 2

SETTING THE SCENE POLARISED PERSPECTIVES

and other community services will have far-reaching and long-lasting social, economic andenvironmental effects.

Although Africas commodity exports could help

to generate some of the resources required to investin urban infrastructure, there are many competingclaims on these funds, ranging from support foragriculture, education and social programmes,to expanded public administrations. Africas fourmain exports are also non-renewable and createfew direct jobs, so the basis of the upswing hasbeen narrow. o create and sustain wealth in thelong-term these diminishing resources have to beconverted into other forms of capital, preferably

tradable industries producing final consumerand capital goods that will outlast the primarycommodities. Otherwise, resource-based growthcould remain an enclave separate from the restof the economy: Commodity exports can leadto high but not sustained economic growth(UNCAD, 2011a, p.4). Te Economist asked:Will Africa continue to rise? Or is this merelya strong upswing in a boom-bust cycle that willinevitably come tumbling back down? (2011).

Since 2008 the triple global crises of rising foodprices, energy prices and financial turmoil havealready eroded some of Africas previous gainsand exposed its vulnerability to external shocks(UNCAD, 2011a). Africa receives less overseasaid than it pays out for oil imports, accordingto the International Energy Agency (Mail andGuardian, 5th April 2012). In addition, thereis evidence that the Africas recent spurt of highgrowth in output has not helped to reduce povertyby an equivalent amount (IMF, 2011).

Africa is the least diversified region in theworld in terms of its exports, and has madeslow progress in the last two decades. Teexport diversification index improved slightlyfrom 0.61 in 1995 to 0.58 in 2009, while in

Asian developing countries it improved from0.32 to 0.26 and in developing America itimproved from 0.36 to 0.33 (UNCAD,2011a). Te IMF (2011) agrees that Africas

goods exports remain unsophisticated (see alsoAfrican Development Bank, 2007; Ajakaiyeand Ncube, 2010), although there has been

some progress with services. Africas problem isthat it still supplies basic inputs to global valuechains that are mostly located and controlledfrom elsewhere. Diversification requires movingup the value chain by refining and processing

natural resources, creating improved productsand integrating different stages of valueaddition to develop more balanced economies.Tis is a contested process threatening foreigncompetitors and challenging the tariff barriersand regulations that affect the way exports from

Africa are treated in advanced economies.

Te recent shift in patterns of foreign tradeand investment from Europe and North America

towards China and other parts of Asia alsoprompts questions about whether this is a newscramble for Africa (Financial imes, 25th

August, 2011). China could be the new imperialpower exploiting African land, labour andminerals in much the same way that Europe didin the past. Te latest evidence suggests that mostChinese FDI to sub-Saharan Africa has consistedof investments in natural resources packaged

with related infrastructure projects (IMF, 2011).

Chinese investors and construction companies tendto employ their own workforces, which limits theextent of technology and skills transfer to Africans.Tere is also a huge imbalance between the patternsof trade between China and Africa.

Yet some observers argue that there aregenuine prospects for more mutually-beneficialEast-South or South-South patterns of tradeand development (Ampiah and Naidu, 2008;Murray, 2008; UNCAD, 2011a). Asianinvestors, including state-owned enterprises,may well take a longer-term view of Africasprospects than Western investors. UNCAD(2011a) believes that Africa could in due coursebecome a supplier of manufactured goods,inputs to infrastructure and agro-industry tothe rapidly expanding middle classes in Chinaand India. Te influential World InvestmentReport also identifies opportunities for

Africa to benefit from new global trends in

foreign direct investment arising from theemergence of a wider array of production andinvestment models (UNCAD, 2011b). Tese


14/62

6


include contract manufacturing and farming,outsourcing of services (such as business processoutsourcing (BPO)), franchising and licensing.

Securing these parts of global value chains

to strengthen domestic productive capacitywill depend on appropriate policy frameworks,reliable infrastructure and sound institutions.Cheap electricity from home-grown oil andgas supplies and renewable energy sources willobviously help. Public procurement couldalso play a role, using the purchasing power ofgovernments over infrastructure and other formsof investment to negotiate higher levels of localproduction by foreign corporations, instead

of imported goods and components. Possibleexamples of technologically realistic productsinclude fertilisers, chemicals, plastics and otherpetroleum products, construction materials, plantand equipment, pipes, electrical cables, pylons,

buses and railway rolling stock. Governmentscould also seek to negotiate a progressive increasein the amount of local manufacturing content formultinationals that are seeking access to expanding

African consumer markets. Tis could cover allkinds of consumer durables, clothing, furniture,processed foods and even pharmaceuticals, healthvaccines and medical devices, where governmentstend to be the biggest customers.


15/62

7

SECTION 3: THE ROLE OF GEOGRAPHYAND URBANISATION

Considerations of economic geography, spaceand location have barely featured in recent reportson African economies, except for the idea ofinfrastructure corridors and lower internal tariffbarriers to promote intra-African trade (EconomicCommission for Africa, 2012). UNCAD (2011a)develops a compelling argument for structural

change from agriculture to manufacturing industryin Africa, combined with the building of strongerinput-output linkages between sectors, generalupgrading of product quality, and more efficientsystems to connect producers to markets. However,it does not mention how spatial proximity, urbaninfrastructure and institutions can contribute to theseprocesses. Similarly, the World Investment Report(UNCAD, 2011b) discusses the need to increasethe stickiness of foreign investment in host

economies by building up domestic suppliers,technological learning, upgrading of skills andentrepreneurship. Again no reference is made tothe spatial dimension of this process of integrationand embedding.

Te IMF economic outlook (2011) includes table1 below which shows that, in a sample of countries,the average rate of employment growth in urbanareas over the last decade was more than double thenational rate. Yet the accompanying analysis says verylittle about sub-national development patterns, andthe policy recommendations ignore urban growth

opportunities or constraints. It is striking how suchreports can neglect both the extra costs of productionand trade arising from dispersed geographical patterns(the friction of distance), and the multiple benefitsof company co-location in strengthening industrialsynergies, productivity, innovation and access todiverse skill-sets. Place is where complementaryactivities come together on the ground, wherephysical obstacles such as access to serviced landare most apparent, and where business interactions

and public infrastructure investment are mostefficient. Geography, space and cities should notbe regarded as mere outcomes of industrialisation,inert containers of economic activity, or passiverecipients of investment.

Source: IMF, 2011, based on national household surveys. Employment is defined as all income generatingactivities rather than just formal employment. Despite strong growth, the employment rate remains verylow by standards elsewhere in the world.

