Download Global City Economies: Largest Cities by GDP in 2008 and Projected Growth to 2025 and more Study notes Urbanization in PDF only on Docsity!
20 • PricewaterhouseCoopers UK Economic Outlook November 2009
III – Which are the largest city economies in the world and how
might this change by 2025?
largest city economies by GDP in 2008
and 2025 in Annex B.
III.1 Long-term historic
trends in city populations
Urbanisation has been one of the major
global themes of the past century and all
the indications are that major cities will
provide an increasing focus for global
economic activity over the course of this
century. In 1900, there were only 16 cities
in the world with more than 1 million
inhabitants, mostly in the advanced
economies; now there are over 400 such
cities according to United Nations (UN)
estimates, around three-quarters of which
are in low and middle-income countries.
In 1950, the rural population of the world
was around twice the urban population,
but by 2010 the UN estimates that the
urban population will be greater and
by 2030 it projects a total global urban
population of around 5 billion compared
to just over 3 billion in rural areas (see
Figure 3.1).
Systematic rankings of urban
agglomeration populations have been
produced by the UN for the period since
1950. Table 3.1 shows the top 30 urban
agglomerations by population in 1950,
1990, 2007 and 2025 to illustrate how
these rankings have evolved over time
and how they are projected to change
by 2025. Notable points are that:
- Section III.1 provides a long-term
historic perspective on population
trends for the largest global cities
- Section III.2 introduces our approach
and methodology;
- Section III.3 presents and discusses our
estimates of the largest city economies
in 2008;
- Section III.4 presents and discusses
our illustrative projections for how these
rankings might change between 2008
and 2025, with a particular focus on the
rise of emerging economy cities;
- Section III.5 highlights the uncertainties
surrounding our projections and
discusses some of the key factors
underlying the relative growth rates
of city economies;
- Section III.6 considers the potential
impact of an alternative de-globalisation
scenario on the growth rates of city
economies; and
- Section III.7 summarises and draws
conclusions from the analysis.
A more detailed description of the
data and methodology used in the
analysis to estimate the size of city
economies as measured by GDP is
provided in Annex A. This is followed
by a full listing of our rankings of the
This article updates an article
published in our March 2007 UK
Economic Outlook^1 giving estimates
for 2005 and projections to 2020
of the size of the largest 100 city
economies in the world. The updated
analysis and illustrative projections
of GDP for different cities show how
the GDP rankings of cities might
change by 2025 taking into account
the impact of the current economic
downturn and the impact of a
potential de-globalisation scenario.
Rankings of global cities by
population are common, but while
population statistics are important,
they are only part of the story:
leading cities such as London, New
York, Paris and Tokyo are major
economies in their own right, of
a size greater than medium-sized
national economies such as Sweden
and Switzerland. Cities are also
centres of innovation, creativity
and culture, as well as focal points
for government, finance, business
services and corporate headquarters
in their respective countries (and
sometimes also their regions in the
case of financial centres like London
in Europe, or political centres like
Brussels in the EU). However, data
are much less readily available on
the overall size of city economies in
terms of their total output, particularly
outside the OECD countries^2.
This analysis fills this gap and
provides a significantly different
picture from rankings by population,
with the advanced economy cities
ranking much higher by GDP than by
population due to their higher average
income levels. Our analysis also
allows us to consider how far fast-
growing cities in emerging market
economies like China, India and Brazil
could challenge the dominance of
current leading global cities such as
New York, Tokyo, Paris and London
by 2025.
The discussion below is organised
as follows:
(^12) The present article was written by John Hawksworth, Thomas Hoehn and Anmol Tiwari. It forms part of PricewaterhouseCoopers’ wider research and consultancy programme on city economies. Some data are available for selected OECD and non-OECD cities on relative wages and costs of living, but no systematic global data source is readily available for GDP per capita at a city level as far as we are aware.
Figure 3.1 – Global urban and rural population trends and UN projections
Source: UN
Rural
Urban
Population (billions)
PricewaterhouseCoopers UK Economic Outlook November 2009 • 21
- Tokyo and New York remained the two
largest urban agglomerations between
1950 and 2007 (although swapping
places after around 1965), although
Mexico City and Mumbai had caught
up with New York by 2007. By 2025
New York is expected to be in 7th place
on a par with Kolkata;
- London was still the third largest city
in 1950, but has slid down the rankings
progressively since then to only 26th
in 2007 (with its population remaining
broadly unchanged between these
dates); Manchester and Birmingham
were in the top 30 cities in 1950 but
would not rank in the top 100 by
population now^3 ;
- Other leading European cities seeing
sharp declines in their population
rankings between 1950 and 2007
include Paris (5th to 20th), Moscow
(6th to 18th) and Berlin (from 13th to
well outside the top 30), a trend that
will continue to 2025 when Paris is
expected to rank 27th and Moscow
23rd;
- Conversely, major risers between 1950
and 2007 include Mumbai (18th to
4th), Sao Paulo (24th to 5th) and ‘new
entrants’ like Jakarta (23rd in 2007),
Dhaka (9th), Karachi (12th) and Lagos
(22nd), all of which were well outside
the top 30 in 1950;
- Notably, however, the major Chinese
cities have not seen such rapid
population rises as those in other
leading emerging markets; both
Shanghai (4th to 7th) and Beijing
(10th to 16th), while increasing their
populations significantly in absolute
terms, have slid down the rankings
between 1950 and 2007, particularly
in recent decades due to China’s one
child policy. Shanghai is expected to
continue to slide, dropping to 9th, but
Beijing should climb back to 15th;
- Overall, the aggregate population of the
top 30 cities is expected to rise from
308 million in 2007 to 391 million in
Population, however, is only one of
the factors determining the size of city
economies as measured by GDP: the
other being average income per capita.
