Chapter 3. Building competitive and liveable metropolitan areas in Japan

This chapter looks at policies aimed at strengthening the competitiveness and liveability of Japan’s major cities, particularly the three great metropolitan areas centred on Tokyo, Osaka and Nagoya. It begins with a description of Japan’s urban system and an analysis of recent trends in urban settlement patterns and in economic and environmental performance. This is followed by a discussion of Tokyo’s global competitiveness and its future. The future of Tokyo is intimately connected with plans to forge Tokyo, Osaka and Nagoya into an urban mega-region, united by super-high-speed magnetic levitating trains. The chapter reviews the prospects for this effort, looking at international experience with such mega-projects and at the “soft” policies that should accompany this infrastructure effort. Finally, it examines ways in which Japan can make the most of its investment in hosting the 2020 Olympic Games in Tokyo and at social policy for cities.


Since 2012, the government has emphasised the need to ensure the international competitiveness of Japan’s major cities. It also emphasises the need to revitalise smaller cities and towns by stimulating private and business investment. There are two distinct strategies by types of cities; for major cities, international competitiveness is clearly the priority, while for local cities, liveability and sustainability are prioritised. This chapter is primarily focused on the former: it explores the challenges facing Japan’s large metropolitan areas, with a particular focus on the Tokyo metropolitan area, which in 2012 was home to more than one-quarter of Japan’s population and was slightly larger than the next nine Japanese metropolitan areas. However, some of the findings presented here have implications for second- and third-tier cities as well. Chapter 4, by contrast, concentrates on the challenges of “non-metropolitan Japan”, but some of its conclusions are relevant to the metropolitan areas. The chapter begins with an overview of Japan’s urban system and a discussion of trends in economic performance, the evolution of the urban system and environmental outcomes. This is followed by a focus on the global competitiveness of the largest cities, above all Tokyo, and a look at some of the ways in which urban policies in the major cities can help address such cross-cutting challenges as fertility, ageing and poverty.

Japan’s urban system

Japan is overwhelmingly urbanised and has a highly concentrated urban system

On the official data, Japan’s urbanisation rate now exceeds 93% (UNDESA, 2014), making it one of the most urbanised countries in the world, second only to Belgium in the OECD area. This figure is, however, somewhat misleading, inasmuch as the Japanese data are based on the proportion of the population living in municipalities defined as shi (cities). To qualify as a shi, a municipality must satisfy the following conditions: a population of at least 50 000, with at least 60% of dwellings located in the main built-up areas and at least 60% of employment in “urban activities” (manufacturing, retail or wholesale trade, etc.). This means, first, that the urbanisation rate does include some people who are de facto rural dwellers and, secondly, that adjustments to municipal boundaries can affect the urbanisation rate substantially, even in the absence of any change in the settlement pattern. As seen in Chapter 2, the Heisei municipal mergers of the early 2000s reduced the number of municipalities by almost half. This was the major factor underlying the 16.7-point jump in Japan’s urbanisation rate between 2000 and 2015. A more meaningful, if restrictive, measure is the share of population living in so-called “densely inhabited districts”, defined as clusters of adjacent census blocks with total populations of at least 5 000 and population densities of at least 4 000/km2. In the 2010 census, some 67.3% of the population lived in such districts. That means that over two-thirds of Japan’s population lived in places with population densities higher than those found in such western cities as Naples, Berlin and Manchester.1

Because differences in statistical methods across countries make such global comparisons rather problematic, most of the comparative analysis in this chapter will rely, to the extent possible, on the method for calculating functional urban areas (FUAs) described in OECD (2012a), which is outlined in Box 3.1 below. This method aims to define cities as functional economies, in terms of actual settlement patterns and labour market flows, instead of on the basis of administrative boundaries. The FUAs so defined exclude urban dwellers in small towns and cities, while including rural dwellers living within the commuter belts of large urban areas. For this reason, they do not provide an alternative measure for the urbanisation rate in Japan or anywhere else – that it not their purpose. However, they do provide a basis for comparisons within and across countries, and they confirm that Japan is highly urbanised and has a concentrated urban system.

Box 3.1. Defining functional urban areas

This review relies, where possible, on the OECD Metropolitan Database, using the method for defining functional urban areas set out in OECD (2012a). This is similar to the calculation of statistical metropolitan areas in the United States and some other countries, but the method is somewhat simpler, in order to allow its application across OECD countries, many of which do not generate the data that would be needed to apply the US method. FUAs are defined in terms of human settlement and economic activity rather than administrative borders. This matters, because, even within countries, the relationship between the city as an administrative unit and the city as a socio-economic entity can vary widely. In Italy, for example, a city like Bologna largely corresponds to its administrative definition, whereas the “functional” city of Venice is far larger than the Comune di Venezia itself and comprises more than 120 municipalities (OECD, 2015a).

The FUA method, constructed in collaboration with the European Commission’s Directorate-General for Regional and Urban Policy, involves a three-step approach that can be summarised as follows:

  • Defining urban cores through gridded population data. Urban cores are constituted by aggregations of contiguous municipalities that have more than 50% of their population living in high-density clusters. The latter are made of contiguous 1 km2 grid cells with a population density of at least 1 500 inhabitants per km2 (1 000 inhabitants per km2 in and Canada and the United States) and a total population of at least 50 000 people (100 000 in Japan, Korea, Mexico).

  • Connecting non-contiguous cores belonging to the same functional area on the basis of commuting data. Two urban cores are considered integrated, and thus part of the same metropolitan system, if more than 15% of the working population of any of the cores commutes to work in the other core (taking polycentricity into account).

  • Identifying the urban hinterlands. The worker catchment area of the urban labour markets, outside the cores is composed of municipalities which send to the cores 15% or more of their employed residents. Municipalities surrounded by a single functional area are included and non-contiguous municipalities are dropped.

This common FUA definition allows for meaningful comparisons within and across countries (comparing like with like) and also makes it possible to identify levels of mono- or polycentricity of FUAs, as well as the extent of concentration.

Source: OECD (2012a), Redefining “Urban”: A New Way to Measure Metropolitan Areas, OECD Publishing, Paris,

Japan has 76 FUAs with populations of at least 50 000 (Figure 3.1); these were home to 77% of the population in 2012. Six FUAs have populations in excess of 1.5 million; collectively, these 6 account for 51% of the population. Greater Tokyo is by far the largest FUA in Japan – and on some measures, the largest urban area in the world – with a population of over 35.4 million in 2012, up from 32.8 million at the turn of the century. Near it on the western (Pacific Ocean) side of the island of Honshū are the second and third metropolitan areas of Japan, Osaka and Nagoya, with 17.3 million and 6.5 million people, respectively, in 2012. Taken together, the Tokyo, Osaka and Nagoya FUAs constitute a massive urbanised region with a population of almost 59.2 million (2012). Among OECD countries, only Korea has a larger share of population living in metropolitan areas of 1.5 million or more (Figure 3.2).

Figure 3.1. Functional urban areas in Japan with populations above 50 000

Note: This map is for illustrative purposes and is without prejudice to the status of or sovereignty over any territory covered by this map.

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Figure 3.2. Distribution of population by size of FUA, selected OECD countries, 2012

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Japanese FUAs are comparatively dense by OECD standards. As can be seen in Figure 3.3, this holds good even when allowing for the well-known fact that larger cities tend to be denser: across the size spectrum, Japanese FUAs tend to be denser than most FUAs of similar sizes elsewhere in the OECD. Greater Tokyo is the fourth-densest large metropolitan area in the OECD (the top three are all in Korea). The density of Japanese cities is even more apparent further down the urban hierarchy. When the nearly 1 200 FUAs in OECD countries that have populations in excess of 50 000 are ranked according to density, only 7 of Japan’s 76 FUAs are in the bottom half of the distribution.

Figure 3.3. Population densities in OECD metropolitan areas

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Urbanisation in Japan is continuing

While the very high urbanisation rate cited above is largely the product of a jump caused by the heisei mergers, both the administratively-defined urbanisation rate and the share of population living in “densely inhabited districts” continues to rise. The UNDESA projections see Japan’s urbanisation rate rising to more than 97% by mid-century as a result of increasing migration to shi. These projections should probably be viewed with some caution, not least in view of the problems with the definition of urbanisation based on population in shi.

  • While the methods employed for estimating further rural-to-urban migration are well known, there has been some concern that they tend to over-predict urbanisation rates, particularly in Asia, and are thus subject to regular downward revisions (Kundu, 2014).

  • No large country has such a high share of population already living in such high-density areas, nor is any other country experiencing the kind of dramatic demographic change that is unfolding in Japan. It is possible that an increasing elderly population will concentrate in cities but, as noted in Chapter 1, it may also be that the “third age” elderly (65-75) will move to less dense places and the “fourth age” elderly (75+) will tend to move back to the cities.

Much will depend on the pace of development of information and communications technologies and of other technologies that will have a disproportionate effect on older people – not only in healthcare but in such areas as transport (e.g. driverless cars). In other words, Japan’s urban and demographic situations are without precedent, so UNDESA’s long-term projections should be taken with some scepticism. While there is no obvious reason to expect de-urbanisation, it may be that the urbanisation rate will stabilise sooner than the projections suggest.

Yet even if the projections should prove high of the mark, the FUA data suggest that the urban system continues to grow more concentrated: there has been a clear relationship between FUA size and population growth since the turn of the century (Figure 3.4). During 2000-12, the 6 FUAs with populations in excess of 1.5 million recorded average annual population growth of 0.4%, while the 30 FUAs with 500 000 to 1.5 million inhabitants shrank by an average of 0.2% per year and the remainder (with under 500 000 inhabitants) shrank by 0.3% per annum. This result is in line with the prefectural-level trends seen in Chapter 1, which showed that people were, on the whole, tending to move from less populous, less dense places to prefectures with higher populations and population densities. There were exceptions, however, and it is noteworthy that the highest growth rates were recorded in relatively small FUAs – notably Anjō (0.84% per year), Toyota (0.72%) and Okazaki (0.67%).

Figure 3.4. FUA size and population growth

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Trends in economic performance

Like its settlement pattern, Japan’s economy is overwhelmingly urban

Concentration of population is largely matched by concentration on other dimensions. Japan has the highest share of GDP generated in cities of 500 000 or more in the OECD and is second only to Korea in the concentration of population and employment. The 36 FUAs with populations of at least 500 000 (hereafter simply “metropolitan areas”) accounted for 69.8% of total output in 2010 (Figure 3.5). Moreover, such metropolitan areas generated roughly 75% of Japanese growth over 2000-10, the second-highest metropolitan growth contribution after Greece, where Athens alone accounted for 79% of GDP growth over the decade. The 75% figure recorded for Japanese FUAs’ growth contribution stands somewhat above their population share (68% in 2012), as well as their share in GDP in 2001 (69.5%). This confirms that they remain the major sources of dynamism in the economy. A close look at the data reveals a slow but steady increase in the concentration of activity during the years leading up to the global crisis, followed by a partial reversal in 2009-10, as the major metropolitan areas bore the brunt of the downturn.

Figure 3.5. Concentration of population, GDP and employment in OECD metropolitan areas, 2010

Note: Countries with only one metropolitan area above 500 000 population are not shown.

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Tokyo’s contribution to growth, at more than 43% over 2000-12, was far in excess of its share of Japan’s GDP (Figure 3.6) and was exceptionally large for the leading city of a large country. Though lower than the contributions recorded by Greater Paris and Seoul-Inchon in their respective countries, it was larger even than the contribution to national growth of London in the United Kingdom or of any metropolitan area in the United States, Germany or Mexico. In contrast to Tokyo, Osaka – Japan’s second-largest FUA – seriously under-performed relative to its size over the decade. Its contribution to aggregate growth was far smaller than its relative weight in the economy. Nagoya, with a population not much more than one-third that of Osaka, made a larger contribution to national growth. A significant proportion of Japan’s other metropolitan areas likewise performed relatively well compared to their size. However, because this was a decade of relatively weak overall growth, the change in the spatial distribution of activity resulting from this very concentrated growth pattern was quite limited. Tokyo’s share of aggregate GDP rose just under 0.9 percentage points over a decade, while the share of non-metropolitan Japan fell by less than 0.4 percentage points. Given faster growth overall, a similar concentration dynamic would have resulted in a far larger structural shift.

Figure 3.6. Initial GDP share and contribution to growth, 2001-10
Japanese FUAs

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Disparities among major metropolitan areas and between large cities and other areas are limited

Viewed in per capita rather than aggregate terms, Japan’s urban economy differs from those of most other OECD countries. In particular, the per capita GDP gap between metropolitan areas and the national average is exceptionally small, and many Japanese metropolitan areas in 2010 reported levels of GDP per capita that were below the national average (Figure 3.7) – though in an economy as urban and as concentrated as Japan’s, this average was itself heavily influenced by the very largest cities. On average, metropolitan areas in OECD countries with populations in excess of 500 000 generate GDP per capita that is about 40% higher than elsewhere. In Japan, this gap stood at just 10.1% in 2010. This is a further confirmation of the small urban-rural income gap described in Chapter 1. Moreover, non-metropolitan Japan experienced substantially faster growth of GDP per capita during 2001-10 than did the metropolitan areas: the former grew at an annual average rate of almost 1.1% and the latter at a rate of just under 0.6%. The crisis, of course, played a role here: its impact was far greater in metropolitan areas, with the result that the metropolitan/non-metropolitan gap in GDP per capita shrank somewhat over the course of the decade (Figure 3.8). The other key factor was movement of labour: areas that made outsize contributions to aggregate growth sometimes saw slower growth in per capita or per worker terms. Even among the metropolitan areas themselves, the contrast between aggregate and per capita growth is striking. Larger metropolitan areas, which have been attracting population, tended to record larger gaps between aggregate and per capita growth (Figure 3.9).