TABLE 1: Indicators of employment growth for selected African countries

Country Period Total employment(annual change)Urban employment

(annual change)

Formal employ/working age pop.

at latest dateCameroon 2001-07 2.7% 5.6% 9.5%

Ghana 1999-2005 3.4% 6.1% 13.3%

Mozambique 2003-09 4.4% 7.4% 16.7%

Tanzania 2000-09 3.3% 8.8% 9.5%

Uganda 2002-09 7.5% 9.8% 13.9%

Zambia 1998-2004 1.9% 5.1% 13.8%

Overall sample 3.3% 6.8% 13.6%


16/62

8


African urbanisation remains a sensitivepolitical issue because of its complicated historyand negative consequences for contemporaryliving conditions (Freund, 2007). In manycountries urban growth seems to have been driven

more by political than economic factors, such ascivil wars and post-independence public spending(Bekker and Terborn, 2012; Bryceson and Potts,2006). Most governments are wary of appearingto encourage rural-urban migration in the beliefthat it is harmful and not rational for householdsor the country as a whole.

Ignoring spatial patterns can undermine theviability of economic development and jeopardise

the sustainability of human settlements. Animmediate issue is whether the location ofpopulation and resource-based economicgrowth coincide, bearing in mind the statisticalprobability that the distribution of mineralsand other natural resources will not correspond

with the location of most existing urban centres.Migration flows tend to adjust slowly andimperfectly to uneven economic growth. Tisspatial misalignment means that the first priority

for public investment is likely to be long-distancepipelines, transport networks and power generationto facilitate the extraction and exporting of the rawmaterials, rather than infrastructure to improve theoperational efficiency and social conditions in cities.Governments and commercial investors will beinclined to focus on the former because it will yieldthe biggest financial returns in the short run.

In addition, experience suggests that growthbased on extracting commodities and using theexport proceeds to pay for imported consumergoods does not generate proportionate amountsof employment, either within cities or anywhereelse (UNCAD, 2011a). Tere is a large literaturesuggesting that mineral exploitation is a uniquelydifficult form of national development, oftenproducing dependent or truncated outcomesrather than diversity and long-term sustainability.For example, there is considerable evidence thata resource curse may undermine broad-based

economic development efforts by crowding outthe growth of other sectors through an inflatedexchange rate, high capital intensity and barriers to

entry. In addition, factional or predatory politicalstructures often arise in the rentier economies ofresource-abundant states, which inhibit otherforms of economic development (Bridge, 2008).

Something analogous may be occurring rightnow in many African cities. Well-endowedpolitical and economic elites and expatriatesemployed by foreign corporations are forcing upthe price of housing, vehicles, consumer goodsand food. Large amounts of money in circulationalso encourage speculative land acquisition andluxury property development. Te effects arereinforced by land market inefficiencies and aconstrained supply of serviced sites available

for development. Inflated property costs andhigher consumer prices make the cost of livingless affordable to ordinary citizens and constrainthe growth of the real economy and domesticindustry. Uncertain property rights and informaltransfer systems can also cause distortions, such ashousing market bubbles. And the lack of tenuresecurity makes it easier for poor households to beevicted from well-located land, with nothing toshow from rising asset prices (UN Habitat, 2010).

City and national authorities may support suchevictions on the grounds that informal settlementsare out of place and shouldnt be tolerated in acontemporary, modern urban setting, wheregeneral environmental standards should be higher.

For example, as a result of Angolas oil boom,Luanda is reputed to have become the mostexpensive city in the world according to cost-of-living surveys produced by agencies suchas Mercer. Tis reflects the very high cost ofaccommodation, groceries, fuel, restaurants,hotels, care hire and local taxes (Redvers, 2012).Tere has been a construction boom consistingof up-market residential property and offices,

with government marketing efforts describingthe city as an African Miami and a West AfricanDubai under construction. Yet 75% of Luandasresidents have no access even to piped water onsite and 90% have no waterborne sewerage,despite the countrys vast oil wealth (Bekker

and Terborn, 2012). Grant (2009) describessomething similar in Accra, where a minority ofnouveau riche Ghanaians and foreign workers


17/62

9

SECTION 3

THE ROLE OF GEOGRAPHY AND URBANISATION

live in sprawling gated communities, while thepoor majority of the population are confined toseverely overcrowded squatter settlements lackingrudimentary infrastructure and services.

Tere is a growing realisation that Africaneconomies need above all to diversify by addingmore value to their natural resources before theyare exported (Economic Commission for Africa,2012; UNCAD, 2011a; IMF, 2011). Similarly,

African cities need to become much more thanarenas of luxury consumption and centres of publicadministration. Diversification would involvedeveloping upstream and downstream activitiessuch as refining, processing, beneficiation and

supplying inputs to mining and manufacturing.Te idea would be to gradually shift up the valuechain over time towards more technologically

advanced components, producer services (such asengineering, research and marketing) and productdesign and development. Tis would create moreintegrated economies in terms of backward andforward linkages, and ensure that any stimulus

to growth generates larger multiplier effects,substantially more jobs and a broader spread ofincomes throughout the economy. Diversification

would also strengthen economic resilience byreducing exposure to volatile commodity pricesand unpredictable levels of demand. It wouldtake a concerted effort and require extensiveinvestment in infrastructure, institutions andskills, as well as negotiations with global originalequipment manufacturers to invest directly

in domestic production or to establish jointventures with local firms to share knowledge,experience and risk.


18/62

10


SECTION 4: CAN CITIES HELP TO BUILD MOREINTEGRATED ECONOMIES?

One of the urgent questions to consider is whetherurbanisation can help this process of building morecoherent economies? Can the expansion of Africancities create conducive environments to fosterbroad-based growth through lower transaction andtransport costs, more intense local trading patterns,improved access to skills and expertise, stronger

collaboration between firms, mutual learning, andother dynamic interactions between economicagents which help to raise productivity and spurinnovation? In a context of widespread informality,can the rapid expansion of population in Africancities be managed in such a way that this becomesa positive force helping economies to develop andlifting people out of poverty? More specifically:

How can the productive potential of cities,

particularly their infrastructure and institutions,be harnessed to support value addition,diversification and structural change in Africaneconomies?

How can the concentrations of consumerspending power in big cities stimulate increasedlocal production of goods and services, ratherthan sucking in cheap imports?

Can the provision of shelter and householdinfrastructure in informal settlements beorganised in such a way as to support incomegeneration and economic expansion?

How can the climate change agenda, rising oilprices and environmental pressures be used ascatalysts to grow the green economy, developcleaner renewable energy sources and createenvironmentally-sustainable employment?