Table 3.1 – Trends in top 30 urban agglomerations by population: 1950-
Source: Population Division of the Department of Economic and Social Affairs of theUnited Nations Secretariat, World Urbanization Prospects: The 2007 Revision.
Pop. (m) 1950
- New York
- Tokyo
- London
- Shanghai
- Paris
- Moscow
- Buenos Aires
- Chicago
- Kolkata
- Beijing
- Osaka/Kobe
- Los Angeles
- Berlin
- Philadelphia
- Rio de Janeiro
- St Petersburg
- Mexico City
- Mumbai
- Detroit
- Boston
- Cairo
- Manchester
- Tianjin
- Sao Paulo
- Birmingham
- Shenyang
- Rome
- Milan
- San Francisco
- Barcelona
Pop. (m) 1990
Tokyo New York Mexico City Sao Paulo Mumbai Osaka-Kobe Kolkata Los Angeles Seoul Buenos Aires Rio de Janeiro Paris Cairo Moscow Delhi Shanghai Manila London Jakarta Chicago Beijing Karachi Istanbul Dhaka Tehran Bangkok Lima Tianjin Hong Kong Chennai
Pop. (m) 2007
Tokyo New York Mexico City Mumbai São Paulo Delhi Shanghai Kolkata Dhaka Buenos Aires Los Angeles Karachi Cairo Rio de Janeiro Osaka-Kobe Beijing Manila Moscow Istanbul Paris Seoul Lagos Jakarta Chicago Guangzhou London Lima Tehran Kinshasa Bogotá
Pro- jected Pop. (m) 2025
Tokyo Mumbai Delhi Dhaka São Paulo Mexico City New York Kolkata Shanghai Karachi Kinshasa Lagos Cairo Manila Beijing Buenos Aires Los Angeles Rio de Janeiro Jakarta Istanbul Guangzhou Osaka-Kobe Moscow Lahore Shenzhen Chennai Paris Chicago Tehran Seoul
Ranking in 1950
Ranking in 1990
Ranking in 2007
Projected Ranking in 2025
Table 3.2 – Data sources for city GDP estimates and projections Variable Sources for 2008 estimates Sources for 2025 projections Urban area population
Source: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Urbanization Prospects: The 2007 Revision; extrapolated from 2007 to 2008 using 2005 – 2010 average annual growth rate
UN projections to 2025
GDP per capita for OECD urban areas
OECD Competitive Cities report (2006) estimates for 2002, extrapolated forward to 2008 using OECD data for 1995-2002 and IMF for 2005-2008, plus data on the city-national differential where available from individual national statistical offices
National projections for GDP per capita growth from PwC World in 2050 model to 2025, with adjustments to reflect historic differentials between city and national growth where OECD data available (for 44 countries in 1995-2002 period). Further adjustments made to short term growth rates due to recent global economic downturn GDP per capita for non-OECD countries
Direct estimates from national statistical offices where available (e.g. China, Brazil) or adjusted World Bank national data to reflect typical ratios of GDP per capita in major cities relative to national averages based on comparators with similar characteristics (e.g. cities of similar population in countries with similar income levels). Asian Development Bank data used for some Asian cities
National projections for GDP per capita growth from PwC World in 2050 model to 2025 for countries where available, with other countries being based on closest available comparators, with some judgemental adjustments to reflect particular national characteristics where appropriate. City GDP per capita growth assumed to be in line with national average for non-OECD countries due to lack of city-level data. Further adjustments made to short term growth rates due to recent global economic downturn
(^3) Although, as shown in Annex B, Manchester and Birmingham still rank in the top 100 cities by GDP in 2008.
PricewaterhouseCoopers UK Economic Outlook November 2009 • 23
Mexico City and Sao Paulo are the only
emerging economy cities in the top
10 when ranked by GDP but Buenos
Aires is not far behind in 13th place and
Moscow in 15th. Shanghai and Mumbai
have jumped into the top 30 with their
strong growth between 2005 and 2008.
Moscow has significantly jumped from
25th to 15th since 2005 due to new data
being available from the World Bank
on the percentage share of Moscow in
total Russian GDP. The full list in Annex
B shows that there are also a number
estimated GDP significantly higher
than national economies such as
South Africa, Belgium, Sweden and
Switzerland as illustrated in Table 3.4).
