This paper will explore the question of why city size increases productivity.
At the end of the paper it will be clear that city size results in such
economic stimulants and societal benefits as reduced transaction costs,
shared inputs of production, larger and more specialized labor pools, and
scales of economy. While constraints in productivity that result from urban
growth certainly exist, the benefits of urban agglomeration outweigh the
costs.
At the outset I will examine Alfred Marshall's theories on city size that
have provided the basis for much of subsequent theories and research on
the subject. Following that I will illustrate how city size affects productivity,
compare the effects of agglomeration in cities in high-tech and low-tech
economic sectors, and consider how city size affects demographics, specifically
the age structure of cities. I will then present two case studies where
empirical evidence has been gathered on the effects of city size. Finally,
I will consider the potential constraints of population growth on economic
development.
Section 2: Alfred Marshall
Many of the arguments about how city size effect the level of economic output
that a city will achieve were first put forth by Alfred Marshall in his
essays Principles of Economics and Industry and Trade, written at the end
of the nineteenth century (Quigley 130). The core of his thesis is that
dense populations of people will facilitate the flow of ideas, information
and skills between people (Glaeser 2). Cities increase the speed and ease
with which people are able to interact, and the result of this interaction
is the stimulation of innovative ideas. Cities do not just result in the
generation of new technologies, but also the faster rate of contact between
individuals will create learning opportunities for individuals who can then
gain skills. There will be a conglomeration and higher levels of skilled
people in industrial centers. Cities create human capital.
Recent scholarship has paid especially close attention to his three pronged
theory that illustrates the reasons that urban populations are ideal locations
for industry: 1) access to specialized suppliers (of both intermediate goods
and producer services; 2) pools of workers with specialized skills; 3) knowledge
spillovers (Fesser 2487).
Marshall asserted that supplier industries within densely populated areas
are likely to be more efficient than producers who chose isolated locations.
First, suppliers are able to minimize the inefficient practice of producing
their own inputs when the local market is large and diverse enough to support
specialized contract suppliers who can do that for them at a lower cost.
The fact that inputs can be obtained from a local market allows businesses
greater flexibility because they can more easily obtain inputs at smaller
quantities on an as-needed basis. Especially in a volatile market, a business's
ability to supplement quickly its orders if its level of demand increases
is a considerable advantage. Marshall also pointed out that buyers who work
more directly with their suppliers will be able to work more closely and
achieve the best collaborative agreements because they are able to communicate
and coordinate more easily due to their proximity (Fesser 2487).
Greater access to skilled and experienced labor is the second benefit that
Marshall gives for firms to locate in highly populated areas (Fesser 2487).
He points out that the supply of trained labor is reasonably assured in
areas with larger populations, not only because there are more people, but
also because a region with a greater number of producers in the same area
will motivate a larger number of workers to specialize in that industry's
skill set in response to the demand for employment. As a result, firms are
able to maintain a higher level of efficiency because they can constantly
adjust their production level to meet any changes in demand.
Third, Marshall argued that areas with high population densities and a concentration
of industry would result in a high level of knowledge spillovers. Knowledge
spillover is the concept that techniques and technology will be shared among
agglomerated firms. In cities with high population densities there will
be a larger amount of information and innovation to share as well as an
increased level of contact between firms. Each firm will benefit from what
they gain through their association with other firms and be able to maximize
their respective efficiency.
As I already mentioned, Marshall's arguments about spatial economies still
receive a great deal of attention to this day and serve as the basis for
much of the ensuing research and conclusions about the relationship between
city size and productivity.
Section 2: City Size and Productivity
City size does relate to the productivity levels of the cities in a particular
city. The evidence gathered from numerous empirical studies varies in the
amount that doubling a city's size would increase productivity in that city.
But all the studies make it clear that productivity will increase as the
city size increases. For example, Shefer analyzed a group of 20 industries
within a metropolitan area and concluded that doubling the city size could
increase productivity up to 27 percent (Quigley 134).
There are four main reasons that city size increases the productivity of
the firms within a city: scale of economies, shared inputs in production,
lower transaction costs, and less fluctuations in the economy (Quigley 130-2).
These reasons are rooted in Marshall's original theory.
Scale of economies are the historical rational for cities in the first place
(Quigley 130).
Economies of scale is the concept that an increase in the number of units produced will decrease the average cost of each unit because of increases in production specialization and efficiency. In large cities firms have access to a wider variety of the goods that they need for their inputs. As a result they don't have to spend as much time producing these goods that they cannot produce as efficiently on their own and can focus their resources on improving the quality of the goods in which they specialize and the efficiency with which they are produced.