And how can the mass of existing small-scale,informal enterprises be developed into moreproductive, growth-oriented businesses that

generate more secure livelihoods, decent jobsand tax revenues to fund improved publicservices and infrastructure?

In short, what can governments do to help Africasfast-growing cities increase their contribution toall-round development, and thereby raise incomes

and living conditions for their residents and forother parts of the country? Tis is likely to requiresome combination of policy action at the local levelaimed at the built environment, institutions andskills (integrated place-making), together withbroader actions aimed at strengthening the positionof local producers in wider economic networks andvalue chains. Te latter is critical bearing in mindthe increasing openness of national economiesunder conditions of globalisation and the growth

of foreign trade, investment and competition.Tese sectoral and spatial dimensions are usuallyanalysed independently. Greater insights andlessons for government policy could be generatedby examining them together and linking them inpractice. Tis short report cannot possibly answer allthese ambitious questions. Instead it seeks to lay outsome of the groundwork by reviewing the relevantarguments and evidence.

A growing body of economic theory andinternational evidence suggests that the geographyof the economy matters for prosperity and humanprogress, and that cities can make a positivecontribution to development. We look in detail atthis evidence in the next section. For the moment,a word of caution is important. Tere are manyexaggerated claims made about the inevitable and

wide-ranging benefits of cities as incubators ofinnovation, centres of civilisation, and so on. Tecity as an engine of growth has become a policy

mantra that ignores other vital conditions. Te titleof leading economist Ed Glaesers recent book is agood example of such hype: riumph of the City:


19/62

11

SECTION 4

CAN CITIES HELP TO BUILD MORE INTEGRATED ECONOMIES?

How Our Greatest Invention Makes Us Richer,Smarter, Greener, Healthier, and Happier (2011).

Another example is provided by Monitor:

the economic future of sub-Saharan Africa

is more connected to the success of its cities, andthe competitive clusters based there, than toits nation states. Cities today generate most ofthe subcontinents wealth, with many thrivingdespite obvious challenges. Rapid urbanizationturbocharges economic growth and diversification,enhances productivity, increases employmentopportunities, and improves standards of living

(Monitor, 2009).

Such titles and slogans are partly designed tocounter the anti-urban bias that exists in manysocieties (and in academic writing), reflecting thesocial and environmental problems that afflictedold industrial cities and continue to affect manydeveloping cities. Tis is understandable. However,their blatant one-sided analysis can also discredit thecase for supporting urban concentration, particularlyin Africa.

Tere is substance to the positive arguments, as Ishow in the next section. At the simplest level, urbandensity improves trade by creating busier (thicker)markets. As settlements get bigger firms specialiseand efficiency improves. Growing concentrationsof activity, resources and ideas have the potential toincrease productivity further, stimulate enterprise,expand output and raise household incomes. Tiscan be beneficial to aggregate growth, i.e. to nationalas well as local economies, although the magnitude ofthese agglomeration economies is subject to debate.Cities also have the capacity to foster creativityand ingenuity, which increases the adaptability ofeconomies, promotes diversification, and avoidslock-in to outmoded activities and stagnant markets.In short, large urban centres can serve as agents ofdevelopment, enhancing technology and economicprogress, and stimulating consumer spending anddemand.

A crucial point to bear in mind, however, is that

these positive externalities are far from automatic.Tey are also not the outcome of market forcesoperating in isolation. Tey depend crucially on

the role of governments in regulating private actors,making sizeable investments in infrastructure andbuilding appropriate support institutions. Cities inthe developed North have a very long history ofsuch government activism and investment, which

cities in the global South cannot suddenly match.Te significance of agglomeration economies incities where enterprises are predominantly informal,general skill levels are low and state capabilitiesare limited is also uncertain. And past research onagglomeration has generally overlooked the unequalbenefits for different population groups, assuminginstead that everyone gains from growth.

Tere is evidence that the direct benefits of

agglomeration can extend beyond city boundariesand peri-urban areas (Kessides, 2006). Smalltowns and rural areas may benefit through cashremittances from migrant labour, dispersal of lowervalue activities from cities, access to expanding urbanmarkets and sources of investment, and connectionsto wider national and international networks oftrade, finance and information. Many nationaland local governments seem insufficiently awareof these spread effects or centrifugal forces, and of

the synergies that can develop between cities andsurrounding areas. Instead urban and rural areas tendto be seen as mutually exclusive and in competition

with each other. Te success of cities is perceived todraw people, investment and other resources awayfrom towns and rural areas, thereby underminingtheir potential. In other words, the centripetal forcesof concentration are perceived to dominate.

Te relative strength of these forces of dispersal andconcentration is the subject of ongoing research anddebate among academic experts. Tere are differentschools of thought on the subject that tend towardscontrasting conclusions. Te New EconomicGeography (NEG) suggests that the spread effectscome to dominate over time, leading in due course tothe elimination of income differences between places(e.g. Krugman, 1991; Glaeser, 2011; World Bank,2009). Evolutionary, Keynesian and institutionalapproaches to economic geography suggest thereverse, namely that spatial disparities tend to

widen over time because of the self-reinforcingadvantages enjoyed by the core cities, which producea cumulative spiral of relative economic prosperity


20/62

12


and in-migration of skills (Garretson and Martin,2010; Pike et al, 2010). Tis is one of a numberof critical issues with a bearing on ultimate policyrecommendations.

Posing the complex dynamics of spatialdevelopment as a simple urban versus ruraldichotomy tends to polarise opinion in favour ofone type of area at the expense of the other. Peopleend up either pro-urban or pro-rural. Tis doesnot help to understand the changing relationships

between cities, towns and rural areas, or to developmore effective policies. Tere are potential conflictsbetween core and peripheral areas, but there arealso synergies and complementary interactions thatshould not be ignored. Te ultimate outcome is not

a foregone conclusion and is likely to vary in differentcontexts and over time. Te policy challenge is todevelop mutually-supportive linkages between cities,towns and rural areas, just as it is to strengthen thelinkages between agriculture, industry and servicesectors of the economy.


21/62

13

SECTION 5: AGGLOMERATION ECONOMIESIN THEORY

Over the last decade, a growing number ofinfluential international organisations have come tothe view that cities promote economic development(OECD, 2006; United Nations, 2007; UN-Habitat, 2008, 2010; World Bank, 2009). Teirarguments are generally broad-brush and theextent of supporting evidence is often limited. Te

connection between urbanisation and economicdevelopment is often portrayed as immutable andinevitable - a kind of universal law. Supporting citiesis sometimes presented as a formula for achievingeconomic success. Insufficient attention is given tothe specific ways in which cities can contribute toeconomic progress, the circumstances in which thesemechanisms may not work, and the strength of theseeffects compared with other drivers and influenceson economic development. Quite different causal

processes are often lumped together in a catch-allassertion, making it difficult to unpack and test thedifferent dynamics.