The most significant changes in top 10
rankings since 2005 have been London
climbing ahead of Paris to 5th place and
Sao Paulo jumping into 10th place^9. The
latter is due to new data being found on
income per capita for all Brazilian cities.
Aside from London and Paris, only
two other European cities (Madrid and
Moscow) make the top 30.
order of magnitude terms and, as noted
above, taking account of income per capita
certainly produces a much better indication
of the relative size of urban economies than
just looking at population data.
Subject to these caveats, Table 3.
shows our estimates of the size of the
top 30 urban agglomerations (on UN
definitions) in 2008, ranked by GDP at
Purchasing Power Parity (PPP) exchange
rates using the methodology described
above and in Annex A. The reason for
using PPPs rather than market exchange
rates is to correct for the currently
significant differences in price levels
between emerging market and advanced
economies, reflecting the relatively low
cost of non-traded goods and services
in emerging economies (this is expected
to be less of an issue by 2025).
It is interesting to note that, in total, our
estimates suggest that the largest 100
cities accounted for around 30% of global
GDP at PPPs in 2008, with the top 30
cities alone accounting for around 18% of
world GDP in that year. This emphasises
the concentration of global economic
activity in the world’s largest cities.
The most striking point to note is that,
while 22 of the top 30 urban areas by
population in 2008 were from emerging/
developing economies (see Table 3.
above), only 7 of these emerging economy
cities (Mexico City, Sao Paulo, Buenos
Aires, Moscow, Shanghai, Mumbai
and Rio de Janeiro) were in the top 30
according to our GDP estimates^8. This
reflects the much higher GDP per capita
levels in the major developed economy
cities than in the major emerging market
cities, as illustrated for a selection of
cases in Figure 3.3. Indeed, based on
OECD and IMF estimates, 23 of the top 30
cities ranked by GDP per capita at PPPs
in 2008 were from the US.
Looking at the top of the 2008 GDP
rankings in Table 3.3, Tokyo has
retained the top ranking we found for
2005 and is narrowly ahead of New
York, with both having economies worth
nearly $1.5 trillion in 2008 (broadly
similar to national economies such
as Spain and Mexico). Los Angeles is
still in clear third place with Chicago,
London and Paris vying for the next
three places (each of which has an
New YorkLondonChicago LA Paris Tokyo Osaka MoscowBuenosAires MexicoCity PauloSao Istanbul JaneiroRio deShanghaiMumbai
Figure 3.3 – Estimated GDP per capita in 2008 in selected major cities
Source: PwC estimates based on OECD, World Bank and national data sources
$k at PPPs in 2008
Table 3.4 – Comparison of estimated GDP of largest urban agglomerations with GDP of selected national economies in 2008
using UN definitions (as in Table 3.3 above). These estimates are from different sources and so^ Source: World Bank for national GDP estimates; PwC for urban agglomeration GDP estimates will not be fully consistent, but should be broadly comparable in order of magnitude terms
Country/Urban Agglomeration Estimated GDP in 2008 ($bn at PPPs) Russia UK Mexico Tokyo Spain New York Canada Los Angeles Australia Poland Chicago London Paris South Africa Osaka/Kobe Colombia Mexico City Philadelphia Sao Paulo Belgium Sweden Switzerland
(^89) This is despite using PPP rather than market exchange rates in order to avoid underestimating the scale of the outputs of the emerging economy cities. New data for GDP per capita of Brazilian cities was taken from the following source: http://www.ibge.gov.br/home/estatistica/economia/pibmunicipios/2006/tab01.pdf
24 • PricewaterhouseCoopers UK Economic Outlook November 2009
of fast-growing emerging economy
cities just outside the top 30, including
Istanbul (34th), Beijing (38th), Manila
(40th), Cairo (42nd), Guangzhou (44th),
and Santiago (53rd).
Table 3.5 shows the top 5 cities in 2008 in
their respective country income brackets,
using the World Bank’s classification^10
of countries by income. We can see that
Mexico City tops the list of cities in the
group of Upper-Middle Income Countries
followed closely by Sao Paulo. Shanghai
tops the list of cities in Lower-Middle
Income Countries followed by Mumbai.
Dhaka tops the Low Income Country
category followed by Ho Chi Min City.
In the next section, we consider how far
these and other emerging economy cities
might rise up the rankings by 2025.
III.4 Projected urban economy growth rates and GDP rankings in 2025
Rankings by economic size in 2025
Table 3.6 shows our projections of the top
30 urban economies in 2025 measured
by GDP at PPPs (in 2008 US dollars),
with the rankings in 2008 shown in the
first column for comparison. The full GDP
rankings for both years are given in Table
3.10 in Annex B. These are based on UN
definitions and population projections.