Figure 3.7. GDP per capita, 2012

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Figure 3.8. Trends in GDP per capita: Metropolitan and non-metropolitan Japan

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Figure 3.9. GDP growth, aggregate and per capita rates, 2001-10
Annual average real growth rate, %

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

The labour market seems to play a critical role in limiting disparities

The contrast between aggregate and per capita dynamics implies that migration and labour market flows serve to limit the growth of inter-regional or urban-rural wage disparities. Some other labour market indicators reinforce this conclusion. Unemployment rates have remained comparatively low across Japan even during the worst period of the crisis (generally below 5% and never above 5.5%), and the gap between the unemployment rates in metropolitan areas and non-metropolitan Japan did not exceed 0.4 percentage points during the period for which data are available (2000-11). The unemployment rates in the two labour markets moved more or less in lock-step (Figure 3.10). Moreover, Japan had the second-smallest variance in unemployment rates across its metropolitan areas in the OECD in 2012 (Figure 3.11).

Figure 3.10. FUA/non-FUA unemployment gaps, 2000-12
Difference between FUA and non-FUA unemployment rates, in percentage points

Note: The years indicated for each country are the years in which it recorded the maximum/minimum unemployment gaps.

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Figure 3.11. Countries ranked by variation in unemployment rates across metropolitan areas

Note: Only countries with at least five metropolitan areas (FUAs with 500 000 people or more) are included.

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

The data on employment creation and destruction make clear that this result did not reflect broadly uniform labour market conditions. Rather, it stemmed in large part from internal mobility in response to very different labour market circumstances in different places: employment creation (and, during the downturn, employment destruction) was concentrated in Tokyo. Over the decade, the capital region came close to matching the reduction in jobs that occurred outside the 36 major metropolitan areas (Figure 3.12). Only ten metropolitan areas recorded increased employment, while 26 recorded declines, in addition to the roughly 1.3 million jobs that disappeared in non-metropolitan Japan. The result was an overall decline in employment, but one that would have been far more dramatic absent Tokyo’s role as a centre of job creation.

Figure 3.12. Contributions to employment growth, 2001-12

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

The foregoing suggests that internal migration has led to increased spatial concentration of activity in aggregate but has also served to limit spatial disparities. The growth of Tokyo, in particular, has aroused some concern among policy makers, but it appears to have been critical in ensuring that decades of slow growth did not lead to high unemployment in the regions and a sharp increase in inter-regional income disparities. The problem, in that sense, lies not in Tokyo but in the lack of dynamism observed elsewhere.2 This is an issue to be borne in mind when contemplating policies to check the inflow of labour to the largest metropolitan areas. It suggests, in particular, that policies that stimulate business entry and job creation outside the large cities would be the most important way to offset the centripetal forces now at work in the labour market.

Changes in the size and composition of population mean that the relationship between labour productivity and GDP varies somewhat across FUAs, as it does across regions. GDP per worker has grown faster than GDP per capita, in 28 of Japan’s 36 metropolitan areas – a reflection of population ageing in urban areas, which accelerated over the course of the decade (OECD, 2015c). In only 2 Japanese FUAs did either figure rise at an annual average rate of 2% or more, and in the great majority, GDP per capita growth was below 1% per annum. Far more sobering, though, is the evolution of the productivity gap between FUAs and the national average (Figure 3.13). Many Japanese FUAs show productivity levels that are below the national average. To be sure, this is in part because that average is heavily skewed by the large relative weight of Greater Tokyo in the economy. As can be seen from the figure, the crisis appears to have arrested the growth of productivity differentials in many places, which widened somewhat between 2001 and 2008 before shrinking in the 2008-12 period.

Figure 3.13. Evolution of the productivity gap: Japanese FUAs relative to the national average, 2001-12

Note: The productivity gap is here defined as the difference between FUA, GDP per worker and national GDP per worker.

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Japanese cities do still generate strong agglomeration benefits

A growing body of evidence and analysis confirms that, other things being equal, larger cities make their residents more productive (OECD, 2015d). In part, this can be explained by the characteristics of the workforce and the firm population in cities. Larger cities tend to attract highly educated and experienced residents, who would have high levels of productivity wherever they choose to work and live, and they also attract more productive firms. However, the evidence is now clear that cities themselves can contribute significantly to their residents’ productivity. Residents of larger cities are more productive than if they were living in smaller cities due to “agglomeration benefits”, the positive productivity spillovers that arise when highly productive firms and people are brought closer together. Typically these benefits are created through shared inputs, better “matching” between firms and employees, and mutual learning among firms and residents (Box 3.2). “Knowledge spillovers”, in particular, are considered critical in explaining the benefits of cities in modern service-oriented economies.

Box 3.2. Agglomeration economies

Three main mechanisms work to produce agglomeration economies:

1. Mechanisms that deal with sharing of:

  • Indivisible facilities such as local public goods or facilities that serve several individuals or firms. Some examples, other than public goods, are facilities such as laboratories, universities and other large goods that do not belong to a particular agent but where some exclusion is implicit in providing them.

  • The gains from the wider variety of input suppliers that can be sustained by a larger final goods industry. In other words, the presence of increasing returns to scale along with forward and backward linkages allow firms to purchase intermediate inputs at lower costs.

  • The gains from the narrower specialisation that can be sustained with higher production levels. Several firms specialise in producing complementary products, reducing overall production costs.

  • Risks. This refers to the idea that an industry gains from having a constant market for skills. If there are market shocks, firms can adjust to changes in demand if they have access to a deep and broad labour market that allows them to expand or contract their demand for labour.

2. Matching mechanisms by which:

  • Agglomeration improves the expected quality of matches between firms and workers, so both are better able to find a good match for their needs.

  • An increase in the number of agents trying to match in the labour market also improves the probability of matching.

  • Delays are alleviated. There is a possibility that contractual problems arising from renegotiation among buyers and suppliers will result in one of the parties losing out to the other party in a renegotiation. However, if the agglomeration is extensive enough, agents can find an alternative partner.

3. Learning mechanisms based on the generation, diffusion and accumulation of knowledge. This refers not only to the learning of technologies, but also the acquisition of skills.

OECD metropolitan regions benefit from agglomeration effects and thus tend to display higher levels of productivity, higher rates of employment and higher levels of GDP per capita than other regions. These benefits, however, are limited by congestion costs, diseconomies of scale and oversupply of labour, among other potential negative elements, and many metropolitan regions have in recent decades tended to underperform national economies.

Sources: Duranton, G. and D. Puga (2004), “Micro-foundations of Urban Agglomeration Economies”, Handbook of Regional and Urban Economics, Vol. 4/4, February; OECD (2009a), OECD Economic Outlook, OECD Publishing, Paris,

To disentangle the contribution of employees’ characteristics from the contribution of the urban environment to local productivity, the OECD uses individual earnings as a proxy for productivity.3 The analysis uses the wages and characteristics of employees from the “Basic Survey on Wage Structure”, a large-scale linked employer-employee survey that covers active firms with at least ten employees in mining, manufacturing or the service sector (excluding government officials and the military). To calculate the differential productivity benefits across cities, workers’ wages are decomposed into a part that can be explained by their personal characteristics and their jobs’ characteristics, on one side, and a remainder that is attributed to the city in which they work, on the other. The characteristics that are taken into account include education, experience, occupation and gender. Because the analysis relies on microdata, it extends to all 76 Japanese FUAs with at least 50 000 residents. The results of this analysis show that Japanese cities generate relatively strong agglomeration benefits, comparable to those observed in the United States (Figure 3.14). The contrast with the United Kingdom is striking, since London’s dominance of the UK economy is often compared to Tokyo’s. Proximity to London seems to matter more than size in explaining the extent of such benefits.

Figure 3.14. Estimated agglomeration benefits across cities in three OECD countries

Note: City productivity is defined as a wage premium associated with each city once the characteristics of the city workforce are taken into account. Individual level wage regressions are estimated with controls for the characteristics of the workers, in order to account for sorting of individuals to cities.

Sources: OECD calculations based on national sources: “UK Annual Survey of Hours and Earnings” (ASHE) for 2003-10; US Census from 1990 and 2000 with the “American Community Survey” for the years 2005 to 2007; for Japan calculations are based on Ministry of Health, Labour and Welfare (n.d.), “Basic Survey on Wage Structure”, (accessed 15 July 2015) and Ahrend, R. et al. (2014), “What Makes Cities More Productive? Evidence on the Role of Urban Governance from five OECD Countries”, OECD Regional Development Working Papers, No. 2014/05, OECD Publishing, Paris,

Urban governance challenges remain to be addressed

The second stage of the analysis looks at whether and to what extent Japan conforms to the findings reported in OECD (2015d, 2015e) concerning the impact of metropolitan fragmentation on performance. The governance fragmentation indicator used is that discussed in Chapter 2 – municipalities relative to population. Two things are striking about the results (Figure 3.15). First, the negative relationship between fragmentation and productivity across the 76 FUAs is remarkably tight – a much closer fit than similar analyses have yielded in other countries. Secondly, the result is not driven by the largest cities, which have lower fragmentation indexes, but by a large number of small and medium-sized cities that spread over municipal boundaries. As observed in Chapter 2, such a result should not necessarily be seen as grounds for another wave of mergers but it does suggest a need for greater co-ordination among localities at the scale of effective functional urban economies. This reinforces the emphasis of the Grand Design on linking up cities and it also provides evidence in support of the Ministry of Internal Affairs and Communications’ (MIC) efforts to enhance horizontal co-operation at local level. It is important not to neglect the governance dimension in seeking to strengthen cities outside the biggest metropolitan areas. Connective infrastructure may also help but a growing body of evidence suggests that governance co-operation is essential, particularly in areas like land use, transport planning and urban development, where co-ordination is particularly critical (OECD, 2015d, 2015e).

Figure 3.15. Governance fragmentation and performance

Note: Size of the bubble corresponds to the population of the FUA.

Source: Author’s elaboration based on the data in Figure 3.14.

Japanese metropolitan areas are failing to fulfil their potential as drivers of innovation-led growth

These productivity outcomes are likely to be linked, at least in part, to weaknesses in the innovation system. As noted in Chapter 1, there is a striking imbalance between the resources devoted to knowledge creation in Japan and the observed outputs in terms of productivity. The problem is not primarily one of knowledge creation per se: Japan compares relatively well to other OECD countries in terms of patent intensity (Figure 3.16), for example, even if it lags behind some of the advanced economies of Northern Europe. In aggregate terms, Tokyo and Osaka rank first and third among OECD metropolitan areas in terms of patents filed. However, no Japanese metropolitan area ranks even in the top 20 for patent intensity (i.e. patents relative to population) – Osaka and Tokyo rank 30th and 31st, respectively on this measure. On the other hand, all Japanese metropolitan areas rank in the top half of OECD metropolitan areas for which data are available (Kumamoto ranks 103rd of 218). Japan is also active in pursuing demand-side policies for innovation, in an effort to stimulate, among other things, green innovation and innovation in support of an ageing society (OECD, 2011).

Figure 3.16. Patent intensity, selected OECD countries and their metropolitan areas, 2008

Sources: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

The striking thing about Japan’s urban innovation systems is how little dispersion in performance is observed, at least in terms of patents (Figure 3.17). The most patent-intensive Japanese metropolitan area (Osaka) generates just over twice as many patents per head of population as the least (Kumamoto). This is the second-lowest figure among the ten OECD countries for which data are available.4 For larger OECD countries, this differential typically falls in the range of 8-20, rising close to 100 for the United States. Overall, the concentration of knowledge creation thus reflects but does not appear to exceed that of the urban system. This is a striking result, given that science- and technology-intensive innovation processes benefit greatly from agglomeration economies and thus tend to concentrate within countries. It suggests that there is potential for innovation-driven growth across much of Japan. The key will be commercialising and diffusing the results of such research.