Shared inputs in production are the second reason the productivity of firms
increase within cities (Quigley 131). Shared inputs contribute to the economies
of scale discussed above. In a city firms will be able to collectively benefit
from the ready availability of workers, and particularly of specialized
workers like lawyers, accountants, advertisers, etc. who can produce many
of the necessary inputs and services more efficiently than firms would be
able to do on their own and can thus reduce the cost of doing business.
The third reason why a metropolitan area facilitates the increase in economic
efficiency is that in a city transaction costs are reduced (Quigley 132).
Cities reduce the transaction costs of production because the costs associated
with searching for the right workers are reduced. Cities support a wide
variety of workers who can meet the differentiated demands of firms. Firms
will be able to spend less resources investing in either finding or training
workers to fill the business's needs and so can invest more into capital
deepening (expansion, research and development, etc.), which results in
increased production.
Finally, large cities will reduce the fluctuations in the market. For example,
fluctuations in purchases of inputs are imperfectly correlated across firms.
As a result employment can be stabilized because some firms are hiring while
other firms are not. Also, since consumption is uncorrelated among buyers,
firms will be able to carry fewer inventories because, even if some consumers
are not buying, other consumers are (Quigley 132). The productivity of the
economy will increase as a result of the diversification that results from
larger cities.
All of these factors contribute to the conclusion that economic productivity
does increase as populations become increasingly dense.
Section 4: High-Tech Vs. Low-Tech Industries
It is important to point out that not all sectors are equally affected by
population agglomeration. A more defensible assessment of how specialized
suppliers, deeper labor pools, and knowledge spillovers positively effect
economic efficiency can be developed when considering the effects that population
agglomeration has on high-technology industries versus low-technology industries
(Fesser 2486).
High-technology industries refer to the range of industries in the sectors
of an economy with very high levels of technological advancement such as
the aerospace industry, search and navigation equipment industries, and
missile and space vehicle producers as well as moderate technological sectors
such as appliance manufacturing (Fesser 2486). City size will result in
a considerable increase in the productivity of high-technology industries.
These industries will benefit greatly from the increased access to producer
services, specialized labor, and knowledge spillovers that result when population
agglomeration occurs (Fesser 2503).
However, lower-technology industries do not reap the same magnitude of benefits
from city size as higher-technology industries. Lower-technology industries
are those that are final markets, like the farm or garden machinery industry,
which is largely producer durable and consists of a range of comparatively
standardized products (Fesser 2486). Lower-technological industries do not
require the diversity and higher level of skill that results from the specialization
of workers in larger cities. There will also be less knowledge spill over
within lower-technological industries because the technology and production
techniques in these industries are not being constantly improved and built
upon as they are in high-technology industries. Low-technology industries
will benefit from increased city size, but their productivity will not be
nearly as drastic as the growth in productivity in high-technology industries
that benefit the most from the byproducts of city growth.
Section 5: The Effect of City Size on Urban Demographics
It is clear that city size positively affects the productivity of firms
within those cities. It is now important to explore how this increase in
productivity has on the demographics of cities. The general population trend
is that cities tend to attract young people and repel older, less skilled
individuals (Glaeser 17). As a result cities have relatively young, more
educated populations than non-metropolitan areas. The 1990 U.S. census illustrated
that 39.7% of the metropolitan population in the United States were between
the ages of 18-34. Only 24.4% of the metropolitan population was between
the ages of 50-65. In non-metropolitan areas 37.1% of the population was
between the age of 18-34 and 27.5% of the population was between the ages
50-65. Thus in 1990, the percentage of young people living in metropolitan
areas was 2.6 percentage points higher than those living in non-metropolitan
areas while the percentage of older people was 3.1 percentage points lower
in metropolitan areas than in non-metropolitan areas.
The primary reason that young people are attracted to the city is that,
although the price of living in cities is considerably higher than the price
of living in non-metropolitan areas, the wages that comparatively higher
workers receive in metropolitan areas more than makes up for the higher
costs of living. Cities facilitate a greater marginal productivity of labor
and so firms are willing to pay specialized workers more (Glaeser 16). However,
workers do not necessarily come to cities with the specialized skills that
will allow them to attract higher paying jobs. Instead they move to the
city to experience faster wage growth over the next five years and eventually
earn as much as long-term residents do (Glaeser 16). As I have already discussed,
cities facilitate learning. City residents acquire skills and specializations
through the interactions and opportunities to learn from each other more
rapidly than those in non-metropolitan areas. Cities are the best location
for younger populations looking to accelerate their skill acquisition, make
themselves more marketable, and eventually receive a higher wage.