The largest five urban economies (on UN
definitions) remain the same as in 2008,
although London overtakes Chicago
to move into 4th place. As might be
expected, however, the dominant trend
is for emerging economy cities to rise
up the rankings: Sao Paulo climbs from
10th to 6th, Mexico City rises from 8th to
7th and Buenos Aires from 13th to 10th.
The largest climbers within the top 30 are
Shanghai (leaping into the top 10 from 25th
to 9th) and Mumbai (racing from 29th to
11th). Istanbul (34th to 28th), Beijing (38th
to 17th), Delhi (37th to 19th), Guangzhou
(44th to 21st) and Cairo (42nd to 30th) are
all notable ‘new entries’ in the top 30.
Lower down the list (see Annex), notable
‘climbers’ between 2008 and 2025 include
Manila (40th to 33rd), Kolkata (61st to
37th), Bangkok (54th to 44th), Jakarta
Table 3.5 – Top 5 urban agglomerations in different country income brackets by estimated GDP in 2008 (using UN population estimates and definitions)
Source: PricewaterhouseCoopers GDP estimates drawing on data from UN, World Bank, OECD and national sources
City GDP in 2008 ($bn at PPPs) Mexico City Sao Paulo Buenos Aires Moscow Rio de Janeiro
Rank
Top 5 Cities in Upper-Middle Income Countries 1 2 3 4 5
City GDP in 2008 ($bn at PPPs) Shanghai Mumbai Delhi Beijing Manila
Rank
Top 5 Cities in Lower-Middle Income Countries 1 2 3 4 5
City GDP in 2008 ($bn at PPPs) Dhaka Ho Chi Min City Hanoi Yangon Chittagong
Rank
Top 5 Cities in Low Income Countries 1 2 3 4 5
Table 3.6 – Top 30 urban agglomerations by estimated GDP in 2025 using UN population definitions and projections
Source: PricewaterhouseCoopers projections
Average real GDP growth (% pa: 2008-2025)
Population in 2025 (millions)
Estimated GDP in 2025 ($bn at 2008 PPPs)
2025 GDP City rank ( in brackets) 1 (1) 2 (2) 3 (3) 4 (5) 5 (4) 6 (10) 7 (8) 8 (6) 9 (25) 10 (13) 11 (29) 12 (15) 13 (9) 14 (16) 15 (11) 16 (7) 17 (38) 18 (12) 19 (37) 20 (14) 21 (44) 22 (21) 23 (17) 24 (30) 25 (18) 26 (19) 27 (20) 28 (34) 29 (22) 30 (42)
Tokyo New York Los Angeles London Chicago Sao Paulo Mexico City Paris Shanghai Buenos Aires Mumbai (Bombay) Moscow Philadelphia Hong Kong Washington DC Osaka/Kobe Beijing Boston Delhi Dallas/Fort Worth Guangzhou Seoul Atlanta Rio de Janeiro San Francisco/Oakland Houston Miami Istanbul Toronto Cairo
Rising by more than 3 places
(^10) For the purpose of this calculation, cities are classified as emerging market cities by the countries in which they are based and how the countries are classified by the World Bank according to income levels (see http:// go.worldbank.org/D7SN0B8YU0 for further information)
26 • PricewaterhouseCoopers UK Economic Outlook November 2009
than larger Asian cities like Bangkok,
Manila and Jakarta. Similarly, Dubai has
been more successful than Cairo. The
same author notes that Mexico City, the
largest emerging economy city in the
It should also be noted that economic
size, although significant, is not a
panacea. As noted by Joel Kotkin^13 ,
Singapore has established itself as a
global financial centre to a greater extent
III.5 Key uncertainties and factors underlying relative city growth rates
It should be recognised, however, that
even though we believe that our general
conclusion on the rise of the emerging
market economies and cities should be
robust, any such growth rankings can only
be illustrative for individual cities. Given
the objective of providing a comprehensive
global ranking, our analysis is necessarily
somewhat mechanical and relies both
on the UN population projections, which
are subject to widening margins of error
over time as with any such long-term
projections^11 , and on the assumption
that our earlier work on national GDP per
capita projections provides a good basis
for city-level projections. We also need
to acknowledge that economic size as
measured by GDP may not fully reflect the
level of well-being in a city given it ignores
other relevant socio-economic factors
(including income inequality, the value of
home production, the quality and quantity
of leisure time and environmental indicators
such as air and water pollution).^12
In practice, some cities may do
significantly better that their national
economies and some may lag behind.
Equally, not all of the emerging economies
may fulfil the potential identified in our
World in 2050 report, whether due to
political and/or macroeconomic instability,
infrastructure constraints, energy supply
problems or environmental crises.
Avoiding these pitfalls, both at national
and local level, will be critical to the long-
term economic success of these cities.