Figure 3.17. Patent intensity of Japanese metropolitan areas, 2008

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

The evolution of the urban system

Population densities continue to decline in most metropolitan areas

Trends in urban density in Japan have shifted over time. In the 1970s and 1980s, most Japanese cities lost density steadily and rapidly, due to expansive development on their peripheries – generally car-oriented low-density suburbs. Between 1970 and 2010, the average density of the densely inhabited districts (DIDs) in 47 prefectural capital cities fell by 23.8%, from 8 546 to 6 512/km2 – though it should be noted that the latter are still very high densities by OECD standards. The population of DIDs rose by 44% over the period, while the spatial extent of such districts expanded by 76% (Figure 3.18). Second-tier cities saw a far sharper drop in densities than the largest ones: the high-density urban areas in the 38 capital cities outside the 3 metropolitan areas defined by Statistics Japan5 expanded by 123%, compared to a figure of 41% for the 7 capital cities in the 3 metropolitan areas. The decline in urban densities has been more or less continuous throughout the period in most cities, due to continued urban expansion and, in more and more places, population decline in recent decades. However, the rate of decline has slowed with each successive decade, and densities have actually risen since 1995 in the three major metropolitan areas. Even so, some places continued to spread out rather quickly. Between 2000 and 2006, the built-up areas in several Japanese urban areas grew by more than 1% per year, although the population stagnated or even declined (Figure 3.19). By contrast, densities have increased in the three metropolitan areas (as defined by Statistics Japan), particularly in Tokyo. The average urban density in these areas recovered from 10 339/km2 in 1995 to 11 017/km2 in 2010.

Figure 3.18. Size and density of densely inhabited districts of prefectural capitals

Source: OECD calculations based on the Census data for 1970-2010 provided directly by Official Statistics Japan (Government of Japan).

Figure 3.19. Fastest spatial expansion observed among OECD metropolitan areas with population growth under 1% per annum (2000-06)

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

This two-way trend seems set to continue. On one hand, given the national trend of population decline, most Japanese cities are going to be less dense. By 2040, Japan’s population is expected to return roughly to the level of 1970. Most cities will lose density, as their built-up areas, which have nearly doubled in extent since 1970, will not shrink nearly as fast as population. On the other hand, densities in the three metropolitan areas may continue to increase or at least stabilise.

Falling densities should not be seen as bad news, always and everywhere. Changes in density imply important trade-offs (Table 3.1), but if well managed, lower densities might allow for lower house prices, some reduction in congestion and local environmental problems, and the expansion of public green space. Moreover, given Japan’s demographic situation, some degree of de-densification could be beneficial and efforts to maintain, let alone increase, urban densities in the major cities should be approached with great caution. Lower densities might in fact contribute to increased fertility rates. Historically, rising incomes are associated with increasing demand for private space more or less everywhere in the world, even when cities have remained relatively compact, dense and focused on public transit. When the supply of private space is inelastic, the easiest way for households to enjoy more private space per person as they grow wealthier is to opt for smaller families. That is one reason why places like Hong Kong, China; Macau, China; Singapore and Tokyo have among the lowest levels of fertility in the world. It is also one of the reasons why Manhattan has a lower population density today than it did a century ago (Yglesias 2014).

Table 3.1. Advantages and disadvantages of high urban densities

Potential advantages of high urban densities

Potential disadvantages of high urban densities


  • Reduce fossil fuel emissions/carbon footprint by decreasing the total number of vehicle trips and the number of kilometres travelled per trip.

  • Enhance accessibility, as people live closer to where they work, shop and play.

  • Make transit more economically viable and efficient.

  • Enable public health benefits from more walkable and bike-friendly environments.

  • Create efficiencies in mixed-use developments through shared parking.

  • Exacerbate traffic congestion, parking problems; increased traffic accidents.

  • Create pedestrian congestion and congestion in public transport.

  • In compact, monocentric cities may only have significant positive environmental effects when a greater share of commuters use mass transit.

Land/resource use

  • Make better use of existing resources and infrastructure.

  • Reduce development pressure on green spaces, agriculture and industrial land.

  • Create a greater mix of land uses.

  • Limit recreational opportunities and reduce the availability of green/open space.

  • Reduce an area’s capacity to absorb rainfall because of open space/recreational loss.

  • Exacerbate pollution, possibly because of reduced area for trees/vegetation.

Social equity

  • Reduce segregation and social exclusion.

  • Add diversity, safety, vitality, and opportunities for creative and social Interaction.

  • Provide access to facilities (e.g. schools, employment, shops), without the need for a (costly) private vehicle.

  • Reduce crime by increasing pedestrian activity and fostering a 24-hour community (more ‘eyes on the street’).

  • Lead to loss of privacy and increased noise, nuisance, etc.

  • Reinforce social inequality and social segregation.

  • Increase crime.

  • Generate cramped living environments.

Economic development

  • Enable investments in community amenities as well as better quality and more attractive building materials.

  • Promote a critical mass necessary to support local retail and service areas.

  • Additional cost to build and maintain high-density projects and city-centre infrastructure.

  • Higher relative prices for land, housing, and many other goods and services.

Environmental sustainability and energy

  • Preserve green open spaces, clean air and water, fauna and flora systems (when higher-density development occurs elsewhere).

  • Facilitate innovative green design and district energy; reduce water and energy consumption.

  • Facilitate the technological and economic viability of certain energy technologies and transport systems.

  • Higher energy consumption during the construction of high-density buildings.

  • Higher exposure to particulate matter and other pollutants subject to local concentration.

  • Limit some forms of ambient energy systems.

  • Increased noise.

Sources: Adapted from Boyko, C. and R. Cooper (2011), “Clarifying and re-conceptualising density”, Progress in Planning, Vol. 76, No. 1, pp. 1-61, July,; Fundación Idea/Cámara de Senadores/SIMO Consulting (2014), “México compacto: Las condiciones para la densificación urbana inteligente en México”, Fundación Idea, Cámara de Senadores and SIMO Consulting, Federal District, Mexico, available at:; Gaigné, C., S. Riou and J.-F. Thisse (2012), “Are compact cities environmentally friendly?”, Journal of Urban Economics, Vol. 72, No. 2-3, pp. 123-136, September-November,

A significant body of empirical research from the 1960s to the 2000s confirms that density is negatively correlated with fertility. This holds true in a variety of settings, including rural and urban, and at a range of different scales from national to local.6 Using fixed effects models on time series data covering 145 countries and controlling for key socio-economic variables, Lutz et al. (2006) find a consistent and significant negative relationship between fertility and population density. Their findings also suggest that individual fertility preferences decline as population density rises – i.e., that people prefer to have fewer children when living in denser environments. Using time series data for 187 countries over the period 1960-2000, Lutz and Ren (2002) find that when explaining the level of fertility, population density is far more important than such traditionally studied factors as female labour force participation, income or food security, though it is far less important than female literacy.7

Prefectural-level data for Japan suggest something similar at work (Figure 3.20). To be sure, other factors do matter, but the close fit for the high-density variable is striking.8 Moreover, in a Japan with a rapidly growing population of over-65s, some reduction in densities may make it easier and more attractive for some families to return to a more traditional three-generation model of the household, which could help address both fertility and ageing challenges. The prefectural-level data suggest that the greater incidence of three-generation households is associated with higher fertility and higher female labour force participation.

Figure 3.20. Fertility and concentration of population in high-density areas
Japanese prefectures, 2010

Source: OECD calculations based on data provided directly by Official Statistics Japan (Government of Japan).

The decline in urban densities must be managed

To recognise that lower densities may have certain advantages is not to suggest that concerns about compact development should simply be set aside.

  • Any reduction in density will need to be managed. Japan has no interest in allowing uncontrolled urban sprawl, nor does it want declining densities to destroy the cohesion of its major urban areas. This would risk leaving them as “perforated cities”, patchworks of density and abandonment that would be economically inefficient and environmentally problematic.

  • There is far more to compact urban development than density (OECD, 2012a). On the contrary, many of the principles underlying compact development policies (mixed use, transit-oriented development, careful planning of green spaces, avoidance of leapfrog development) can be employed to allow some reduction in densities while mitigating many of the problems that lower densities might bring. The emphasis on compactness will also underscore the need to maintain the cohesion and coherence of the urban space.

  • This is certainly not an argument against density as an option for those who want it – many will doubtless prefer high-density living in lively urban cores. It is rather an argument against seeing de-densification always and everywhere as bad news.

  • Density is related to size. Other things being equal, larger cities will tend to be denser than smaller places, and this is economically and environmentally appropriate.

This is in keeping with one of the themes of this report, which is that the evolution of Japan’s settlement pattern will need to vary from place to place, creating a variety of location options for households and firms. Too much standardisation will reduce the attractiveness of most places, as it will undermine their ability to differentiate themselves according to their own assets.

Locally tailored approaches are needed when implementing “compact city” policies

As seen in Chapter 2, Japan is promoting a compact and networked urban structure to address its population decline and ageing challenges. The 2014 revisions to the Urban Renaissance Act and Local Public Transport Revitalisation Act aim to maintain population density, to promote the use of public transportation and to guide the location of urban services in designated urban areas, with various instruments (Box 3.3). Although it is relevant to all types of cities, the initiative is mainly understood as a spatial strategy for local cities, and is so listed in the Revitalisation Headquarters’ Comprehensive Strategy in December 2014. The Strategy envisages that master plans for the location of urban functions and local public transport network development plans based on the two Acts should be developed in 150 and 100 municipalities, respectively by 2020.

Box 3.3. Policy initiatives for promoting “compact and networked” urban structure

As Japanese cities have expanded, traditional urban centres in Japan have suffered from depopulation and a hollowing-out of urban functions. Since the 1990s, the national government introduced various initiatives to revitalize urban centres. The 1998 City Centre Revitalisation Act introduced comprehensive measures to revive urban centres, including extended national grants for public investment and housing projects and a funding mechanism to stimulate private investment in designated areas. The 2008 revision of City Planning Act and City Centre Revitalisation Act introduced stringent land-use regulations to control the location of large-scale retail stores in suburban areas. Such efforts to protect city centres have come at a price, however: as OECD (2008) observes, they are among the reasons why Japan’s productivity performance in the retail sector has been among the worst in the OECD since the 1980s.

Most recently, the 2014 revision of Urban Renaissance Act introduced comprehensive measures to achieve compact and networked urban structure. It encourages municipalities to develop comprehensive master plans to guide private and public investment in the target areas supported by public transport networks. Various incentives and disincentives will be provided in order for the relevant actors to follow the plan. Each plan will specify:

  1. Districts to guide the location of urban functions. The location of urban functions such as welfare, medical, commercial and other vital facilities is to be promoted in this district. Tax incentives, subsidies and financial assistance are provided for the facilities to locate in the district. Floor area ratio bonuses may be awarded for the reconstruction of these facilities. The plan should specify instruments to promote the effective use of under- or unused land. Planning regulation is also used to promote walkable environments. Moderate control over the deployment of urban functions outside of the designated area is possible.

  2. Districts to guide the location of residential functions. The location of housing developments is promoted in this district to maintain population density. Moderate control on residential development outside of the designated area is possible.

  3. Public transport. Public transport networks should be maintained in keeping with the above designations of districts, so that private developers and individuals are aware of them when making location choices.

  4. Management and effective use of demolished-building sites. The plan also describes how to better manage and use effectively demolished-building sites. Agreement system for the maintenance of demolished-building sites by NPO etc. is introduced. Urban farming, including direct sales store, in demolished-building sites will be encouraged.

In order to support implementation of these provisions, an inter-ministerial support team was established in March 2015 to discuss possible support tools.

From the public transport network perspective, the 2014 revision of Local Public Transport Revitalisation Act encourages municipalities to develop local public transport network development plan, which facilitates coordination between public transport policies and land use policies to promote a compact city (Ministry of Land, Infrastructure, Transport and Tourism, 2015).

Source: Ministry of Land, Infrastructure, Transport and Tourism (2015), Brochure of the revised Regional Public Transport Revitalisation Act, 8th edition, July 2015, available at:

The desire to avoid too much decay in the spatial integrity of cities that are losing population is surely correct, but it will require locally tailored approaches. In Tokyo, and perhaps also Osaka and Nagoya, urban density is going to increase, at least for some years yet, as they still have strong development pressure.9 High urban density can enhance quality of life and economic competitiveness, but only if certain other challenges are addressed, particularly ensuring quality public transport, promoting mixed-use development and, especially in Japanese conditions, developing age- and family-friendly urban spaces. Traditional planning instruments are still important in these cities, so proposed new urban expansion (green-field development) needs to be carefully compared with infill development alternatives, particularly in terms of environmental impact. It is important to understand compact city policies not just as densification, but as a comprehensive strategy to promote efficient land use in existing built-up areas and to minimise the disadvantages of high densities, such as congestion and pollution.

Most other cities in Japan will experience population decline and hence lower urban densities. Indeed, by 2012, 16 of Japan’s 36 FUAs with populations above 500 000 were losing population. While lower urban density can bring some benefits (less congestion, more affordable housing, more public space per capita, etc.), it also implies a number of potential challenges. In terms of mobility, lower urban density will make it harder to sustain the efficiency of public transport networks. Unless other solutions are found, this is likely to lead to greater reliance on private automobiles. In Japan, public bus and railway networks are already shrinking; since 2007, more than 10 200 km of bus routes and 186 km of railway networks has been abolished (Ministry of Land, Infrastructure, Transport and Tourism, 2015). In terms of the local economy, it will be difficult under lower urban density to maintain the critical mass necessary to support local retail and service areas, especially in medium-sized cities. As noted in Chapter 2, population decline will also make it harder for them to maintain their infrastructure. In this context, managing and limiting the loss of density would be an important objective for sustainable urban management. Some decline in urban densities can be accepted, and may even be beneficial, but cities will want to maintain sufficient density to sustain good public services and quality of life.