Te resulting messages for government policyhave ended up rather broad and generalised, withlittle sense of priorities and sequencing, and thereforedifficult to follow through and put into practice. Terecommendations have often amounted to a call fora greater focus on cities because of their potential tomaximise national economic growth, irrespectiveof the way public resources are actually distributedspatially within that particular country. Te existenceof an anti-urban bias is readily assumed, even thoughpro-rural sentiments may amount to little more thanrhetoric in practice. Another recommendation isfor better planning and management of urbangrowth to avoid the chaotic proliferation ofinformal settlements. Tis is commonsense, butit doesnt help governments to work out how tostrengthen urban economies. For example, it is

not obvious that upgrading informal settlementsis the most cost-effective way of acceleratingeconomic development.

A common refrain is that: Te prosperity ofnations is intimately linked to the prosperity oftheir cities. No country has ever achieved sustainedeconomic growth or rapid social development

without urbanising (UN Habitat, 2010, p.x).Similarly,

No country has grown to middle incomewithout industrializing and urbanizing.None has grown to high income withoutvibrant cities. Te rush to cities in developingcountries seems chaotic, but it is necessary

(World Bank, 2009, p.24).

Economic progress is said to have both dependedon more people living in cities (to expand the supplyof labour and entrepreneurs, and to stimulate

mutual learning and creativity) and it has generatedthe resources to support continued urbanisation(through essential infrastructure and services). Teoutcome of this virtuous circle is said to have beenrising national productivity and prosperity, andreduced overall poverty. While this argument mayseem plausible at a general level, the explanationfor the connection between urbanisation andeconomic development is limited and there is a nave(and incorrect) assumption that one thing followsautomatically from the other. Africas experienceshows that many other factors are involved thatcomplicate the relationship and can cause contraryoutcomes in the real world. Urbanisation does notnecessarily go hand in hand with economic growth,employment and higher living standards. Indeed, the2010/11 State of the Worlds Cities acknowledgesthis: when accompanied by weak economic growth urbanisation results in local concentration of poorpeople rather than significant poverty reduction(UN Habitat, 2010, p.x). Tis phrasing even

suggests that urbanisation may have little to do ina causal sense with economic growth it is a mereaccompaniment.


22/62

14


Sometimes the arguments are historical - structuralshifts in the economy, particularly industrialisation,have gone hand-in-hand with a spatial redistributionof the population the urban transition (WorldBank, 2009). Te industrial revolution in the 18th

and 19th centuries was certainly accompanied bylarge-scale, and in many cases forced, movements ofpeople off the land and into burgeoning cities. Tislesson of history probably cannot be reproduced inmany (democratic) countries today without a severepolitical backlash from rural areas. More importantly,the strong spatial concentration of industry duringthe 18th and 19th centuries occurred whentransportation systems were undeveloped andtransport costs were very high. Contemporary

conditions are quite different in most places.

China is widely regarded as the best contemporaryexample of how rural-urban migration can fuelindustrialisation and boost living standards (WorldBank, 2009; Ravallion, 2009). China passed thehistoric milestone of 50% of its population livingin cities in 2011, up from only 20% in 1980(Financial imes, 18th January 2012). Te speedof urbanisation has reflected above all the strength

of employment growth in cities without jobspeople wont go. Average household incomes inChinese cities are now almost three times higherthan in rural areas, reflecting higher productivityaccompanying strong employment growth. TeChinese governments approach to urbanisation,particularly its commitment to invest heavily inurban infrastructure, is also contrasted with themore informal, disorganised approach of India andother developing countries, where congestion, watershortages, squalor, disease and conflict are rife: Forevery pound Indian authorities invest in urbaninfrastructure, their Chinese counterparts spendseven (Observer, 22nd January 2012). Althoughit is not a democracy, the Chinese government hasalso been forced to counter concerns about ruraldisadvantage with major commitments to invest inrural economic and social development in recentyears, including infrastructure, schools and pensions.

Te basic insights into the advantages of economic

concentration have been around for a long timeand can be traced back to many of the founding

fathers of economics. What has changed recentlyis that the analytical techniques have become morecomplicated and the theoretical frameworks moreelaborate. Work on geographical economics hasalso developed a much higher academic profile than

traditional economic geography, reflected in theaward of the Nobel Prize to Paul Krugman in 2008for his work in this field. Many of these economicmodels are derived from first principles, rather thanempirical observation. Tey are specifically designedto apply anywhere and everywhere, and to includeonly a few fundamental forces: the key imperativeof NEG is to devise a universal grammar of spatialagglomeration in terms of formal models thathave general applicability but also representational

parsimony, that is as few basic causal factors andrelationships as possible (Krugman 2000)(Martin, 2008, p.7). Tis gives them a law-likecharacter, which has contributed to the kinds ofover-simplification and generalisation noted at thebeginning of this section by policy-makers seekingto interpret this work. Tese models cannotexplain historical and geographical variations, andapparent anomalies such as African urbanisation.Tey are not meant to represent or explain any

specific real-world situation: they consist ofhypothetical worlds built upon and defined bysimplified assumptions and idealized geographicallandscapes (Martin, 2008, p.7).

Te economic rationale for urbanisation isunderpinned by two basic concepts - the divisionof labour and economies of scale. Te former wasintroduced by Adam Smith and explains the benefitsfor productivity and therefore growth that arise fromspecialisation among producers. Specialisation meansindividuals and firms concentrating on particularproducts or tasks, which yields greater efficiencies,skills, expertise and variety. Tis accounted for thegreat leap forward from craft production to factoryproduction that gave rise to the industrial revolution.Te idea of specialisation may be part of whatsrequired to improve the performance of informaltraders in Africa today, bearing in the mind thelevels of duplication and imitation that often exist.Te concept of specialisation also has some relevance

at the city-wide level, in terms of the benefits offocusing on particular functions for which they have


23/62

15

SECTION 5

AGGLOMERATION ECONOMIES IN THEORY

some inherent advantage. Specialisation arguablybecomes more important for cities to prosperunder conditions of increasing external trade andglobalisation, when competition is more intense.

