This document discusses several paradoxes and puzzles regarding common assumptions about cities, regions, and productivity. It summarizes recent evidence that challenges some of these assumptions, such as: not all cities display a productivity premium; larger cities are not always more productive or resilient; the link between productivity growth and wage growth is weakening; regional re-allocations and innovation diffusion are not universally driving growth as assumed; and urban economic models based on land prices and population growth do not always apply given population declines in some cities. The document examines how the experiences of European cities both support and contradict some textbook frameworks.

1. Cities, Regions and Productivity: Some
Paradoxes and Puzzles
Philip McCann
University of Sheffield
1

2. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Textbook assumptions and underlying frameworks:
• Cities display a productivity premium and drive economic
growth
• Larger cities are more productive and resilient
• Productivity growth drives wage growth
• Regional re-allocations can drive productivity
• Innovation and entrepreneurship diffusion is critical for
growth
• Bid-rent curves (productivity spikes) are downward-sloping
• Transport infrastructure drives productivity growth
• Urban economic models based on land price and/or
population growth
• Henry George Theorem 2

3. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
3
Labour productivity in PPS in metro regions compared to the rest of their country, 2008
ES DE
UK
IT
NL FR
BEATSE
FI
BG
RO
LT
LV
PL
HU
EE
SK
CZ
MT DK
GRSI
PT
0
40
80
120
160
200
240
280
LabourproductivityinPPS,non-metroregionscombined=100
Capital metro region
Second tier metro region
Smaller metro region
Non-metro regions combined
IE

4. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
4
Change in labour productivity in pps, 2000-2008
MT
DK
DE
IE
BEFRAT
SE
IT
FI
UK
NL
ES
BG
RO
LV
LT
PL
EE
HU
CZ
PT
SK
SI
-40
-30
-20
-10
0
10
20
30
ChangeinProductivityrelativetothenationallevelinindexpoints
Capital metro region
Second tier metro region
Smaller metro region
Non metro regions combined
60

5. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
5
Population change in metro regions, 2000-2008
IE
UKNLFR
SK
DK
PLIT
RO
BE
PT
CZ
EE
HU
ES
SILT
DE
BG
SE
AT
MT
LV
GR
FI
-12
-8
-4
0
4
8
12
16
Changeinshareofnationalpopulationin%
Capital metro region
Second tier metro region
Smaller metro region
Non-metro regions combined

6. 6
OECD Extended Regional Typology of European Regions: Two Year Moving Average
Growth Rates in GDP among TL3 regions, 1995-2011
-3%
-2%
-1%
0%
1%
2%
3%
4%
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
PU IN PRC PRR

7. 7
OECD Extended Regional Typology of European Regions: Two Year Moving Average
Growth Rates in GDP per Capita among OECD TL3 regions, 1995-2011

8. 8
Effects of the crisis in OECD TL3 European Regions (Extended OECD Regional
Typology) by Period

9. 9
GDP per head growth in EU metro regions 2000-2008 and 2008-2011
2000-2008 2008-2011
Average
annual real
change in %
GDP per
head growth
=
Productivity
growth +
Employment
per head
growth
GDP per
head growth
=
Productivity
growth +
Employment
per head
growth
EU-15
Capital metro 1.44 0.88 0.56 -0.79 0.34 -1.13
Second-tier
metro 1.29 0.70 0.59 -0.76 0.15 -0.91
Smaller metro 1.20 0.67 0.53 -0.59 0.24 -0.83
Non-metro 1.15 0.75 0.40 -0.77 0.20 -0.98
Total 1.27 0.76 0.51 -0.70 0.24 -0.94
EU-13
Capital metro 5.49 3.64 1.85 -0.26 1.04 -1.30
Second-tier
metro 4.85 4.08 0.78 1.43 1.30 0.14
Smaller metro 3.66 3.56 0.09 1.38 1.17 0.21
Non-metro 4.47 4.45 0.02 0.57 1.70 -1.13
Total 4.88 4.31 0.56 0.66 1.44 -0.78
Source: Eurostat and authors’ calculations

10. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Assumptions: Cities display a productivity premium and drive
economic growth; → Recent evidence: While this is true for
many cities it is also true that many cities display no (or even a
negative) productivity premium; many cities are not
(productivity or population) growth drivers; in some countries
productivity-scale relationships are zero or close to zero
• EU experience is very different to US - Triumph of the City
• In Europe urban advantages relating to employment and
productivity post-crisis are oriented towards EU13 economies
while EU15 face severe urban disadvantages
10

11. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Assumption: Larger cities are more productive and resilient →
EU cities display both resilience and vulnerability: they
exacerbate national post-crisis trends with growing countries
driven by growing cities and declining countries weighed
down by declining cities
• Effects of the real estate-related debt on the real economy are
more pronounced in cities; real estate shock effects are
dominated by cities → induced effects in the real economy
• Assumption: Productivity growth drives wage growth →
Recent evidence: the link is rapidly weakening due to job
polarisation, especially in cities
11

12. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Assumption: Regional re-allocations can drive productivity →
Recent evidence: increasing divergence in many countries;
falling US interregional migration; static UK interregional
migration (2 of the 3 most geographically mobile societies);
increasing spatial dispersion of yields on investments
• Assumption: Innovation and entrepreneurship diffusion is
critical for growth → Recent evidence: falling US rates of
entrepreneurship; innovation diffusion processes
stalled/broken in many countries UK/US; costs of innovations
increasing?
•
12

13. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Assumption: Bid-rent curves (productivity spikes) are
downward-sloping → Recent evidence: upward sloping bid
rent curves in European many cities; flat bid-rent curves in
many developing country mega-cities
• Assumption: Transport infrastructure drives productivity
growth → Recent evidence: no specific observable link which
can be generalised: only general broad statements are
possible based on comparisons between regions with and
without good infrastructure
13

14. Cities, Regions and Productivity: Paradoxes,
Puzzles and Policy
• Assumption: Urban economic models based on land price
and/or population growth → Recent evidence: between one
quarter and one third of European cities are facing population
decline; a majority of Japanese and Korean cities are declining
• Little or no knowledge regarding how to ‘manage decline’ →
random vacancies/insolvencies increase marginal costs of
infrastructure and network-based service provision and
inhibit/prohibit necessary land use consolidation and
coordination
• Assumption: Henry George Theorem → Recent evidence:
human capital rewards in cities are 50% economic rents
rather than returns to knowledge investments (Collier and
Venables 2018)
14