Environmental performance of Japanese metropolitan areas

CO2 emissions from most Japanese metropolitan areas are relatively low by OECD standards…

Most Japanese metropolitan areas have comparatively low CO2 emissions per capita by OECD standards (Figure 3.21), with only two falling in the top 10%. The highest per capita emissions in Japan are in small FUAs with strong energy sectors. Indeed, there is relatively little variation in emissions across Japanese metropolitan areas, apart from those generated by the energy sector (Figure 3.22).

Figure 3.21. CO2 emissions per capita, OECD metropolitan areas

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Figure 3.22. Major sources of per capita CO2 emissions in Japanese metropolitan areas, 2008

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

…but important environmental challenges remain

On two other environmental indicators, Japanese metropolitan areas perform rather less well compared to OECD peers.

  • Particulate matter. As noted in Chapter 1, fine particulate matter (PM) is in some ways the air pollutant that poses the most immediate and direct threat to human health and it is much more local in impact than carbon, which is in essence a global externality. Most of Japan’s major cities fall in the mid-to-upper range of the distribution for exposure to PM2.5 (Figure 3.23), though there is a good deal of variance among Japanese metropolitan areas on this indicator. All Japanese metropolitan areas – like the vast majority of OECD metropolitan areas – score above the WHO standard for average exposure to PM2.5 (10 μg/m3 annual mean). In part, this is likely to be related to urban density, since PM exposure tends to be positively correlated with population and traffic density (Hixson et al., 2012; Brezzi and Sanchez-Serra, 2014; Demographia, 2015).

  • Green space. Green areas such as parks and natural vegetation contribute to reducing pollution, improving health and quality of life, and making metropolitan areas more attractive. Not surprisingly, given the density of Japanese cities, the country’s metropolitan areas offer far less green space per capita than most OECD metros (Figure 3.24). While the numbers derived from satellite-based measures of land cover do not reflect the quality or placement of green space, such low provision is a concern. Indeed, Nagoya, Fukuoka, Tokyo and Anjō are among the 22 OECD metropolitan areas with estimated green areas below the minimum level of 9 m2 recommended by the WHO.

Figure 3.23. Exposure to particulate matter, OECD metropolitan areas

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

While the Japanese authorities are keen to ensure that population decline does not lead to the hollowing out of the cities, it does represent an opportunity to rectify the problem of lack of green space. The key will be careful planning of where and how much green space to restore. In a recent cross-country study of urban land use, Stott et al. (2015) conclude that, when it comes to urban parks, size matters: high-density neighbourhoods seem to generate better environmental outcomes than low-density areas characterised by small residential gardens and limited public space. On the basis of nine case studies across urbanisation gradients, the authors argue that “land sparing” – intensive and relatively compact urban development alongside separate, large, contiguous green space – is better at sustaining a majority of ecosystem services, including carbon storage (both above and below ground), water infiltration, pollination, pest control, noise reduction, air purification and temperature regulation. However, they also find that some “land sharing” – lower-density development, where built land and natural space are interspersed – may also be necessary to ensure that people benefit from urban green space: the positive effects of nearby green space on physical and mental health make an excellent case for incorporating smaller parks too. Indeed, human health was a category where land sparing was found to perform less well. A certain balance will thus be needed in managing development intensity across Japanese cities.

Figure 3.24. Availability of green space in OECD metropolitan areas

Source: OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

Efforts are under way at multiple levels to make Japanese cities greener…

The government continues to promote the development of greener cities in conjunction with the concentration of urban functions and improvements to public transport and the efficiency of urban energy systems, as well as countermeasures against the heat island phenomenon. In keeping with the place-based character of many urban greening policies, which must be designed so as to reflect local conditions, the 2012 Low Carbon City Promotion Act requires local governments to develop Low-Carbon Development Plans and to facilitate actions to integrate various urban functions, to promote the use of public transportation, to enhance efficiency of energy use, to preserve urban green areas, and to promote urban greening. The Act offers tax breaks for certified energy efficient buildings (exceeding an energy efficiency standard, with other low-carbon measures, water saving measures, use of sustainable building materials, and heat island prevention measures). The Urban Green Space Conservation Act also addresses such issues as the improvement of urban parks, greening of roads and public facilities such as ports, and securing of private green space using green facility development plans based on the Urban Green Space Conservation Act.

Tokyo and other major cities have also been active on their own in addressing local and global environmental concerns (Okata and Murayama, 2011). Since 2000, Tokyo, for example, has been taking measures to mitigate the impacts of the urban heat island effect, including covering roofs and walls with greenery in order to lower the surface temperature of buildings. To further these efforts, the Tokyo Metropolitan Government (TMG) passed a Nature Conservation Ordinance in 2001, requiring the greening of building roofs and walls in addition to ground-level greenings for all new construction, as well as existing buildings undergoing renovations or being extended to an area larger than 1 000 m2 for private facilities or 250 m2 for public facilities. Plans must be also submitted to include rooftop greenery for new construction with a total floor area exceeding 10 000 m2. Tokyo’s goal is to install 1 000 ha (10 million m2) of greenery by 2016, both on rooftops and at the ground level. By 2015, TMG reported that about 180 green ha had been added on rooftops alone under the ordinance. Tokyo is also piloting more specifications of the type of greening required, looking at cooling and other biodiversity benefits of various species, and looking at additional ways to put large scale, long-term urban heat island mitigation projects in place.

In 2005, TMG introduced an emissions-reduction programme that required large-scale factories, offices, commercial facilities and public institutions to develop five-year greenhouse gas reduction plans. In addition, Tokyo launched a carbon cap-and-trade system in 2010 and it co-leads the C40 Private Building Efficiency Network, a group of cities committed not only to tackling energy efficiency in their own buildings, but also to collaborating with others so they can take faster action, and have more impact.

Many other Japanese cities overcame serious environmental degradation in the 1960-70s, and developed advanced environmental technologies. The city of Kitakyūshū is among the pioneers. Once a heavily polluted industrial zone, Kitakyūshū is now a modern industrial city pursuing green growth. Under sustained pressure from its citizens, the city made a concerted effort to reduce industrial pollution and clean up accumulated environmental degradation, dramatically improving water and air quality. Remarkably, this transformation was achieved even as industrial output increased. The city’s recent green growth initiatives include an “Eco-Town” recycling cluster and ongoing investments in green city demonstration projects, such as the “smart community” trial in the Higashida area (OECD, 2013a). In 1996, the city of Kyoto launched a biodiesel fuel project that collects used cooking oil from the residents for making fuel. The city’s 92 buses and 136 garbage collection trucks now run on this biodiesel fuel. The city also boasts the lowest amount of garbage per person among Japan’s 20 largest cities, with 445g per person per day in 2013. This was down 42% from the peak year (2000). The number of garbage incineration facilities in operation was cut from 5 to 3 and the annual cost of waste collection fell by 42.5%, to JPY 5.4 billion (City of Kyoto, 2015). Both cities have also implemented international city-to-city co-operation for sustainable development in Asia and have built reputations among cities aiming for green growth.

…and more resilient

In many cases, efforts to make Japanese cities greener go hand in hand with initiatives to strengthen urban resilience. The government is anxious to prepare cities for major disasters, such as a major inland earthquake near Tokyo or an earthquake in the Great Nankai Trough, both of which are regarded as very high risks in the coming decades. Projections suggest that a major quake around Tokyo could leave 23 000 dead and cost the economy up to JPY 95 trillion (Cabinet Office, 2015b). A large quake in the Nankai Trough, where the risk of Tsunami is very high, could be far worse, killing as many as 320 000. To this end, the government has set about:

  • making improvements in urban blocks where houses and other buildings are densely concentrated and therefore at high risk of fires and building collapse in the event of a quake

  • securing the refuge sites and evacuation routes needed for quick and smooth evacuation from tsunamis

  • taking measures to support people stranded by disaster, e.g. those in major railway stations unable to return home in the immediate aftermath of a disaster

  • securing urban parks, which may serve as disaster management centres and refuge sites.

Local authorities, too, are part of this effort, engaging in training, disaster drills and the storage of provisions. Governments at all levels are also working to inform the public better about disaster risks and responses. Prior to 2011, many municipalities were reluctant to publish hazard maps, for fear of the impact on property prices. That is now changing.

Promoting earthquake-resistant buildings is among the most important urban resilience strategies. In Japan, structures built before 1981, when stricter building codes were introduced, may lack adequate earthquake-resistant structure. The government thus promotes: i) assessment of earthquake-resistant structure of buildings which were built before 1981, and ii) retrofitting the buildings whose earthquake-resistant level is assessed insufficient. Structures built after 1981 are considered as earthquake-resistant. The Basic Plan for the Tokyo Inland Earthquake, adopted in March 2015, sets a number of nation-wide targets, including that of making 95% of Japan’s housing stock earthquake-resistant by 2020 (Table 3.2). The Basic Plan stipulates that Tokyo could continue to function and the damage would be significantly reduced by preparedness for the disaster and by emergency response plans. It is crucial that such preparatory measures are implemented.

Table 3.2. Targets for earthquake-resistant buildings


Public schools (Japan)

Police buildings

Number of building stock

60 628 600 (2013)

118 504 (2015)

1 815 (Japan, 2013)

Current ratio



84% (average of Tokyo, Kanagawa, Chiba and Saitama)




FY 2014






By 2020

Earliest possible by FY 2015

FY 2018 (average of Tokyo, Kanagawa, Chiba and Saitama)

Sources: Cabinet Office (2015a), “Basic Plan for Urgent Implementation of Measures for the Tokyo Inland Earthquake”, available (in Japanese) at:; Fire and Disaster Management Agency (2015), “Result of the Survey on Earthquake-Resistant Public Facilities for Disaster Management”, press release, 18 February 2015, available (in Japanese) at:; Official Statistics of Japan (2015), 2013 Housing and Land Survey, (accessed 15 July 2015); Ministry of Education, Culture, Sports, Science and Technology (2015), “Result of the Survey on Earthquake-Resistant Renovation of Public School Buildings”, press release, Ministry of Education, Culture, Sports, Science and Technology, 2 June 2015, available (in Japanese) at:

The global competitiveness of Japan’s metropolitan areas

At an international level, the largest cities increasingly compete with one another

In the age of globalisation, it is no longer possible to think of cities or urban systems in a purely national context. The emergence of global value chains has entailed the development of new forms of financial and producer services to manage them, in many cases highly complex and skill-intensive services. Increasingly, therefore, the largest cities in many developed countries have increasingly specialised according to function rather than sector, developing as important nodes in these global networks rather than as producers of a certain range of goods, such as cars or electronics (Duranton and Puga, 2002). Since many of these activities benefit from cluster economies and agglomeration dynamics, they have tended to concentrate in a limited number of cities – those defined by Sassen (2001, 2005) as “global cities”. Strikingly, despite their broadly similar specialisations, these cities do not necessarily compete directly with one another: there are competitive dynamics, to be sure, but also complementarities insofar as they form part of a network of cities offering the specialised financial and producer services that make the globalised economy run (Renn, 2012).

That said, it is important to stress that many other cities around the world have benefited spectacularly from globalisation. Cities need not enter the ranks of the major global centres in order to offer prosperity and opportunity to citizens. For one thing, cities do not exist in isolation: the existence of urban hierarchies and divisions of labour within countries has long been studied – certain functions tend to be concentrated in specific cities, and there are observable patterns with respect to city size and specialisation (Duranton and Puga, 2002). In a sense, the emergence of the “global city” reflects the degree to which globalisation has given rise to a global urban system. The cities dominating the “global city ratings” are those that concentrate some of the most globalised functions of the system, but there are other niches to fill. Not all cities can or will specialise in such activities – nor should they. That does not mean they are not deeply integrated into the world economy: the manufacturing centres of China, for example, have been built on exports and are deeply embedded in global production chains.

For Japan, then, the question of global competitiveness therefore needs to be addressed on two levels: the first concerns the continued role of Tokyo – or, in future, a Tokyo-Nagoya-Osaka urban mega-region – as one of the most important global cities. The second concerns the competitiveness and integration into the world economy of the other Japanese metropolitan areas, and in particular the potential for gaining greater benefit from their location in what is, and is expected to remain for some time, the most economically dynamic region on the planet.

What makes “global cities” competitive and what does this mean for Tokyo?

There is a paradox to consider when confronting the question of what makes a global city successful. While the methods employed by different institutions and researchers vary, global city rankings converge to a surprising extent. Leff and Petersen (2015) present an overview of the top-ten lists emerging from nine internationally recognised city ranking exercises.10 A small number of cities in developed countries dominate the rankings,11 despite the fact that developing countries in recent years have exhibited far more economic dynamism and are now home to many of the largest and fastest-growing cities on the planet. Beijing makes the top ten on two ratings, with Dubai and Shanghai each listed once, but emerging Asia, the Middle East, the former Soviet Union, Latin America and Africa are all otherwise absent. While Tokyo is among the most highly rated cities, appearing on five of the major rankings, it is the only Japanese city represented. By contrast, eight US cities are in the rankings, and a number of other OECD countries have two or three.