Economies of scale is the second principle. It tendsto give this body of work a reductionist character that size matters above all else. Tere are two aspects.Internal economies of scale are internal to the firmand relate to the lower unit costs or efficiencies thatresult from larger scale production. Larger firms canbuy their inputs at lower prices and can spread theirfixed costs (industrial plant, equipment, rent, rates,marketing, management, R&D etc) over a largervolume of output. External economies of scale (or

agglomeration economies) relate to the benefitsthat firms derive from locating near to other firms(their customers and suppliers) in order to reducetransport and communication costs, and to gainfrom network effects (or positive externalities such

as shared information). Te more firms there are inthe network, the more knowledge and intelligencepotentially available to learn from. Agglomerationeconomies also include proximity to a large labourpool, suppliers, customers and competitors withinthe same industry, and firms in other industries.

Alfred Marshall (1920) was the first economist torecognise the benefits for economic agents (workersas well as firms) of having access to a reservoir ofinformation and ideas, skills and shared inputs.


24/62

16

SECTION 6: DIFFERENT ADVANTAGESOF AGGLOMERATION

Tese economic gains from spatial concentrationcan be summarised as three broad functions:matching, sharing, and learning (Duranton andPuga, 2004; Storper, 2010). First, cities enable firmsto match their distinctive requirements for labour,premises, suppliers and business services better thansmaller towns, simply because markets are larger

and there is a bigger choice available. A bigger poolof providers also tends to reduce costs and improvevariety as a result of competition and specialisation. Ina volatile and dynamic economy there is a premiumon flexibility and adaptability to shifts in marketsand technologies, especially as companies tend to beleaner, more focused on core competences, and morereliant on buying-in goods and services rather thanin-house production (Buck et al, 2005; Scott, 2006).

Agglomerations enable firms to mix and match

their resources more easily in an uncertain operatingenvironment. Tese opportunities and interactionslower the cost of transactions, help companies toreorganise and grow more quickly, and thereforeimprove their resilience. Ease of staff recruitment andreplacement are especially important for activitiesthat tend to have a high labour turnover, such as callcentres. Labour market matching can benefit workersas well as firms by ensuring a better fit with their skillsand aspirations, and higher earnings because of theresulting productivity gains. Tis encourages in-migration from smaller towns and rural areas.

Second, cities also give firms access to a biggerand better range of shared services and infrastructurebecause of the larger population of firms andoverall scale of activity. Cities offer better externalconnectivity to national and global customersand suppliers through more frequent transportconnections to a wider range of destinations, andmore efficient logistics systems to handle imports

and exports. Tey tend to have higher capacitybroadband and telecommunications systemsfor electronic communication and marketing.

Tey have a better choice of specialised technicalsupport, professional expertise, financial know-how,engineering advice, or research and developmentorganisations to assist with product design andimprovement, and to assist companies to stay up-to-date with changes in technology and markets. Citiesalso have a wider range of education and training

organisations to help with staff development and theacquisition of specialised skills and capabilities.

Tird, firms stand to benefit from the superior flowsof information and ideas in cities, which promotemore learning, creativity and innovation, and resultsin new and more valuable products and processes(Jacobs, 1969, 1984; Hall, 1998). Agglomerationmay be vital for high-end knowledge-intensivefunctions and technologically advanced activities

that differentiate themselves from competitorsby continuing to create higher quality goods andservices. Proximity can facilitate communicationand sharing of complex ideas between companies,research centres and related organisations (Cookeand Morgan, 1998; Storper and Manville, 2006;Scott, 2006). It enables people and firms to compare,compete and collaborate, which may establish a self-reinforcing dynamic that spurs creativity, attractsmobile capital and talent, and generates growthfrom within. Brainstorming, mutual learning andexchanging tacit knowledge tend to be more effectiveface-to-face than remotely through electroniccommunication (Storper and Venables, 2004).Close contact enables formal and informal networksof technical and scientific staff to emerge, whichcan encourage all sorts of collaborative projects.

And new ideas are generally formed by refining andcombining old ideas products, services and placesthemselves can be reinvented. Cities epitomise theprocess of endogenous growth whereby resources are

used more productively and in new ways (Kessides,2006, page 13). Tese dynamic advantages becomeincreasingly significant over time because they are


25/62

17

SECTION 6

DIFFERENT ADVANTAGES OF AGGLOMERATION

cumulative, compared with the one-off or staticadvantages gained from lower production andtransactions costs. Many authors attach particularimportance to ideas and information for cities inthe West to prosper: Not every city will succeed,

because not every city has been adept at adaptingto the age of information, in which ideas are theultimate creator of wealth (Glaeser, 2011, p.40; seealso Castells, 2000).

Economies of scale also apply to the sphere ofconsumption. Te combined spending power oflarge concentrations of population tends to stimulatenew consumer goods and amenities, such as leisureand recreation activities, which can attract further

rounds of job-creating investment, tourism andpopulation growth (Glaeser and Gottlieb, 2006).Cities are more likely than other places to containthe cultural vitality, social infrastructure and careeroptions to help regions and nations attract the skillsand talent required to generate and exploit knowledgeand build dynamic competitive advantage. Someamenities can only be sustained in large cities, suchas major entertainment venues or specialised centresof education and health. Cities also offer a greater

choice of facilities to attract people to visit, study,live and work: Cities thrive as places of pleasure aswell as productivity (Glaeser, 2011, p.10). Glaeserattributes Londons recent growth to its status as aluxury resort or playground for the rich, while thesuccess of Houston and Dallas is attributable to theirlow housing costs and good weather.

Research on the agglomeration economies ofconsumption has also formed part of a broaderargument that the fundamental driver of urbaneconomic growth is the movement of populationto cities, rather than firms (Glaeser, 2011). Te basicproposition is that jobs follow people rather thanpeople following jobs. Urban amenities and thequality of place are held to be particularly importantinfluences on the location behaviour of individualsendowed with high levels of human capital. Andhuman capital, far more than physical infrastructure,explains which cities succeed (Glaeser, 2011, p.27).Similarly, Te magic of cities comes from their

people, but those people must be well served by thebricks and mortar that surround them. Cities needroads and buildings that enable people to live well

and to connect easily with one another (p.160).Such people have particular consumer or lifestylepreferences that are best satisfied in cities. Henceemployers looking for the entrepreneurial, creativeand innovative energies of these groups have little

choice but to invest in cities. In Europe and NorthAmerica, many city governments have been attractedby the argument and invested heavily in consumeramenities in the hope of attracting youthful talent,professionals, entrepreneurs and other members ofthe so-called creative class (Florida, 2004). Tis hasbeen controversial because of its apparent elitism anddoubts about the fundamental logic that jobs followpeople, which ignores the significance of firms andindustries (Peck, 2006; Storper, 2011). Glaesers

discussion of Londons recent growth makes nomention of the enormous growth of the financialservices industry in attracting people to Londonand generating the income that sustains its manypleasures (Storper, 2011).