Older populations of workers on the other hand tend to have much less incentive
to accelerate their skill acquisition. First of all, they generally will
have less years in the labor force to utilize their skills, but also there
is a great likelihood that older workers already posses skills that they
have acquired over time and that many of these skills are transferable to
other locations. Workers who leave cities experience no visible wage declines
(Glaeser 16). This is in part because many older workers move out of cities
to less urban areas once their skills are transferable. It is also because
the cost of living is generally lower in non-metropolitan areas.
Section 6: India and Portugal
This section will move beyond theoretical explanations and cite two real
world examples of how city size has increased economic productivity.
Arup Mitra examined in 2000 the data from fifteen major states in India
in order to deduce the effects of urbanization on productivity and urban
growth. He illustrated how in over half of the industries, there was "an
association between total factor growth and agglomeration economies"
(Mitra 106). In India, a concentration of activities reduced the labor cost
and also encouraged workers to invest in more specialized capital. Also,
businesses in urban areas were able to take advantage of the infrastructure
base and communication networks that exist in urban areas (Mitra 106).
In 1986, Portugal experienced a large increase in the amount of foreign
direct investment (FDI) as a result of its acceptance into the European
Union (Guimaraes 118). A significant share of this FDI was directed at the
manufacturing sector. Foreign firms had to decide where to place their manufacturing
plants in order to maximize their production. A study of foreign direct
investment reveals that the existence of agglomeration economies were the
major factor in determining location decisions (Guimaraes 122).
Foreign investors chose to invest overwhelmingly in Porto and Lisbon, Portugal's
two principle cities for many of the reasons that this paper has already
explored (Guimaraes 125). Investors chose areas with large concentrations
of suppliers of inputs and urban services. These provide a large pool of
workers, as well as a large conglomeration of thriving foreign firms and
indicate to investors who may be less knowledgeable about the region that
the region's general conditions are favorable for production and investment
(Guimaraes 124). Porto and Lisbon's developed transport infrastructures
were also a major influence on investor decisions because it greatly reduced
transaction costs (Guimaraes 126).
Section 7: The Risk of Crowding
There is little doubt about the positive correlation between city size and
productivity. However, there is some risk that cities can become too crowded
and crowding externalities that are potentially harmful to economic growth
may result. Over-crowding, particularly of a highly educated populous, lowers
the wages of educated workers (Peri 594). As a result, workers become less
motivated to work hard and invest in improving their skill sets due to the
decreasing returns on education. Also, overcrowding increases the price
index as a result of higher demands on goods and housing and affords workers
less opportunity to save and reinvest in the economy.
However, as we have already discussed, a concentration of highly educated
workers also generates agglomeration economies and improves the overall
efficiency and production of an economy. This can offset many of the possible
constraints placed on an economy as a result of overcrowding. Also, as I
discussed in section 5, older and highly educated workers are mobile and
will live in places that give them the highest utility. Crowding is often
caused by the influx of younger workers into a city where the opportunities
for education and training are abundant. Many older workers offset these
trends by moving out of urban areas after they have acquired transferable
skills and want to take advantage of the lower costs of living (Peri 594).
Conclusion
City size has a clear, positive effect on productivity. Externalities of
population agglomerations like increased access of production inputs, lower
transaction costs, reduced economic volatility, larger and more specialized
labor pools, and economies of scale are the reason that cities were first
created and the reason they continue to exist and grow. City size will also
affect the demographic settlement patterns and education level of an economy.
There are some potential constraints on the benefits of city size; however,
for the most part these constraints are greatly outweighed by the benefits
of urbanization.
Feser, Edward J. (2002), "Tracing the Sources of Local External Economies",
Urban Studies, v39, n13 (December 2002): 2485-2506
Glaeser, Edward L. (1997), "Learning in Cities", National Bureau
of Economic Research Working Paper: 6271 November 1997.
Guimaraes, Paulo; Figueiredo, Octavio; Woodward, Douglas (2000), "Agglomeration
and the Location of Foreign Direct Investment in Portugal", Journal
of Urban Economics v47, n1 (January 2000): 115-135.
Mitra, Arup (2000), "Total Factor Productivity Growth and Urbanization
Economies: A Case of Indian Industries", Review of Urban and Regional
Development Studies v12, n2 (July 2000): 97-108
Peri, Giovanni (2002), "Young Workers, Learning, and Agglomerations",
Journal of Urban Economics v52, n3 (November 2002): 582-607.
Quigley, John M. (1998), "Urban Diversity and Economic Growth"
in Journal of Economic Perspectives, Vol. 12, No. 2, pp. 127-138.