This might simply imply that the ratings are backward-looking and fail to reflect recent changes in the world economy. Alternatively, it might be a result of bias, since most of the ratings are produced in North America or Western Europe.12 A closer look, however, suggests that there is more at work:

  • Size is not all that matters. While mega-cities like London, New York and Tokyo turn up again and again, some, like Stockholm and Zurich, are relatively small by global standards, and many of the world’s largest cities are absent.

  • The various ratings are almost all dominated by cities that combine size, strong international connectivity and specialisation in high-value business services like finance, trade and logistics, innovation and knowledge creation. These are all highly globalised activities, and yet many of them have strong place-based characteristics, either because they benefit from cluster effects and/or because they are fields in which global standards and firms dominate and yet must function in a wide range of institutional and legal settings (e.g. finance and accounting).

  • To attract such services, cities require not only the “hard” infrastructure and institutions needed to support them but also a certain quality of life. This is because success in these spheres requires attracting highly skilled (and highly mobile) people. Urban amenities, environmental quality and other non-pecuniary dimensions of well-being are thus important components of their competitiveness – far more important than they would be, for example, in a mining “boom town”.

The foregoing suggests that the developed countries may retain a comparative advantage in high-order urban functions. The developed countries’ quality of infrastructure, human capital and institutions thus gives them an edge in the core “global city functions” of commerce, finance and knowledge creation. So does the quality of life they offer. Well-functioning, liveable cities are not easy to build, and they are even harder to manage when cities are growing very fast, as they are in many developing countries. Emerging-market cities often struggle to compete in this arena, as city growth often continues to outstrip the ability of local or national authorities to provide infrastructure and services or to manage such challenges as congestion, pollution or waste disposal. As a result, the developed countries retain an advantage in providing cities that offer a high quality of life and good environmental quality to residents. It worth noting, too, that they tend to be highly internationalised places, in which it is relatively easy to settle and enter high-skilled employment without being fluent in the local language.

For Japan, this points to a number of priorities for the urban mega-region at the heart of Honshū:

  • National-level legal, regulatory and fiscal regimes must be conducive to the attraction and retention of global players in fields like logistics, finance and knowledge-creation.

  • The gateway role of global cities requires that external connectivity be strong.

  • Liveability matters, so steps to ensure adequate internal connectivity, as well as environmental quality, urban cultural and recreational amenities, etc. are also critical.

  • The international character of the city matters: it must be attractive to high-skilled expatriates from outside Japan and, given falling labour supply, it is likely to require immigrant labour even in less-skilled occupations.

In 2013, more than one-third of London’s population was foreign-born, according to the UK Labour Force Survey, and the list of metropolitan areas with the largest foreign-born population shares overlaps significantly with the major global cities rankings – along with London, it includes Hong Kong, China; Los Angeles, New York, Paris, Singapore, Sydney and Toronto, among others. This suggests that resistance to immigration, population decline and population ageing must be added to the list of issues Japan’s major metropolitan areas, particularly Tokyo, must address in order to sustain their competitive positions in Asia.

A number of these competitiveness priorities are being addressed in the creation of the government’s new “National Strategic Special Zones” (NSSZs), particularly the zone organised in Tokyo, which is discussed below.13 One might also add the risk of major earthquakes and tsunamis to the list of competitiveness challenges – a number of cities are at high risk of a major quake and/or tsunami in the coming decades, and both national and local authorities are actively preparing for this.14 However, while such risks pose major issues for policy, they may matter less in the competition for people and capital in Asia, simply because so many of Asia’s most dynamic cities face similar hazards.

The future of Tokyo

Greater Tokyo occupies a somewhat paradoxical position with respect to Japan’s demographic and economic challenges. On the one hand, it is on many measures the world’s largest metropolitan area and the country’s main driver of growth. It accounts for almost one-third of GDP and is the country’s capital, main financial centre and principal international gateway. Its dynamism and global competitiveness are central to any hope of reviving Japan’s productivity and growth performance. At the same time, in a context of population decline and ageing, Tokyo’s dynamism can seem threatening to other parts of the country, as it acts as a magnet for money and talent. Such concerns underlie much of the emphasis on promoting regional revitalisation. Yet regional revival cannot come at the expense of Tokyo: Japan needs its existing growth engine to grow stronger even as it seeks to ignite other centres of growth and employment creation around the country. Tokyo’s position in the global competition to attract financial and human resources and companies from around the world is critical to Japan’s prosperity.

Tokyo’s demographic problems are unique in Japan

Lützeler (2011) observes that the Tokyo Metropolitan Area (TMA) is about the only place in Japan not yet worried about population decline – projections show the population of Tokyo Prefecture rising to 2020 and then gently declining to a level in 2035 that would still be above that of 2005 (12.5 million). The rest of the agglomeration is projected to experience a sharper decline but should still retain population better than the national average. That said, there are many areas of the TMA that are already experiencing rapid demographic decline amidst continuing TMA growth. Growing areas include waterfront areas (Chūō, Kōtō and Minato wards), residential areas within the other wards and some suburban residential areas. Okata and Murayama (2011) report that many outer suburbs have been experiencing population decline since the turn of the century. Of course, population decline is not a simple matter to interpret in central parts of the agglomeration. At the neighbourhood level, demographic decline can reflect either economic vitality (residential use changes to commercial use) or economic decline (rising vacancies, departing populations). On the periphery, however, it is far more likely to reflect decline.

Tokyo’s population is also ageing less rapidly than the average but the growing population of elderly – and especially of the elderly poor – represents a challenge. Old-age dependency ratios in Saitama and Chiba prefectures are likely to approach the national average by 2030, but Tokyo and Kanagawa are projected to be 3-4 percentage points lower. Ageing in Tokyo does, however, tend to be concentrated in specific places. The fastest-ageing areas are of two types: inner area districts often adjacent to railway lines or other urban disamenities (noise, etc.); and (far more numerous) neighbourhoods dominated by public rental housing,15 often with high unemployment and poverty rates (Lützeler, 2011). The former tend to be quite central and so may be redeveloped in time, to draw back the young, or they may be transitioned to non-residential use. This pattern is consistent with Sassen’s theory of “global cities”, which anticipates growing gaps between city centres, where the new service class elite concentrates, and the rest of the agglomeration (Sassen, 2001:260-2).16

In some ways, though, the more serious concern for the authorities is the impact of Tokyo on fertility. In terms of economic opportunity, Tokyo has enormous attractive power, and in recent years. The government estimates that in 2014, the net inflow of people into Tokyo was over 109 000, a figure that is not out of line with trends since 2000, apart from a sharp downturn for a few years immediately following the onset of the global crisis in 2008.17 The great bulk of new arrivals are young – more than 90% were in the 15-29 age bracket in 2014. Young Japanese come to the capital to study or find work and then stay. The problem for Japanese policy makers is that Tokyo Prefecture also has, by a fair margin, the lowest fertility rates in the country (1.13 in 2013), and the other prefectures around it, such as Saitama, Chiba and Kanagawa were also far below the national average, at 1.31-1.33 in 2013. This problem is present, albeit in less acute form, in the other major conurbations: the prefectures at the heart of the large metropolitan area centred on Osaka also had fertility rates of 1.26-1.32 in 2013.

High living costs and the difficulties in balancing parenthood and career together mean that young people in Tokyo and other large cities work more, marry later and have fewer children. While some officials see redirecting migration flows away from Tokyo as critical to raising the birth rate, there is no evidence that this will occur on any scale that would make a difference nationally. Measures to make Tokyo more “child-friendly” – particularly as regards childcare provision and affordable housing – are therefore critical to addressing Japan’s demographic challenges.

Policies for Tokyo focus on connective infrastructure and redevelopment of the urban fabric

The City Bureau of the Ministry of Land, Infrastructure, Transport and Tourism views connectivity and the urban redevelopment of Tokyo as the main priorities when trying to increase the global competitiveness of the capital, emphasising the need to improve the environment for doing international business, including external access points and the number of international convention centres and other facilities. The Tokyo Metropolitan Government likewise emphasises transport infrastructure. Both point to the need for better connections from the two airports, Haneda and Narita, to the city centre and to one another (via the development of a new railway line between them), as well as an increase in airport capacity, which is expected to reach saturation level by 2020. The city authorities would also like to see the construction of a new port outside the Rainbow Bridge. The bridge is 52m high, which means that big cruise ships cannot enter the port. In addition to these priorities, the TMG lays great stress on internal connectivity, seeing the city centre’s increasingly acute congestion as a primary handicap. The average traffic speed in Tokyo is around 16 km/h (as against a national average of 35). Three major ring roads are under construction.18 These are meant to improve accessibility to airports and seaports and to ease chronic congestion traffic in central Tokyo. The TMG estimates that 60% of the traffic inside Tokyo’s inner loop is through traffic, so the construction of second and third rings could allow more traffic to bypass the centre. This will require central government support, as ring roads and national airports are national priorities – though that does not mean that they will be wholly funded from the national budget. Usually, the government, the TMG and Tokyo’s wards would share the costs.

The emphasis on internal connectivity is a priority that Tokyo shares with many other global metropolitan areas, including Paris, where the French authorities have recently launched a 15-year project to strengthen internal transport connectivity and institutional cohesion within the metropolitan area (Box 3.4). The aim is not only to improve quality of life by reducing pollution and facilitating mobility but also to stimulate new agglomeration dynamics, innovation and the emergence of new centres of economic activity in the Greater Paris metropolitan area. This, in turn, should make the specific assets of the Paris region more visible at metropolitan scale. In Japan, plans for the development of Tokyo and for the creation of an urban mega-region encompassing the Tokyo, Osaka and Nagoya metropolitan areas are intended to achieve many of the same objectives.

Box 3.4. France’s Grand Paris project

The Grand Paris project in France reflects a new approach in spatial planning in France. It places great emphasis on the Paris region as a global asset for the country rather than viewing it as a danger for the development of other regions and cities. This change can be explained in part by the rapid development of other important cities such as Bordeaux, Lyon or Toulouse, which have eased fears concerning the concentration of people and activity in and around the capital, but it also reflects the awareness that countries’ economic performance relies in large measure on the performance of world-class cities such as Paris, London, New York or Tokyo. It is all the more so important in the French case that Paris is tightly connected, via both transport networks and economic ties, to the other major French cities, enabling important spillover effects all across the country (see also Box 3.6 on the French experience with high-speed rail).

The origin of this project lies in the widespread perception that the Paris region does not live up to its economic potential because of the low visibility of its main assets (such as the diversity of its economic activities or the high concentration of researchers) and an under-optimal agglomeration effect that limits efficient matching on the labour market and cross-fertilisation of ideas and activities. Such difficulties are reinforced by institutional fragmentation and a star-shaped transport system that works well inside the city of Paris itself (intra muros) and connects the surrounding communities to Paris, while hindering mobility between cities and towns that surround Paris itself. This explains why the Grand Paris is a two-level project, including spatial planning and institutional dimensions.

The most visible part of the spatial planning dimension consists in the construction of new metro networks (comprising both new lines and extensions to some existing ones) to enable easier and faster connections around Paris. The aim of this network is not only to make the transport system more efficient and coherent with the reality of the urban area and the needs of the inhabitants. It is also conceived as an incentive for renewed urban planning and economic development projects around the new stations. These stations – for which a particular emphasis has been laid on architectural quality and the amenities offered (shops and common services but also co-working places, business incubators, etc.) – are envisaged as a way to facilitate the emergence of new centres of activity capable of reinforcing the general dynamics of the agglomeration. Moreover, this network will carry not only people but also data with high-speed connections and data centres in order to reinforce potential development and strengthen the attractions of the Paris region. Finally, this transport network is also meant to encourage the development of economic clusters (e.g. creative industries around Saint-Denis, aerospace in Le Bourget, health and medical research around Villejuif, etc.) or academic concentration (e.g. in Saclay) around these stations, in order to reinforce cross-fertilisation and critical mass, while making the assets and specificities of the Paris region more visible at an international scale.

The aim, in short, is not only to help mobility but also to reduce discrepancies in economic development inside a metropolitan area that combines extremes of both wealth and poverty, by creating new development centres and enabling people to find jobs more easily. It is also meant to curb urban sprawl and reduce the use of cars, making the development of the Paris region more sustainable. After a period of preliminary studies, the construction of the new lines, supervised by the Société du Grand Paris started in 2015 and should be completed by 2030.

The institutional dimension mainly consists in the creation of a new Métropole du Grand Paris, a metropolitan governance body that will come into being in January 2016. It will bring together Paris and 130 surrounding cities. The new entity will be in charge of economic development, town planning, environmental policies and, to a certain extent, housing. The creation of this métropole is part of a global policy set out in new legislation adopted in 2014, which has created 14 métropoles in France. This process of achieving metropolitan governance has been especially difficult in the case of Paris, because of its global size and such local specificities as high wealth differences among cities and a long tradition of political and institutional fragmentation. This is why federative projects, such as the 2024 Olympic Games or 2025 universal exhibition candidacies, though not directly linked to the Grand Paris project, are also regarded as powerful drivers to move collectively along in that direction, as well as a booster for public works.