Te debate about people or jobs as the basic driverof local growth resonates with another importantdebate in spatial economics between people-basedand place-based policies to promote economic

development (World Bank, 2009; Barca et al, 2012;Glaeser, 2011; Storper, 2010). Briefly, advocates ofpeople-based policies emphasise the importanceof individual attributes, such as education, skills,cultural background and family characteristics, indetermining local economic conditions. However,advocates of place-based policies emphasise widerattributes, such as the industrial structure, physicalconditions and historical development of the area.

We return to this important distinction later.

Te argument about population being a potentialdriver of economic development is relevant tothe context of African urbanisation. It promptsthree questions: (i) does population in-migrationnaturally stimulate a broader range of activities thanconsumer services, (ii) if not, could it be harnessedby government as a force to spur stronger economicdevelopment, and if so (iii) what is currentlyimpeding this process, and therefore requires policyattention? Without providing any evidence, Beall et

al (2010, p.6) argue that new migrants to cities indeveloping countries can create new opportunitiesand new needs; offer new skills and new perspectives,


26/62

18


and generate new requirements for institutionalinnovation. In other words, in-migration can help tostimulate growth and development.

In addition, economies of scale can apply to

many public services and investments in householdinfrastructure. It is obviously cheaper to provideand operate public services such as hospitals, waterreservoirs, sewage treatment facilities, electricitygeneration and refuse collection in large cities than indispersed settlements where population densities arelow and distances are large (United Nations, 2007;Martine et al, 2008). Te network effects of publicgoods can also be maximised where there are largenumbers of service users (Overman and Venables,

2010). Tere is a related argument that urban shelterand infrastructure projects generate greater positiveexternalities in cities than in towns and rural areas

because they both (i) increase the productive capacityof households (particularly through improved healthand life expectancy) and (ii) increase the effectivelabour supply, which enable higher rates of city-

wide economic growth. Kalarickal (2007) provides

examples of how public investment in low costhousing, sanitation, drinking water, electricity andaccess roads in informal settlements in Africa helpedto reduce public health problems and mortalityassociated with overcrowding, and facilitate in-migration, which assisted in addressing some ofthe binding constraints to economic growth.He makes the point that governments andinternational donors must look beyond theindividual project when assessing their costs and

benefits because of the significant externalitiesassociated with investments in cities.


27/62

19

SECTION 7: DIFFERENT FORMS OFAGGLOMERATION

Economic concentrations can also take differentforms, and be driven by different dynamics,depending on the underlying rationale for firms tocongregate together. Recent efforts to disentanglethese different agglomeration processes were partlyprompted by the work of the worlds leadingadvocate of industrial clusters. According to Michael

Porter: Te enduring competitive advantages in aglobal economy are often heavily localised, arisingfrom concentrations of highly specialised skills andknowledge, institutions, rivalry, related businesses,and sophisticated customers (1998, p.90). Porterdefined clusters as geographic concentrations ofinterconnected companies, specialised suppliers,service providers, firms in related industries, andassociated institutions (e.g., universities, standardsagencies, trade associations) in a particular field that

compete but also co-operate (2000, p.15). Te twokey features of this are that companies are linkedin some way (through inter-firm trade, or the useof common inputs, or shared techniques) and arelocated in proximity. Porters central argument wasthat areas with strong inter-firm and institutionalrelationships of this kind are more internationallycompetitive, innovative and capable of growingfaster than places with weaker local connections.Tis argument was not essentially new, but it

was articulated more forcefully and with greaterimpact on the international policy communitythan ever before.

Porters definition set no clear boundaries toeconomic clusters, either in terms of their industrialcomposition or their geographical extent (Martinand Sunley, 2003). It was vague about the degreeof aggregation of firms and industries, the strengthof linkages between firms, and the spatial scale andintensity with which clustering processes operate

(for additional criticisms, see aylor, 2010). Moreimportantly for present purposes, the concept failedto distinguish between different kinds of forces

promoting economic concentration. It conflateddifferent processes operating at different geographicalscales into a single, all-embracing notion, which madeit difficult to validate his arguments empirically.

Tere are three distinctive forces that promoteeconomic concentration (Gordon and McCann,

2000; Storper, 1997; Malmberg and Maskell, 2002).Te first involves relatively stable, tangible tradinglinkages between firms. Te physical outputs of onemeet the input requirements of another, so firmsare bound together in production chains with theirsuppliers of intermediate inputs. With sufficient scale,this co-location of firms within the same value chaincan create relatively integrated industrial complexes.Examples in practice include petro-chemicals, agro-processing, minerals processing, heavy engineering

or steel complexes, where many of the inputs andoutputs are bulky or perishable. Te main reasonfor firms locating close together is to minimise thecosts of transport, distribution and communicationsbetween themselves. Tis may be reinforced bythe growing imperatives in some industries ofshorter product cycles, responsiveness to volatilemarkets and just-in-time production systems. Tesepressures place a premium on locational proximityto ensure security and speed of supply. Tey areoffset by reductions over time in transportcosts and improvements in communicationstechniques and transport technology, such ascontainerisation. Efficient internal and externaltransport infrastructure, reliable logistics systemsand responsive bureaucratic procedures are vitalfor the productivity and competitiveness ofindustrial complexes.

Tere is generally a higher degree of predictability,continuity and planning in the relationships

between firms in industrial complexes than in moreatomised agglomerations. echnologies are likely tobe more mature and processes more routine than in


28/62

20


other concentrations of activity. Consequently thenetworks are likely to be smaller and less open to newentrants. Where it occurs, innovation may also takethe form of joint ventures between firms and theirsuppliers, rather than one-off projects arising from

chance opportunities and spontaneous events. Tereare generally fewer firms and fewer opportunities fornew combinations of inter-firm relationships toemerge within the complex. Te most importantrole for government policy is probably to ensure astable business environment and efficient publicinfrastructure suited to the requirements of thekey corporations.

Te second force for concentration is the localised

network. Te core idea is that there is practical co-operation and exchange of intelligence betweenfirms and related organisations within the networkin order to promote trust and longer-term decision-making. Te emphasis is on intellectual andcognitive resources (human ingenuity, knowledge,competences, etc.) rather than physical resourcesand routine interactions. Te key agents are likelyto be knowledge-intensive firms requiring highlyspecialised skills and communicating complex

information to promote innovation and learning bydoing. Shared understanding and conventions enablecompanies to overcome some of the limitations ofmarket relationships and commercial contracts, suchas short-termism, and to undertake risky or costlyventures without fear of opportunism. Firms are

willing to act together in alliances or partnerships,or as part of a larger group of companies in supportof common, mutually beneficial goals. Tis mayextend to creating institutions to promote theircollective interests by lobbying government or otherregulators. Such institutions may also be created toprovide common support services for members ofthe network, such as marketing, training or technicalassistance. Te greater the degree of shared beliefs,values and assumptions across the group, the higherthe level of embeddedness and social integration(Granovetter, 1985).