Climate change, for example, may pose
significant long-term challenge, particularly
to coastal cities in developing economies
that may be relatively exposed to more
frequent and intense severe weather events
such as typhoons and hurricanes, as well
as a potential long-term rise in sea levels
due to global warming. Some preliminary
analysis we have carried out suggests that
emerging market coastal cities like Dhaka,
Cairo, Karachi, Lagos, Tianjin and Porto
Alegre could be particularly exposed to
any early adverse effects from climate
change. There are huge uncertainties here
as to the scale and timing of any such
effects and the extent to which mitigating
action could be taking in time to avoid
serious consequences.
Table 3.8 – Top 30 urban agglomerations by projected average real GDP growth in 2008-25 (using UN population definitions and projections)
Source: PricewaterhouseCoopers projections using UN population definitions
City
Hanoi Ho Chi Min City Changchun Guangzhou Addis Ababa Xian Surat Beijing Jaipur Lucknow Chengdu Shenyang Kanpur Shanghai Tianjin Pune Chongqing Ahmedabad Kabul Bangalore Hyderabad Dar es Salaam Chennai (Madras) Delhi Lagos Nairobi Kolkata (Calcutta) Mumbai (Bombay) Chittagong Kinshasha
Country
Vietnam Vietnam China China Ethiopia China India China India India China China India China China India China India Afghanistan India India Tanzania India India Nigeria Kenya India India Bangladesh Democratic Republic of Congo
Average real GDP growth in 2008- (% per annum) 7.0% 7.0% 6.9% 6.8% 6.8% 6.7% 6.7% 6.7% 6.7% 6.6% 6.6% 6.6% 6.6% 6.6% 6.6% 6.6% 6.6% 6.5% 6.5% 6.5% 6.5% 6.5% 6.5% 6.4% 6.4% 6.4% 6.4% 6.3% 6.3% 6.3%
Growth rank
1 2 3 4 5 6 7 8 9
Osaka Paris
LA Philadelphia
Tokyo New York
Chicago London
Moscow BuenosAires
MexicoCity PauloSao
Istanbul JaneiroRio de
Mumbai Shanghai
Figure 3.4 – Cumulative projected GDP growth to 2025 for mega-cities
Top 8 emerging economy and Top 8 advanced economy cities Source: PwC analysis
% cumulative real GDP growth: 2008-
(^11) In addition to the earlier research by Bairoch (1982) cited above, this point is also explored in some detail in a more recent paper by Barry Cohen (‘Urban Growth in Developing Countries: A Review of Current Trends and 12 a Caution Regarding Existing Forecasts’, World Development, vol 32, no 1, pp 23-51, 2004).See http://www.stiglitz-sen-fitoussi.fr/documents/rapport_anglais.pdf (^13) J. Kotkin, The City: A Global History (Pheonix: London, 2005). Similar arguments on the potential disbenefits of greater city size beyond some threshold were set out by Bairoch (1982, op. cit).
PricewaterhouseCoopers UK Economic Outlook November 2009 • 27
In this scenario an assumed rise in global
protectionism, notably by the US and
the EU, leads to generally slower world
economic growth with particularly severe
adverse effects on emerging economies,
which typically rely heavily on cross-
border trade and investment flows to
realise their growth potential (as shown
by the experience of the Asian tigers
III.6 Possible impact of de-globalisation scenario on city growth rates
To explore further the uncertainties
surrounding our projections, we have
considered one particular alternative
scenario which could impact the growth of
some cities significantly: de-globalisation.
world based on our analysis, is burdened
by problems of crime, congestion
and pollution that make smaller but
faster-growing cities like Monterrey
and Guadalajara more attractive to
entrepreneurs and ambitious workers.
Within the developed world, it seems clear
that the most successful cities will be
those that have comparative advantages
in intangible business, financial and
consumer services that are not so easily
emulated by the rising stars of China, India
or Brazil. Prominent examples include the
continued pre-eminence of London, New
York and Tokyo in global financial services,
or of Los Angeles in the media and
entertainment sector, but it also applies to
smaller but possibly faster growing cities
that specialise in new technologies where
distance is not an issue and the most
talented individuals are looking for a better
quality of life than the mega-cities can offer.
The comparatively rapid projected growth
rates (by developed country standards) of
cities such as Atlanta, Dublin, Stockholm
and Seattle reflect these kinds of more
qualitative factors.
More formally, our projections show
a negative correlation between initial
economic size (GDP) and subsequent
projected growth, but this is very much
driven by lower initial GDP per capita
in emerging economies (see Figure 3.5,
which shows a significant but non-
linear relationship). After correcting for
differences in initial GDP per capita,
regression analysis does not indicate
any statistically significant relationship
between initial population levels and
subsequent projected GDP growth^14.
These are only projections, of course,
so this is a feature of our analysis that
may or may not be borne out by actual
experience. Without time series of historic
GDP for a sufficient range of cities we
are unfortunately not able to test these
relationships using actual data.
It is also important to note that, while
cities may compete for inward investment
in some respects, they are also important
trading partners for each other to the
extent that they specialise in different
areas of economic activity. A larger global
market can still be of great potential
benefit to those ‘old Europe’ cities even
if the latter are likely to slide down the
relative GDP rankings.