Source: Information provided directly by the Commissariat Général à l’Égalité des Territoires.

In addition to the above, Tokyo is at the centre of two megaprojects which will be discussed below:

  • In 2020, Tokyo will host the Summer Games of the XXXII Olympiad. The authorities’ aim is to use the Games not only to raise the city’s international profile and enhance its image but also to put in place a set of infrastructure and other investments that will support its future competitiveness.

  • In May 2011, the government authorised the construction of a JPY 9 trillion magnetic levitation (maglev) train, the Chūō Shinkansen, which will ultimately connect Tokyo, Nagoya, and Osaka. When completed, it will cut the journey time between Tokyo and Nagoya from the current 95-100 minutes to just 40, and the time between Tokyo and Osaka from around 140 minutes to 67. Its maximum running speed will be about 505 km/hour.19

The City Bureau has also been actively supporting urban redevelopment with a view to enhancing the investment attractiveness of Japan’s major cities. The 2002 Urban Renaissance Law was one of the first comprehensive efforts to enhance Japanese cities’ competitiveness, by stimulating private investment in urban development. Deregulation of land use controls and financial incentives for private investors were the main instruments. A total of 2 806 ha in Tokyo/Yokohama was designated for urgent and focused government support in private urban development projects. Urban renewal was accelerated in downtown Tokyo and the waterfront area. Between 1990 and 2010, the total floor area in large offices increased by 82%, from 20.5 million to 37.2 million m2 (Ministry of Land, Infrastructure, Transport and Tourism, 2015). The land and rent value increased in most of the designated districts relative to other areas in these cities. Urban Renaissance initiatives were also undertaken with City Bureau support in such cities as Fukuoka, Nagoya, Osaka, and Sapporo. By August 2015, 79 such urban renaissance projects had been registered with the Ministry of Land, Infrastructure, Transport and Tourism.

Flagship projects in Tokyo have included developments along the waterfront, around Shinjuku, Shibuya and Shinagawa stations, the Maronouchi Park Building and Toranomon Hills. Altogether, the floor area in large office buildings rose by 16.7 m metres2 (+82%) between 1990 and 2010.20 Support for such projects includes government guarantees to facilitate borrowing, accelerated amortisation, the relaxation of floor-area ratios and height restrictions and a provision for lower tax rates on fixed assets in cases where buildings are built with public spaces. Partial exemption from the real estate acquisition tax is also possible, as are reduced rates for registration and licence taxes. According to City Bureau figures, around JPY 25.8 billion in guarantees and other assistance with financing was extended to end-October 2013, as well as JPY 27.5 billion in assistance via tax breaks.

Regulatory and other policies are also changing but more needs to be done

It is striking that the priorities of both the TMG and the Ministry are so strongly focused on the new construction, mainly of infrastructure but also for the Olympics. There is certainly a need to address congestion problems and scope to improve the connections to and from Tokyo’s major airports. These are both weak points identified in international assessments, such as the Mori Memorial Foundation’s Global Power Cities Index. However, the emphasis on infrastructure solutions is open to question given the stress on public finances, population ageing and the fact that Greater Tokyo’s population is expected to begin declining. Demand-side solutions, such as road-pricing or congestion charges, might merit consideration, given the risk that there will be insufficient demand to justify some of the new infrastructure in coming decades. A balance of supply- and demand-side solutions is needed. In any case, international assessments identify a number of other weaknesses Tokyo needs to address, including regulatory burdens, English proficiency, cost of living and the natural environment.21 Complex and lengthy procedures to start a business remain one of Japan’s competitive weaknesses. According to the World Bank’s “doing business” indicator, it takes an average of 11 days to set up a business in Tokyo, compared with only 3 days in Hong Kong or Singapore. Physical urban development projects will be far more effective, if they are linked to promote start-ups and attract foreign investment, to facilitate innovation, or to attract skilled and talented workforce.

In this connection, the National Strategic Special Zones scheme launched in 2014 is critical for Tokyo, as it aims to create “the world’s most business-friendly environment”, and also for other major Japanese cities. There are a range of menus for deregulation, including one-stop services for legal procedures for business start-ups and visa conditions for foreigners wishing to run businesses. In December 2014, Greater Tokyo (Chiyoda, Chuo, Minato, Shinjuku, Bunkyo, Koto, Shinagawa, Ota and Shibuya wards, Kanagawa Prefecture and Narita City) was designated as the first NSSZ. As of July 2015, there are nine NSSZs approved by the Cabinet Office. The Tokyo NSSZ plan includes various deregulation measures, including support for business start-ups. In April 2015, the Tokyo Metropolitan Government and four ministries opened a “Tokyo One-Stop Business Establishment Centre”. By making the office a one-stop window, it expects to speed up the process for starting companies. A “Centre for Employment and Labour” will also be established by the central government to support global enterprises and start-ups by providing expert consultation on employment conditions and labour regulations in Japan. The TMG already operates a “Business Development Centre”, as a part of “the Special Zone for Asian Headquarters”, another national deregulation initiative started in 2011. Such existing initiatives will be linked and co-ordinated with the new NSSZ measure to generate synergetic effects.

The latest NSSZ plan for Tokyo, approved in June 2015, includes the following provisions which aim to stimulate innovation and attract high-skilled workforce from all over the world:

  • Regulation of the use of medical drugs/methods that are not yet approved in Japan is to be relaxed at six medical facilities in order to provide high-level service more promptly.

  • Regulation of the number of beds is relaxed at six medical facilities in order to provide high-level medical service.

  • In order to attract doctors with high medical skills from abroad, foreign doctors are allowed to work in three medical facilities without obtaining a Japanese licence, based on bilateral agreements with France, the United Kingdom and the United States.

  • Simplified planning procedures are applied to five urban development projects that concern facilities for international business promotion.

  • Regulation of streets is eased at four locations in the zone in order to enhance the attractiveness of the zone and promote tourism.

It will also be crucial for the Tokyo NSSZ to keep revising its deregulation menus in response to the needs of the business community. Ambitious reforms will be needed to make the new scheme more successful than previous ones. For example, the Special Zone for Asian Headquarters launched in 2011 in an effort to make Japan a centre for international business aimed to attract more than 50 corporate headquarters for Asia by 2016. As of early 2015, it had attracted only two (OECD, 2015f). The central government also has an important role in examining the effectiveness of the experimental deregulation and expanding it to other areas of Tokyo and major cities.

In addition to all these efforts, long-term social and institutional investment is indispensable if Tokyo is to be more competitive in an increasingly globalised economy. Tokyo could help skilled foreign workers to find firms in Japan in search of their skills. In Victoria, Australia, the Overseas Qualified Professionals Programme (OQPP) provides recently arrived professionals who acquired their skills abroad with a work-experience placement to enhance their opportunities for employment in their field of study (Quintini, 2011). Many OECD countries and regions have implemented specific measures to address the needs of and demand for high-skilled immigrants. Transforming Tokyo into a foreigner-friendly city not only for visitors but also foreign residents is an essential element of such an effort. Newly arrived foreign workers and their families face various challenges, such as finding housing, education for their children, hospitals for medical service, cultural activities, etc.

A key step is to provide support to newly arrived workers and their families at the community level. For example, multi-language support should be expanded for newly arrived workers in order to provide them with information on education and medical services in the community. Language programmes to support integration are also important. In Sweden, a non-profit organisation called Folkuniversitetet offers a wide range of adult education courses throughout Sweden, and it is especially committed to strengthening the position of immigrants on the labour market. For instance, a special Swedish language programme for immigrant medical and healthcare staff focuses on enabling them to use their professional skills in Sweden (OECD, 2012b).

The maglev project and the formation of a Tokyo-Osaka-Nagoya mega-region

The Chūō Shinkansen is to create an urban mega-region of more than 60 million people

The Chūō Shinkansen maglev project now under way is set to reshape the future of the great urban cluster on the Pacific side of Honshū. The magnetic levitation trains will float about 10 cm above a U-shaped guideway, held aloft and propelled forward by superconducting magnets. This allows for frictionless movement. It should also result in lower greenhouse gas emissions than most other transport. With no moving parts, the maglev has low maintenance costs, and some maglev designs, including JR Tōkai’s, are self-stabilising, which reduces the risk of crashes. The technology has already been used to build short demonstration lines, but Tokyo-Nagoya will be the first functioning intercity route.22 In the first phase, the line will run from Tokyo’s Shinagawa Station to Nagoya, stations in Sagamihara, Kanagawa, Kofu, Yamanashi, Iida, Nagano, Nakatsugawa and Gifu. Operations should begin in 2027. According to documents released by the Central Japan Railway Company (JR Tōkai), the line’s route will not go through water resources, natural parks or communities, and waste soil will be used for farmland or public works projects (Minami and Yagi, 2013). Over 85% of the 286 km line between Tokyo and Nagoya is to be built underground or through tunnels.

JR Tōkai is to fund the project entirely through cash generated by the operation of its current bullet train lines. Under the law on the development of the Shinkansen lines, JR Tōkai would be able to rely on government subsidies, but because Shinkansen lines to be constructed in other parts of Japan require government support, the company plans to finance the Chūō Shinkansen itself. However, the government and the affected municipalities are making significant contributions in other ways. While budgetary funds are not being spent on the project, the government passed legislation stipulating that tunnels more than 40 metres underground (the planned depth of the maglev lines) do not need to compensate landowners (Lippert, 2013). This will greatly reduce costs in building the line through some of the most expensive real estate in the world – indeed, the value of land in the area is a major reason for JR Tōkai’s decision to build as much as possible underground. The stations along the maglev line are to be financed by the municipalities where they are located.

JR Tōkai acknowledges that the project will break even only after decades of operation, and much of its commercial benefit will depend on Japan’s success in exporting its maglev technology.23 However, the company argues that it needs an alternative route to current Shinkansen line, which will have to be renovated extensively by mid-century, and which is in any case close to saturation – schedules are so tight that there is little room to add more trains. In addition, JR Tōkai insists the additional route will also enhance the resilience of the transport network in the event of a major earthquake. In 2013, a subcommittee of the Land, Infrastructure, Transport and Tourism Ministry estimated the economic effect of the Linear Chūō Shinkansen at JPY 870 billion a year once the second phase is completed. About 45% of this is expected to accrue to Greater Tokyo, with a slightly larger share to the Nagoya and Osaka metropolitan areas and just under 7% to the rest of the country (Minami and Yagi, 2013). The spread effect of this development across space will depend to a great extent on how well other transport investments are aligned with it. In Nagoya, for example, the authorities are already planning not only to co-ordinate timetables but also to reconfigure the central station so as to facilitate fast, easy transfers to and from regional services.

The authorities’ expectation is that construction of the line will contribute to the creation of a globally competitive urban mega-region, generating powerful agglomeration economies and attracting investment as a result of its massive market potential. With a population of 60 million people, the region will boast 4 international airports and 2 international container hubs, which should make it one of the world’s largest trading regions. In an early analysis of the proposed maglev, Morichi and Shimuzu (2002) argued on the basis of the historical development of earlier Shinkansen lines that the 60-minute travel radius around Tokyo was a crucial threshold. As faster transport expanded this circle around the capital, the regions drawn into Tokyo’s “orbit” experienced faster growth. They anticipate that the Chūō Shinkansen will have a similar effect.

The government also expects that competitive synergies will emerge as the international functions of the Tokyo area are linked more closely to the manufacturing excellence of the Nagoya metropolitan area, in addition to the cultural, historical and commercial functions of the Osaka metropolitan area. This ambition is not unique: other countries, too, have in recent years grown interested in the potential for transport investments to bring people and businesses closer together, in terms of travel times if not distances, so as to create larger agglomerations of economic activity (Gibbons, 2015). For example, Overman et al. (2009) find that a 20-minute reduction in rail journey times between Manchester and Leeds would yield benefits to Manchester alone of around GBP 4.3 billion. This, moreover, is without factoring in the possibility that a more prosperous Manchester-Leeds agglomeration would attract more high-skilled workers.

This last point, of course, underscores one of the distributional consequences of agglomeration-oriented policies: the attraction of more highly skilled individuals to the big city would be likely, other things being equal, to enhance their productivity and thus improve national performance (OECD, 2015d), but it would come at the expense of other places in the country. The issue of displacement as opposed to new activity is one that needs consideration in the assessment of any proposed infrastructure investment. It is particularly relevant in respect of the Chūō Shinkansen, given the authorities’ concern about over-concentration in Tokyo. If avoiding over-concentration is a policy priority, then it is hard not to question the expedience of a project that is highly likely to strengthen centripetal forces in the labour market. The demographic situation also raises questions about the need for the line: with Japan’s population falling rapidly and even Tokyo’s population expected to begin declining within a decade, it is not clear that there will be sufficient demand for the service.