Geographical proximity fosters some of theconditions for social interaction and collaboration

across the network (Gordon and McCann, 2000;Glaeser, 2011). Being located in the same place canhelp interpersonal relationships and trust to develop,

and promote a sense of common interest andbelonging. Geography can also help collaborativenetworks to build upon the distinctive culturalhistory, traditions and identity of places, and makeit easier for members to organise practical collective

actions. Over time, the result may be the formationof strong city- or region-specific economic clustersrepresented or partly governed by their own businessassociations. Such clusters may even contribute tothe unique image and reputation of places. Classicexamples include Silicon Valley and Wall Street inthe United States, the City of London and the TirdItaly (Piore and Sabel, 1984; Storper, 1997, 2010).Te Cape Winelands in South Africa shares some ofthese characteristics of an emerging business cluster

with strong collaboration between firms and a highlyvisible external profile.

Tird, the classic notion of agglomerationemphasises the external economies of scale or scopethat flow from firms locating within the same area.Te main benefits were originally identified byMarshall (1920): (i) firms gain from access to a moreextensive labour pool, which makes it easier to findspecialist skills; (ii) firms can gain access to a greater

range and quality of shared inputs and supportingindustries, such as specialised maintenance,marketing services, transport and communicationsfacilities and venture capital; (iii) firms gain from agreater flow of information. Tere is efficient transferof trade knowledge and expertise between firmsthrough informal contacts, chance meetings, labourpoaching and voluntary switching of employers byskilled workers and managers. Tese knowledgespillovers help to disseminate good practice andfacilitate the development of new products andprocesses. A further important distinction is betweenlocalisation and urbanisation economies. Teformer are associated with specialised infrastructure,services and skills geared to particular industriesor branches of economic activity. Urbanisationeconomies relate to cross-cutting and generalisedurban assets (such as airports, educational institutionsand municipal services, but also shared suppliers andproducer services) that serve different industries.

An important feature of this form of agglomerationis that there is no co-operation between actorsbeyond what is in their short-term interests. Tere is


29/62

21

SECTION 7

DIFFERENT FORMS OF AGGLOMERATION

no particular organisation or formal structure to thesystem, nor any special loyalty or sense of solidaritybetween firms. Tey are independent units operating

with flexibility in a competitive market environment.Locational proximity increases the opportunities for

them to trade with other firms, to recruit suitablelabour, to benefit from common infrastructure,

and to reduce market uncertainties. Te scale ofactivity and the number of firms determine thesignificance of these economic benefits basicallythe larger the better. Te density and diversity offirms may also be sources of economic dynamism

and vitality (Beall et al, 2010).


30/62

22

SECTION 8: EMPIRICAL EVIDENCE OFAGGLOMERATION ANALYTICAL PRINCIPLES

It is very difficult to disentangle and measure theeffects of agglomeration because of their enormouscomplexity, recursive interactions and feedbackeffects. Te benefits of economic concentration maybe partly absorbed by higher land and labour costs,and offset by increased congestion. Consequently, the

effects may not be apparent in aggregate economicindicators, such as output, employment or averageearnings. Tese variables are affected by a range ofother important factors and forces as well, such as theindustrial or occupational composition of the city.Te effects of agglomeration may not be apparent atthe level of city administrative units, for which spatialdata is conventionally available, because the opennessof city economies means there is considerable leakageof resources and displacement of activity across

administrative boundaries. Government financialtransfers between localities and regions also tend tomask the underlying economic processes, especiallyas they are often specifically designed to compensatefor economic weakness.

Te key element of the local economy that affectslong-term prosperity is arguably the export base ortradable sector. Tese are goods and services whichcan be easily traded with firms in other cities andregions. Te performance of the non-tradable sectoris largely dependent on local demand. Te crucialfeature of a tradable good or service is that it bringsincome into the city or region by providing a goodor service to the outside world, or provides locals

with a good or service which they would otherwisehave to import. Consequently its output does notdepend mainly on the scale of local demand. It isan independent source of income for the city orregion, and therefore a critical source of propulsivegrowth, provided it is competitive in terms of

price and quality, or the service/expertise has otherdistinctive attributes. Examples of tradables includemanufacturing and mining, the producer services

that these activities rely on (such as engineering andmanagement consultants, accountants, surveyors,architects, designers, project managers), and nationalor international tourism, banking and insurance,BPO, call centres, corporate headquarters, researchcentres, and educational facilities for foreign students.

Te crucial indicator of the performance of theexport base, and therefore the key outcome ofagglomeration economies that should be measured,is productivity. Productivity is the single mostimportant determinant of growth in economicoutput and income. It reflects the value of localgoods and services and the efficiency with they areproduced. Unfortunately statistics on productivityare notoriously unreliable at the local level, especially

for the sub-set of activities that make up the tradablesector. Instead researchers employ all sorts of otherdevices and indicators to assess the economicadvantages of agglomeration.

One of the simplest is for researchers to comparethe urbanisation levels of individual countries withsome measure of national economic output, averageincome or social development at a particular point intime in order to test for a simple statistical associationbetween urbanisation and development. Manyresults suggest that a broad empirical regularity doesexist. However, the spread of observations is alwaysextremely wide around the trend line. In addition,the discovery of a simple association does notconstitute evidence of a causal connection, i.e. thatcities are actually driving growth and development.Urbanisation may be more of a consequenceof economic development than a cause, or therelationship may be coincidental.

Tere may be some other process contributingto both urbanisation and higher levels of economicand social development in cities. One of the obvious


31/62

23

SECTION 8

EMPIRICAL EVIDENCE OF AGGLOMERATION ANALYTICAL PRINCIPLES

candidates is the concentration of political power incities, resulting in rent-seeking rather than productiveactivity, as people and firms seek preferential accessto political elites. Te outcomes could includemore public sector jobs and higher wages in cities,

together with higher levels of professional services,construction, embassies, media and consultantsseeking access to government contracts and lobbyingover legislation. Bekker and Terborn argue that thishas been the dominant factor in the growth of manycapital cities in Africa: Proximity to the centre ofpatronage and redistribution, rather than economicdevelopment, has driven the explosive growth of

African capitals since independence (2012, p.193).Glaeser adds a point about the concentration of

political power: Te more centralised a nationsgovernment, the larger its capital city, because peopleare attracted to power as ants are to picnics (2011,p.225; see also Overman and Venables, 2010).21

A series of other studies involving simplecorrelations between some measure of agglomeration(such as city size or variation in density levelsbetween regions) and some measure of economicdevelopment (such as average incomes or growth

in output) have found that no relationship exists,or even an inverse relationship (e.g. studiessummarised in Martin, 2008; see also, urok andMykhnenko, 2008).