Figure 3.5 – Projected urban area GDP growth vs initial GDP per capita
GDP per capita in 2008 ($k PPPs)
Average annual GDP growth (% pa: 2008-25)
Source: PwC analysis
Table 3.9 – Impact of de-globalisation scenario on projected GDP rankings of selected cities in 2025 (using UN definitions and population estimates)
*Relative to baseline projections for 2025 Source: Illustrative PwC projection
Percentage change in projected 2025 GDP
Change in rank*
De-globalisation scenario results New 2025 rank
City
Shanghai Mumbai (Bombay) Rio de Janeiro Manila Cairo Guangzhou Bangkok Jakarta Dhaka Karachi Tianjin Chennai (Madras) Porto Alegre Ho Chi Min City Ahmedabad Alexandria Hanoi Lahore Surat Lagos Chengdu Casablanca Changchun Chittagong Lucknow
(^14) In fact, cities with larger populations have, after allowing for differences in initial GDP per capita levels, slightly higher projected growth rates in our model, but not to a statistically significant degree (t-statistic = 1.1).
PricewaterhouseCoopers UK Economic Outlook November 2009 • 29
Annex A: Data and methodology used to derive city GDP
estimates and projections
and services produced in each urban
agglomeration more accurately. Using
current market exchange rates instead
would tend to understate the scale of the
outputs of goods and services produced
by emerging economy cities, particularly
estimates of City GDP in 2008. By
2025 we can expect the increase in
productivity in emerging markets to
raise both GDP per capita and the real
exchange rate, thus closing at least
partially the gap between GDP estimates
based on PPP exchange rates and
market exchange rates.
As outlined in Section III.2 above, our
primary estimates of city output are
based on combining UN population
estimates for 2008 with estimates of
income per capita, The population
estimates for 2008 are interpolated by
using the 2007 UN population estimates
and applying the average annual
population growth rate between 2007
and 2010. For the OECD countries,
we began with the city-level GDP per
capita estimates for 2002 in the OECD’s
Competitive Cities report (2006) and then
projected these forward to 2008 based
on national GDP per capita growth over
this period plus an adjustment to reflect
the observed city-national GDP per
capita growth differential in 1995-
for OECD cities for which these historic
data were available (in other cases,
unadjusted national growth data were
used). The 2005 estimates were further
extrapolated to 2008 using national
income per capita growth rates from the
IMF World Economic Outlook database.
For non-OECD cities, data are not readily
available from a single source. In some
cases (e.g. China, Brazil) GDP per capita
estimates at city level were available
from national sources, but in many cases
we were only able to make approximate
estimates based on plausible ratios of
city to national GDP per capita (the latter
sourced from the World Bank) based
on comparisons with cities at similar
income levels for which direct income
per capita estimates were available^17.
As such, the 2008 urban agglomeration
GDP estimates should only be taken as
it appears that the UN adopts narrower
definitions than Brinkhoff, which tends
to make the UN estimates correspond
more closely to what might generally be
considered to be a city, as opposed to a
cluster of closely-related cities or towns.
But there is no ‘right’ answer here, so it
is important to recognise that our GDP
rankings are sensitive to the particular
definitions used.
To establish our list of candidate urban
agglomerations to be ranked in the
global top 100 by economic size in either
2008 or 2025, we first included all urban
agglomerations (using UN definitions)
with a population of more than 3 million
in 2008 (115 areas in total). We then
added:
- Other urban agglomerations projected
to be in the top 100 by population in
2025 using UN projections; and
- Other OECD urban agglomerations
with populations over 1 million, as
covered by the OECD report on
Competitive Cities (2006).
This procedure gave a total of 151
candidate urban agglomerations for
further analysis. Based on a review of
our results, we are confident that this
should cover all urban agglomerations
(on UN definitions) likely to rank in the top
100 by GDP in 2008, and probably also
in 2025 (although the latter is obviously
subject to more uncertainty). Table 3.10 in
Annex B shows results for all of the cities,
although it should be noted that we
cannot be sure that these are the largest
151 city economies given that our aim
was just to identify the top 100.
We chose to use GDP at Purchasing
Power Parity (PPP) exchange rates
as our measure of economic size.
The reason for using PPPs rather
than market exchange rates is to
correct for differences in price levels
between economies, which are due
in particular to the relatively low cost
of non-traded goods and services in
emerging economies. By using PPPs,
we can compare the volume of goods
We describe below how we have gone
about producing GDP estimates and
projections for the leading cities in the
world.
The first question to be addressed in
any study of this kind is: how should you
define a city? While national boundaries
are clear and change relatively rarely,
city definitions differ significantly across
countries and evolve over time as the
city expands and absorbs surrounding
neighbourhoods. For the purposes of
this study we have generally adopted UN
definitions^15 of ‘urban agglomerations’
(for short, these are sometimes also
referred to below as ‘urban economies’
or just as ‘cities’ where the context
makes this appropriate), but it should
be recognised that the UN population
estimates rely on information provided
by national statistical agencies and are
therefore not based on fully standardised
definitions across countries.