Ex-ante assessment of the impact of infrastructure mega-projects is notoriously difficult

The degree to which transport infrastructure can serve as a driver of growth remains open to question. OECD work suggests that investment in network infrastructure does indeed have a positive effect over and above the addition to the capital stock but that the effect is non-linear and tends to be stronger at lower levels of provision (Sutherland et al., 2009). This is precisely what one would expect if such investment were subject to diminishing returns (Demetriades and Mamuneas, 2000).24 Where infrastructure gaps are large, the returns to filling them are, too, but when infrastructure is well-supplied, the marginal benefits of additional investment tend to be low or even non-existent (OECD, 2013b). Much depends on how infrastructure investments interact with other elements of the policy package: OECD (2009b) finds that the impact of infrastructure provision on regional performance depends to a significant extent on policies relating to human capital development, innovation and the business environment. On its own, infrastructure may deliver little but in tandem with other policies it can make an important contribution.

Even more difficult than estimating the impact of transport investment in aggregate is ex-ante evaluation of large individual projects. It is important to distinguish here between the methods appropriate for routine infrastructure investments and those required for mega-projects that are conceived to be “game-changers”, involving long-term dynamic feedback that may affect economic structure and competitiveness (OECD, 2014a). For example, fairly standard cost-benefit analysis methods may provide a basis for assessing an investment that will ease a constraint on a well-travelled route, but they are ill-suited to gauging the impact of something like the Öresund Bridge that connects southern Sweden with Denmark’s capital region. The bridge altered existing flows of people, goods and services, gave rise to substantial new flows, and altered the settlement pattern substantially. Even at smaller scales, such things can be hard to predict: the addition of new commuter rail lines in Greater Paris is meant to reduce inefficiencies (direct impact) but will also change land-use patterns and possibly restructure the housing markets and local economies around the new routes. There is a growing body of evidence on the pitfalls involved in evaluating such projects (and the possible solutions), but as yet no standardised method that is accepted as best practice. Recent OECD work does, however, point to some basic steps that should be followed (Box 3.5).

Box 3.5. Ex-ante assessments of infrastructure mega-projects

The literature on assessing the economic impact of large transport infrastructure investment focuses on the most effective ways to estimate the benefits of infrastructure investment compared to other scenarios, the usual counter-factual being the lack of any investment (i.e. the status quo) or, less frequently, investment in an alternative solution. The measures used to assess the projects vary, but tend to include one or more of the following:

  • the potential for increased productivity through agglomeration economies

  • the benefits of easing transport bottlenecks

  • the benefits of greater network reliability and predictability

  • the benefits of expanding capacity for transport, housing, and job creation (and by extension individual choice)

  • the impact on human capital/skill formation.

Recent assessments have emphasised broader measures of benefits that go beyond direct benefits to users (those directly connected with or affected by the project), paying greater attention to indirect (those that supply or benefit from those connected to the project), and induced economic impact (benefits that accrue to the wider economy, up to the national scale). Some estimates seek to identify dynamic induced effects. The ratio of all cumulative effects to the direct effect is usually the multiplier effect used to estimate the overall benefits of large projects. In addition, there is an increasing effort to measure spillover effects spatially as an important factor affecting the nature and distribution of benefits.

Because of the complexity of assessing these broader impacts, OECD (2014a) points to the need for a number of essential steps.

  • The assessment needs to include sensitivity and robustness analyses, because there is less confidence in any modelling results.

  • There needs to be a clear counterfactual. In most such cases, this is just as challenging as estimating the outcomes under the investment scenario. Overman (2011), points out that, in the case of very large infrastructure projects, the baseline (not building anything) is difficult enough to estimate, but that policy makers must also consider the opportunity cost of using the public funds involved for the project rather than in some other way.

  • The assessment should start by establishing a baseline to estimate the marginal return to additional transport infrastructure given the structure of the economy. Efforts to link the impact of transport investment to agglomeration benefits face a number of obstacles in both measurement and identification. However, combining land-use and transport models can make it possible to arrive at better estimates of productivity effects.

  • When the nature of the project means that the ex-ante assessment will necessarily lack precision and confidence, it can help to draw on ex-post studies of other projects.

When it comes to high-speed rail (HSR), in particular, OECD (2014b) proposes a four-stage test. The first concerns commercial viability. The trouble is that such investments rarely appear to be commercially sustainable on their own, as the time horizons and scales involved typically necessitate some degree of public support.25 Japan is something of an exception here, in view of the success of its Shinkansen network. However, the commercial case for the maglev is finely balanced and depends on public resources to a significant extent. The social returns to the investment are therefore critical, and the other stages of the OECD test focus on them. The second stage concerns social returns that come from transport alone. This benchmark, usually assessed in terms of travel times, anticipated passenger numbers and modal shift (from dirtier to clean transport modes and/or from more to less congested ones), is often achieved. However, the threshold set by the implementing agency can vary substantially depending on its goals, so there is a real risk of bias, depending on who does the assessment. The third stage focuses on the benefits to the entire transport network, which inevitably involves much greater complexity in assessment. Finally, the fourth stage looks at the broader economic and social benefits, which are not always tangible.

Some recent work casts light on the issue of dynamic induced benefits, in particular. Chen and Haynes (2013) present a new method to estimate the macro-economic impact of large investments in public transport infrastructure, combining spatial econometrics and comprehensive general equilibrium estimation.26 Their model estimates both demand- and supply-side effects and it looks at the entire transport system – i.e. it looks at the interactions between different modes of transport. They find that the magnitude of the impact of transport infrastructure investment is typically much smaller than previously believed. In part, this may simply reflect the use of more recent data from countries with well-developed transport networks: this would be consistent with the findings of Sutherland et al. (2009). The high level of aggregation of the model (at the national level) also means that the estimates are based on broad definitions of each sector (e.g. all roads and highways), which limits its utility in assessing the benefits of a single large localised project (such as a new HSR line).

Perhaps of greater direct utility for assessing specific large-scale projects is the model presented in Australian Department of Transport (2012). It examines four major infrastructure investments ex-post and argues for a broad measure of benefits, using a dynamic general equilibrium approach that includes feedback loops and land-use outcomes, while remaining cognisant of the cost and difficulty involved in identifying costs and benefits beyond the first-round effects. The main variable the authors use is “effective job density”, a composite variable that measures accessibility based on the density of employment and the time it takes to reach employment centres.

The economic impacts of the maglev are difficult to foresee

The authorities expect that the new maglev line will drive the formation of an urban mega-region at the heart of Japan. While international experience with such mega-projects is too limited to generalise with confidence about their impact, a review of international experience suggests that the speed and extent of this integration should not be taken for granted. Other policies must be supportive of integration, particularly those that favour innovation and entrepreneurship (OECD, 2009b), and the cities involved must take steps to ensure that the regions around the new line are integrated with it (e.g. the efforts by Nagoya to ensure speedy connections). A large number of recent and not-so-recent examples of economic development strategies founded on the introduction of new transport infrastructure confirm that the effects of such investments are likely to be substantial but are difficult to predict and far from automatic (Box 3.6). They depend on the degree to which entrepreneurs see opportunities to use the new transport option to exploit opportunities previously unavailable to them – to cut transport times and costs, to penetrate new markets or to establish new links with partners.

Box 3.6. International evidence on the spatial impact of large transport infrastructure projects


After the Tōkaidō Shinkansen came into service, traffic growth exceeded all forecasts, rising from 10.7 billion passenger-kilometres in 1965 to almost 46.1 billion in 2008; passenger traffic density1 rose fivefold. It did result in some reallocation of roles among the cities it connected. Prior to the introduction of the Tōkaidō, Nagoya played the role of an “intermediary capital” between Tokyo and Osaka (ECMT, 1992). Advanced service activities were located there. After the Tōkaidō came into service, it seems that Nagoya began to lose this role. The number of jobs in such activities fell in Nagoya, while rising in Tokyo and even more so in Osaka. Nagoya did not decline, however; rather, it became more specialised in high-value manufacturing.


The introduction of high-speed rail (train à grande vitesse, TGV) in France between Paris and the country’s second city Lyon (a distance of about 464 km – comparable to that between Tokyo and Osaka) led to a steep increase in both directions.2 This reflected both a shift of about one-third of passengers from air to rail and an increase in the average number of journeys per passenger each year of about 30%. The Lyon region, far from being absorbed by the Paris region as some had feared, actually extended its markets. Some firms from the Lyon area had advantages over Parisian competitors that they had previously been unable to exploit (it took too long to get there; transport costs were too high, etc.). Chen and Hall (2012) examine the effect of increased connectivity, in the form of reduced travel time, and the degree to which it facilitates economic restructuring in de-industrialising regions in a comparison of Nord-Pas-de-Calais (France) and Lancashire (United Kingdom). They find that the roll-out of HSR in France has a more pronounced and broader regional impact than the mere upgrading of existing infrastructure, such as occurred in the United Kingdom, but that the benefits still tend to go to the dominating region (in this case, Paris).


Ahlfeldt and Feddersen (2015) examine the impact of the construction of an HSR link between Frankfurt and Cologne. They argue that the location of intermediate stations was effectively exogenous to local economies,3 which allows them to isolate the impact of rail connection on different places, as well as the agglomeration effect in surrounding areas and the spread of those effects in space. They find that the intermediate stops benefited substantially from the HSR as locations of economic activity but that they benefited far more as potential places to live, creating new commuting possibilities to the main centres. These benefits were highly localised, decaying rapidly with distance from the stations.4 This suggests that the maglev might have a similar effect, leading to greater concentration of economic activity along the line, and especially in the three major cities, with far greater impacts on housing markets in the intermediate places and those located near the line but lacking stops. Even Nagoya, the smallest of the three main cities involved, may lose activity to Tokyo, something officials there acknowledge as a real risk.


The Eurostar connecting London to Paris provides a salient case study with potential impacts at various geographic scales. Though services began in 1994, the project did not reach its full potential until 2007, when the HSR was fully extended to London. The benefits of HSR to the English County of Kent are, to date, ambiguous. The construction of intermediate stations (especially Ebbsfleet) connected to local rail network did lead to important investment in housing and commercial property, but these investments remain limited in scope and scale. Looking at the period when the Eurostar’s HSR stopped at the tunnel and continued on conventional lines to London Waterloo, Hay, Meredith and Vickerman (2004) find only limited impact on Kent and little potential for further development. This is consistent with other research suggesting that a region’s economic potential prior to large infrastructure investment is a critical determinant of the investment’s impact. Thus, Kveiborg’s (2014) comparative case study of Eurostar and the proposed Fehmarn Belt fixed link emphasises that the impact on the intermediate regions is likely to be limited and will accrue only over the long-term, except where those intermediate regions have pent-up potential that could be exploited through increased connectivity. The main exceptions are places where intermediate stops are located. In the case of the Eurostar, Lille and Ashford (before the re-routing of Eurostar via Ebbsfleet) enjoyed better economic performance than their respective regions in the first decade of Eurostar’s operation.

The Öresund

The opening in 2000 of a bridge connecting the southern Swedish region of Skåne to the Danish Capital Region has had profound and sometimes unexpected consequences for the region around the Öresund Strait (OECD, 2012c). Passenger numbers roughly tripled over the course of a decade and the number of daily commuters rose roughly seven-fold to around 20 000 per day. A housing boom ensued on the Swedish side, where property prices were far lower, while employment was concentrated in and around Copenhagen, where salaries are far higher – effectively a degree of integration by specialisation (Decoville et al., 2010). However, Malmö did not simply become part of Copenhagen’s commuter belt and some land-intensive activities relocated from Denmark to southern Sweden in search of lower costs. During 2000-09, employment growth in the Malmö region for basic services was 7 percentage points higher and for advanced services it was 14 percentage points higher compared to the Stockholm region. The rate of growth of new jobs in the Malmö region between 2000 and 2009 exceeded both the Stockholm and Gothenburg regions and Sweden as a whole. Cross-border integration stalled for a time following the global financial crisis, and – in contrast to the other cases cited here – the cross-border nature of the integration remains a constraint. Even in the European single market, the persistence of differences in tax legislation, labour codes and other regulatory regimes constitutes a barrier to deeper integration across the Öresund Strait.

1. Defined as passenger-kilometres per kilometre of line length.

2. Passenger numbers rapidly jumped from 1.5 to 3.7 million per year, rising at around 6% per annum thereafter.

3. Since the intermediate stops were the results of exogenous decisions, the authors exploit conditions approaching a natural experiment, assigning places with stops as the “treated counties” and comparing them to control cases (counties that were not treated). Once the cases are established, the authors use a difference-in-difference specification to assess the economic impact of the HSR.

4. In addition, the spread of economic activity to intermediate cities may be a result of the pent-up potential in the dense urban area of the Cologne-Frankfurt corridor and is unlikely to be consistently reproduced.

Sources: ECMT (1992), Transport and Spatial Distribution of Activities: Report of the Eighty-Fifth Round Table on Transport Economics Held in Paris on 5-6 April 1990, ECMT Round Tables, No. 85, OECD Publishing, Paris,; Hay, A., K. Meredith and R. Vickerman (2004), “The Impact of the Channel Tunnel on Kent and Relationships with Nord-Pas de Calais”, Centre for European Regional and Transport Economics, University of Kent, Canterbury; Chen, C. L. and P. Hall (2012), “The Wider Spatial-Economic Impacts of High-Speed Trains: A Comparative Case study of Manchester and Lille Sub-regions, Journal of Transport Geography, Vol. 24, pp. 89 110; OECD (2012c), OECD Territorial Reviews: Skåne, Sweden, OECD Publishing, Paris,; Kveiborg, O. (2013), Economic Effects of Large-Scale Infrastructure Projects, BeltTrade, Copenhagen; Ahlfeldt, G. M. and A. Feddersen (2015), “From periphery to core: Measuring Agglomeration Effects Using High-Speed Rail”, SERC Discussion Paper, No. 172.