2 For a critical interpretation of this view and a restatement ofthe economic logic of urbanisation, see Satterthwaite (2010)

Leading academics acknowledge that thepresent level of understanding of the dynamics ofagglomeration and the strength of the cause-effectmechanisms in practice remains quite limited,despite considerable theoretical development.

According to Storper, for example:

All in all, work on agglomeration has progressedconsiderably in the NEG, urban economicsand regional science. But it remains far froman adequate causal account of the dynamicsof agglomeration and de-agglomeration. Deciphering the causes of agglomeration willultimately require a much better understandingof the complex interdependencies between

agents that lead them to congregate together(2010, p.322).

According to Garretson and Martin, seriousweaknesses in the main theories of agglomerationarise because they embody crude conceptionsof geography and history (2010, p.130). Oneof the consequences is that they are unable toexplain differences in the strength and nature ofagglomeration effects in different places and atdifferent points in time. Despite being one of leading

proponents of formal economic models based onartificial assumptions, Glaeser recognises that: Notonly is there no one formula toward urban eminence,but also the sources of success are often highly nationspecific (2011, p.225).


32/62

24

SECTION 9: EMPIRICAL EVIDENCEOF AGGLOMERATION ECONOMETRIC STUDIES

Despite these shortcomings in existing conceptualframeworks, a variety of sophisticated econometricstudies in the United States and Europe have beencarried out in recent years in order to estimate theeffects of agglomeration. Tis work is complextechnically and relies on modelling and estimation

procedures that are generally inaccessible to peopleoutside this specialised field. It is also difficultto compare these studies because of significantmethodological differences, varied methods ofestimation, model specification, types of data,levels of aggregation, territorial units and of coursedifferences in local and national context. Perhaps notsurprisingly, these studies generate quite contrastingresults. Te broad finding of many of them isthat cities do indeed offer measurable economic

advantages (Eberts and McMillen, 1999; Ciccone,2002; Duranton and Puga, 2004), although they arenot as substantial or widespread as often suggested.

A useful summary of the international evidencedrawn mostly from the United States concluded thatthe elasticity of city productivity with respect to citysize is somewhere in the range 0.04-0.11 (Rosenthaland Strange, 2004). A simpler way of expressing thisis that doubling city size increases productivity by

between approximately 4% and 11%. Alternatively,for an increase of 25% in a citys population, theoutput per worker (and consequently income) risesby between 1% and 2%. Studies based on earningsdata for individuals typically find somewhat smaller,although still significant, impacts of agglomerationin big cities (summarised in Rice et al, 2006). Someevidence suggests that the impacts are larger for cities thatspecialise in particular industries or occupations. hissuggests that localisation economies may be stronger

than urbanisation economies. his finding is supportedby the various studies summarised in Graham (2007,2009). hese studies also found that the averageelasticities of productivity ranged widely between 0.01and 0.20, although most were under 0.10 (see table 2).


33/62

25

SECTION 9

EMPIRICAL EVIDENCE OF AGGLOMERATION ECONOMETRIC STUDIES

For Europe, Ciccone (2002) estimates anelasticity of productivity in relation to employmentdensity of around 0.05, i.e. at the lower end ofthe estimates for the United States. Tis meansthat for an increase of 25% in the density of jobsin a city, the output per worker rises by about1%. Fingleton (2003) reports an elasticity figureof only 0.015 for Britain, which suggests a veryminor effect. A subsequent, more detailed studyin Britain measured the elasticity of productivity

with respect to economic mass at 0.05 (Rice et al,2006). Tis is the same finding as that of Cicconefor Europe, although the methods are different.Te relevant indicator of agglomeration is the size

of the working age population in the city ratherthan the jobs density. Teir interpretation of thescale of the agglomeration advantage is that: Tis

seems modest, but its impact is important as thereare large variations in areas access to economicmass (Rice et al, 2006, p.745).

A novel feature of the Rice et al (2006) study isthat it also sought to measure the rate at which theeconomic advantages of proximity diminish withdistance from the core city. Tey found that thebenefits are greatest within 40 minutes driving time ofthe city core, tapering off quite sharply thereafter andhaving little or no effect beyond about 80 minutes.Te effects of agglomeration are four times stronger30 minutes driving-time away than 60 minutes away,and 17 times stronger than 90 minutes away. Tis is

consistent with the few other studies addressing thisissue (e.g. Graham, 2009). One of the implications isthat urban sprawl and residential decentralisation can

Source: Graham, 2007. Note: MSA = Metropolitan Statistical Area

TABLE 2: Estimates of agglomeration economies

Author Unit of analysis Independent variable Elasticity

Aaaberg (1973) Swedish cities City size (population) 0.02

Shefer (1973) US MSAs City size (population) 0.20

Sveikauskas (1975) US MSAs City size (population) 0.06

Kawashima (1975) US MSAs City size (population) 0.20

Fogarty & Garofalo (1978) US MSAs City size (population) 0.10

Moomaw (1981) US MSAs City size (population) 0.03



Nakamura (1985) Japanese Cities City size (population) 0.03

Tabuchi (1986) Japanese Cities City size (population) 0.04Louri (1988) Greek Regions City size (population) 0.05

Sveikauskas et al (1988) US MSAs City size (population) 0.01

Nakamura (1985) Japanese Cities Industry size (employment) 0.05

Henderson (1986) Brazilian Cities Industry size (employment) 0.11

Henderson (1986) US MSAs Industry size (employment) 0.19

Henderson (2003) US MSAs Industry size (no. of plants) 0.03

Ciccone & Hall (1996) US States Employment density 0.06

Ciccone (2002) EU regions Employment density 0.05

Rice et al (2006) Britain sub-regions Economically activepopulation

0.05


34/62

26


undermine productivity and growth by lengtheningtravel-to-work distances and times. Another is thattransport improvements to reduce travel timesand

Unleashing the Economic Potential of Agglomeration in African Cities

Documents

Transcript of Unleashing the Economic Potential of Agglomeration in African Cities