To explore the effect of adopting
alternative definitions, we also
considered the impact on our 2008
GDP estimates of using an alternative
set of urban agglomeration population
estimates compiled by Professor
Thomas Brinkhoff (see his website
at www.citypopulation.de for details)
that also provide global coverage
and have been used in a number of
previous studies^16. However, 27 of the
top 30 cities were the same using the
Brinkhoff data as in our analysis using
the UN urban population data (from
its 2007 World Urbanization Prospects
report). Given these broadly comparable
results, the UN data were selected
as our primary source because they
have the advantage of providing both
a time series of historic data by city/
urban area back to 1950 and population
projections to 2025 for individual cities/
urban areas. We also used UN national
population projections in deriving our
national GDP per capita projections, so
it was more consistent to use UN data
here than the Brinkhoff estimates, which
include some historic estimates back
to 1970 but not forward projections. In
the majority of cases where they differ,
(^15) The UN defines an urban agglomeration’s population as follows: ‘The de facto population contained within the contours of a contiguous territory inhabited at urban density levels without regard to administrative boundaries. It 16 usually incorporates the population in a city or town plus that in the sub-urban areas lying outside of but being adjacent to the city boundaries’ (http://esa.un.org/unup/index.asp?panel=6)The other alternative we considered was to use the OECD definition of metropolitan areas from their recent report on ‘Competitive Cities’ (2006). But, unlike the UN data and the Brinkhoff estimates, this would not have 17 covered non-OECD countries and also did not provide historic and projected population estimates on a consistent basis.Typically, these ratios are in the range from 1.5 to 3, with higher values tending to be observed in the lowest income countries where urban-rural income differentials are particularly large.
30 • PricewaterhouseCoopers UK Economic Outlook November 2009
were available, we assumed that
differences between national and
urban GDP per capita growth rates in
2008-25 were half those in 1995-2002.
This was based on the assumption
that historic growth differentials would
be gradually eroded over time, since
otherwise there would be implausibly
rapid or slow growth of the major
cities relative to their economies
as a whole. For all the other urban
agglomerations, including non-OECD
cities, we assumed (in the absence of
other data) that their income per capita
growth would be in line with national
average projections. This is, in fact,
in line with the average historic trend
for the OECD cities for which data
are available^18. In practice, of course,
income per capita growth rates will
vary more than this at city level, but
we have no readily available data on
which to predict such variations.
the income per capita growth rates (this
has a particularly large downward effect
in 2009-10). As illustrated for selected
countries in Figure 3.2 above, these
projections show consistently higher
income per capita growth in the emerging
economies, particularly China and India.
It is notable here that US GDP per capita
growth is projected to be slower than that
in the other major economies. This is due
to the assumption in our model that other
countries will tend to catch up gradually
with initially higher economy-wide labour
productivity levels in the US. It should
be noted, however, that after taking
account of its higher projected population
growth (including immigration), overall
US GDP growth is nonetheless projected
to be higher than in any of the other G
countries.
For the OECD urban agglomerations
where historic income growth trends
broadly indicative of relative economic
size for the non-OECD countries.
Nonetheless, they provide a much better
indication of relative economic size than
just looking at population data. For most
of the non-OECD cities, we extrapolated
the 2005 GDP per capita estimates to
2008 using national income per capita
growth rates from IMF. However, for
some cities such as Bogota, Moscow,
Mumbai and Sao Paulo, new data were
found on income per capita and used to
calculate 2008 GDP city estimates.
As Table 3.2 in the main text shows, our
illustrative projections for city GDP in 2025
combine UN population projections with
our own estimates of national income per
capita growth trends from our previous
World in 2050 report (see earlier footnote
6 for reference). We then incorporated
the short term and long term impacts of
the recent global economic downturn on
(^18) It should be noted here that, particularly for smaller economies, the largest cities may play a dominant role in their overall national economies, so one would not expect a large divergence between income growth in these cities and the average for their economies as a whole.
32 • PricewaterhouseCoopers UK Economic Outlook November 2009
Table 3.10 – Full listing of urban agglomeration GDP rankings in 2005 and illustrative projection to 2025 (using UN definitions and population estimates)
Barcelona Melbourne Delhi Beijing Denver Metro Manila Montreal Cairo Rome Guangzhou Baltimore Milan Tehran St Louis Tampa/St Petersburg Vienna Tel Aviv-Jaffa Busan Santiago Bangkok Cleveland Brasilia Portland Johannesburg Lima Riyadh Kolkata (Calcutta) Cape Town Monterrey Bogota Pittsburgh Lisbon Athens Vancouver Berlin Jakarta St Petersburg Birmingham Fukuoka Manchester Brussels Guadalajara Dhaka Karachi Hamburg
Rank
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
Cities ranked by estimated 2008 GDP at PPPs 177 172 167 166 165 149 148 145 144 143 137 136 127 126 123 122 122 121 120 119 112 110 110 110 109 107 104 103 102 100 99 98 96 95 95 92 91 90 88 85 83 81 78 78 74
Est. GDP in 2008 ($bn at PPPs)
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
Rank
Singapore Sydney Kolkata (Calcutta) Phoenix Minneapolis San Diego Tehran Barcelona Melbourne Bangkok Jakarta Denver Tianjin Dhaka Riyadh Lima Brasilia Santiago Rome Montreal Bangalore Johannesburg Karachi Bogota Tel Aviv-Jaffa Chennai (Madras) Monterrey Baltimore Cape Town Ho Chi Min City Milan Busan Vienna St Louis Hyderabad Chongqing Tampa/St Petersburg Cleveland Portland Guadalajara St Petersburg Lisbon Ahmedabad Jiddah Athens
Cities ranked by projected 2025 GDP at PPPs 312 298 298 271 265 260 252 248 245 241 231 226 218 215 214 213 210 207 203 203 203 198 193 192 191 191 188 187 183 181 178 177 175 172 170 170 168 153 152 150 149 149 145 143 142
Est. GDP in 2025 ($bn at 2005 PPPs)
2.2% 2.0% 6.4% 1.8% 1.8% 1.8% 4.1% 2.0% 2.1% 4.2% 5.5% 1.9% 6.6% 6.2% 4.2% 4.0% 3.9% 3.3% 2.1% 1.9% 6.5% 3.5% 5.5% 3.9% 2.7% 6.5% 3.7% 1.8% 3.5% 7.0% 1.6% 2.2% 2.1% 1.8% 6.5% 6.6% 1.8% 1.9% 1.9% 3.6% 3.0% 2.5% 6.5% 4.1% 2.4%
Real GDP growth rate (% pa: 2008-25)
95 103 27 120 119 122 55 102 98 49 37 113 15 33 53 57 63 78 100 110 20 73 38 60 83 23 66 116 75 2 139 93 96 117 21 17 115 114 109 68 80 87 18 54 90
GDP growth ranking (out of 151)
PricewaterhouseCoopers UK Economic Outlook November 2009 • 33
Table 3.10 – Full listing of urban agglomeration GDP rankings in 2005 and illustrative projection to 2025 (using UN definitions and population estimates)
Tianjin Jiddah Stockholm Lyon Bangalore Warsaw Turin Chennai (Madras) Porto Alegre Munich Belo Horizonte Dublin Leeds Hyderabad Ankara Ho Chi Min City Helsinki Chongqing Auckland East Rand Budapest Zurich Wuhan Naples Medellin Ahmedabad Prague Copenhagen Pune Amsterdam Rotterdam Alexandria Algiers Curitiba Shenyang Daegu Hanoi Izmir Puebla Caracas Oslo Lahore Cologne-Bonn Surat Lagos
Rank
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
Cities ranked by estimated 2008 GDP at PPPs 74 72 70 69 69 68 68 66 66 64 61 61 60 58 58 58 58 57 55 54 53 52 52 51 50 49 49 49 48 47 46 46 45 44 44 43 42 42 42 41 40 40 39 36 35
Est. GDP in 2008 ($bn at PPPs)
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
Rank
Pune Pittsburgh Hanoi Vancouver Shenyang Porto Alegre Berlin Fukuoka Birmingham Belo Horizonte Ankara Brussels Manchester Alexandria Warsaw Surat Dublin Lahore Wuhan Lagos Chengdu East Rand Lyon Medellin Algiers Stockholm Xian Luanda Hamburg Turin Khartoum Auckland Curitiba Changchun Izmir Munich Budapest Helsinki Puebla Leeds Kanpur Prague Zurich Caracas Jaipur
Cities ranked by projected 2025 GDP at PPPs 142 136 134 133 132 118 117 117 114 112 111 109 108 108 107 107 106 102 102 101 100 98 97 97 96 95 93 93 93 89 86 84 83 81 81 81 80 79 78 78 76 75 73 72 71
Est. GDP in 2025 ($bn at 2005 PPPs)
6.6% 1.9% 7.0% 2.0% 6.6% 3.5% 1.3% 1.7% 1.4% 3.6% 3.9% 1.6% 1.4% 5.2% 2.7% 6.7% 3.3% 5.6% 4.1% 6.4% 6.6% 3.6% 2.0% 4.0% 4.6% 1.9% 6.7% 6.3% 1.3% 1.6% 5.5% 2.5% 3.8% 6.9% 4.0% 1.4% 2.4% 1.9% 3.8% 1.5% 6.6% 2.5% 1.9% 3.4% 6.7%
Real GDP growth rate (% pa: 2008-25)
16 112 1 105 12 72 150 134 147 69 61 137 146 40 82 7 77 36 56 25 11 71 104 58 45 111 6 31 149 140 39 85 64 3 59 148 88 108 65 144 13 86 107 76 9
GDP growth ranking (out of 151)