Gibbons (2015) provides an overview of some of the issues involved in the decision to invest in large transport infrastructure for purposes of supporting regional development. The evidence for using transport investments as an efficient means to decrease regional disparities remains more than a little ambiguous. First, the superiority of transport over other types of investments to promote regional growth is not clearly established, and, secondly, there is a real risk of displacement going the “wrong” way, triggering the movement of firms and people from periphery to core (what OECD, 2012c, calls “leaking by linking”). That said, fears of a “hollowing out” of intermediate regions are probably exaggerated. Relatively few firms actually relocate in response to new transport connections, though connectivity issues certainly influence the location choices of firms that chose to move for other reasons (ECMT, 1992; Ahlfeldt and Feddersen, 2015). Transport considerations usually come into play only once the decision to move has been taken. Even then, the firm’s choice of location is dictated primarily by other things. Thus, while the quality of transport infrastructure is important, it is rarely the decisive factor. Even if firms develop new strategies as a result of the construction of new infrastructure, the spatial pattern of activities may be unaffected or little affected, because the new strategies may entail the movement of persons and goods rather than changes of firm location. It is thus important to take proper measure of the changes taking place in passenger and freight traffic.

The Chūō Shinkansen can stimulate innovation and entrepreneurship

In assessing the potential impact of the Chūō Shinkansen and related links, the first thing to bear in mind is that the changes it brings about will probably be marginal in the short term, if only because Japan’s transport networks are already so well-developed. The journeys that it will offer are already possible now, even if they take longer. The Chūō Shinkansen certainly will provide an important new option on a very well-travelled route, but it will not connect up regions that were previously poorly connected, let alone isolated from each other. This suggests that the economic impact will take time to be felt; the new line’s most immediate effect will probably be to intensify intermodal competition (chiefly with air travel) and to increase the number of journeys taken. To be blunt, if the train fares are at all competitive, then few travellers will opt to fly between Tokyo and Osaka unless they have connecting flights to other destinations.

The longer-term impacts depend in large measure on its success in integrating regional investment and innovation systems rather than regional labour markets. Transport costs are at the core of most regional economic development models (Puga, 2012). A reduction in transport costs increases the size of the accessible market for firms, which in turn increases the concentration of firms and production in in attractive locations. The maglev is unlikely to reduce the cost of transporting goods by a significant margin and, as noted above, it will not sustain large-scale commuting flows. Its main attraction for firms is thus likely to lie in providing access to and for highly-skilled employees with a high opportunity cost for their time, such as consultants, researchers and managers. High-speed trains, at least those that are well-integrated into the local transport network, reduce the time spent in transit compared to air travel; compared to car travel, they offer both greater speed and the opportunity to work comfortably during the trip. This is less about labour market integration than new opportunities for business-to-business connections – above all, about learning and investment flows.

First, the Chūō Shinkansen should facilitate learning. Researchers and managers are only a small fraction of the workforce, but employees who are engaged in high-value added parts of the production chain have the potential to create significant spillovers both within firms and in their cities and regions. The decisions taken by senior managers affect the productivity of all their workers and ultimately the success of their companies. Empirical evidence shows that high-tech jobs in a metropolitan region create employment opportunities in local service industries for both high-skilled (e.g. teachers or doctors) and low-skilled workers (e.g. cleaners or taxi-drivers) workers. The estimated impact can be substantial: for the United States, estimates suggest a five-fold multiplier effect from employment in tradable high-tech industries. This means that the 12 000 workers employed by Apple in Cupertino generate an additional 60 000 jobs in the wider metropolitan area (Moretti, 2012). The larger accessible area created by the Chūō Shinkansen will generate greater potential for knowledge spillovers via the interactions among high-skilled individuals. Employees and firms learn from each other by observation and face-to-face interactions, for which telecommunications solutions are an important complement but a poor substitute (Box 3.7).27 The new line will offer them more opportunities for such interactions, enabling them to learn about the most efficient production methods and to adapt accordingly.

Box 3.7. Globalisation, communication and transaction costs

There is a widespread perception that globalisation in general and the advance of information and telecommunications technology in particular have led to the “death of distance”, creating a global economy that is flat and borderless (Cairncross, 2001; Friedman, 2005). If this were the case, then the rationale for the concentration of economic activity in cities would be far weaker than at any time in centuries, and the case for expensive physical connective infrastructure (as opposed to ITC infrastructure) linking relatively proximate places would be hard to sustain. In fact, the evidence to date suggests a much more complex picture of the impact of globalisation on the spatial organisation of economic activity. Spatial transactions costs for routine, standardised and non-knowledge intensive activities have fallen; this drop in transaction costs has formed the basis for much recent outsourcing and the organisation of ever more complex global value chains. However, it would appear that spatial transactions costs for non-routine, non-standardised and knowledge-intensive activities have risen, a fact that is of huge significance in a world in which value creation is increasingly linked to knowledge rather than physical resources. A number of studies find that transactions costs have increased for many knowledge and time-related activities (Bouhol and de Serres 2010; McCann 2007; Duranton and Storper 2008; Disdier and Head 2008). At the same time, McCann (2007) finds that in knowledge-intensive activities, the optimum frequency of face-to-face interaction has increased. It is findings such as these that point to the potential of very fast connections like the Chūō Shinkansen to foster the integration not of urban labour markets but of urban and regional innovation systems.

Secondly, it will also allow for more venture investment, as it will make close monitoring of such investments easier. The evidence on venture capital (VC) markets suggests that distance matters. The number of VC financing rounds a start-up requires increases with the distance between the company and its investors and the amount of capital raised each round decreases with distance (Tian, 2011). The ability to visit their investments on day-to-day basis makes a significant difference for VC investors. Investors bring more than money; especially in early stages of development they can provide expertise, guidance and contacts, as can be observed in the role played by early venture capitalists in choosing the location and business model of the pharmaceutical firm Amgen (Box 3.8). The kind of high-speed connections made possible by the maglev should make it easier and cheaper for VC (and other) investors to engage frequently and face-to-face with the firms they are backing.

Box 3.8. Amgen: Conceived in San Francisco, grown in Los Angeles

Amgen is a multinational biopharmaceutical company ranked 12th globally in terms of 2014 revenue. Its success is the combination of cutting-edge technology, top researchers and investors experienced in creating marketable products. The company was founded in 1980, on the initiative of several collaborating venture capitalists from the San Francisco Bay area. Initially they tried to find local collaborators, but a saturated market prompted them to look further south, and, as a result, the company in based in Thousand Oaks, California, part of the Los Angeles metropolitan area. This location was chosen, because it was roughly equidistant to the universities of its scientific advisory board and because local housing costs were then relatively cheap and amenities relatively high. Importantly the contribution by the venture capitalists included finding a CEO who could bring together a group of academics for its advisory board, the identification of a law firm that accompanied the incorporation of the company and the introduction of several individuals to the company’s board to establish credibility and guide the early stages of development.

Sources: Rathman, G. B. (1999), “‘Biotechnology Startups’ in Springham”, in D.G. and Moses, V. and Cape, R.E. (eds.) Biotechnology - The Science and the Business, Taylor & Francis; Bowes (2009); Staton, T. (2015), “The top 15 pharma companies by 2014 revenue”, FiercePharma, 18 March,, (accessed on 1 August 2015).

The capacity to build well-functioning networks among key local actors can create and sustain economic growth. Starting from very similar fundamental strengths in the 1970s, the metropolitan regions around Los Angeles and San Francisco in California have experienced strongly divergent trends. Both were among the wealthiest US metropolitan regions in the 1970. However, while San Francisco and its Silicon Valley have maintained a leading position, Los Angeles (LA) has slipped from 4th in the country to 25th in 2010. Storper, Kemeny, and Makarem (2015) argue that the greater capacity for organisational change and therefore a greater capacity to adapt to a changing economic environment has been the driver of San Francisco’s success. This capacity is largely due to the formation and strengths of interrelated networks of actors and institutions. Storper (2009:49) concludes that “the actor-networks function differently in the two regions – in the Bay Area they are decentralised but criss-crossing, leading to powerful regional identities and the formation of pragmatic coalitions to get things done. In LA, they are decentralised but not criss-crossing, leading to weak regional identities, and a culture of distrust from one locality to another (and especially of the ‘big dogs’ of LA County and City).”

The contrasting fortunes of Los Angeles and San Francisco underscore the important role played by the cities themselves and their private sectors in fostering the kind of business environment that will enable them to profit fully from the new connections made possible by the maglev. For policy makers in the Tokyo-Nagoya-Osaka mega-region, this will primarily concern enhancing the environment for innovation, entrepreneurship and venture investment. In short, the economic benefits of the new maglev line will depend as much on getting complementary policies right as on the building of the line itself.

The 2020 Olympics

The growth and employment effects of the Olympics are difficult to anticipate…

In September 2013, Tokyo was selected to host the 2020 Olympic and Paralympic Games. Since then, much attention has been focused on the infrastructure development this will entail and on the ways in which these events can enhance the Tokyo area’s international competitiveness over the longer term. In keeping with Japanese urban policies, the Tokyo Olympic and Paralympic Games are to feature the principles of compact urban development. For example, most of the venues are to be located within eight kilometres of the Olympic Village. The major infrastructure development projects to support transportation to Olympic venues include the Harumi Line of the Tokyo Metropolitan Expressway, the national road designated as Route 357 and a second ring road. An August 2012 study conducted for the Tokyo 2020 Bid Committee estimated that the Games would generate economic activity in Japan worth JPY 3 trillion (USD 37.9 billion), equivalent to 0.64% of 2013 GDP, and would create more than 150 000 jobs. Spread over many years, that would imply a stimulus of less than 0.1% of GDP per year – far smaller than the estimated effect of the 1964 Games (Kang, 2007). Even so, Mizuho (2014) argues for an even larger impact, of close to 0.3% per year. This, however, is based on simple comparisons of trend growth in countries before the bid was awarded with performance during the run-up to the Games; it leaves out a range of factors found to be relevant in assessing such impacts elsewhere (e.g. Fourie and Santana-Gallego, 2011).

There are in any case good reasons for viewing these impact assessments with great scepticism. There has long been considerable controversy about the economic impact of the Olympic Games on host cities and countries, with critics arguing that bid backers routinely overstate the potential benefits and some assessments finding that the net economic outcomes are negative (see Box 3.9 for an overview of this debate). Even assessments that acknowledge a strong stimulus effect can be critical, as in cases where the environmental and social costs are discounted or where – as has happened for many sporting mega-events – the pattern of public investment was distorted so as to delay or prevent many more economically productive investments (Whiting, 2014). Yet even positive evaluations emphasise that the costs of hosting a major event can rarely be justified in terms of the event alone: it is the success of the event plus the value of the legacy that justify the costs (OECD, 2010a). The Games need to be set in a longer-term strategic vision and plan for building on the Olympic legacy to enhance Metropolitan Tokyo’s global competitiveness.

Box 3.9. Assessing the impact of hosting “mega-events”

There has long been controversy in many cities about the value of hosting major sporting events like the FIFA World Cup or the Olympic Games. In many cases, the willingness of cities to invest heavily in hosting such events has contrasted with the lack of strong evidence that they yield the kind of tangible long-term benefits that might justify such investment. This contradiction extends to the body of research on the Olympics, in particular, which tends to be divided between studies commissioned by host cities, which tend to identify clear benefits to hosting the Games,1 and the academic literature, which, for the most part, finds few discernible long-term benefits to hosting the games beyond the satisfaction they provide to residents and officials.2 In part this may stem from the tendency of the former to rely on input-output models, while academic researchers have increasingly used computable general equilibrium (CGE) models.3 Collins et al. (2009) emphasise the importance and complexity of assessing the environmental impact of mega events and draw attention to the very difficult methodological issues involved, as well as problems with data availability.4 Ultimately, of course, there is an additional problem: any assessment is only as good as the implicit counterfactual – the assessment of what would have happened without the games.

However, there are important divergences in the scholarly literature as well. While the consensus as to the elusiveness of an “Olympic dividend” is strong (Madden, 2002; Kasimati and Dawson, 2009; Zimbalist, 2012), estimates of the specific effects and the magnitude of specific costs/benefits vary widely. Some studies find a significant growth effect in the preparation phase but little lasting impact after the games (Kasimati and Dawson, 2009). While many studies acknowledge a significant impact on employment in the run-up to the Games, the employment effects following the games are minimal (Madden, 2006; Kasimati and Dawson, 2009).

Rose and Spiegel (2009) accept the consensus concerning the lack of direct economic benefits of hosting mega-events, but argue that there is a substantial positive impact on trade. They find that trade is around 30% higher for countries that have hosted the Olympics and that this effect is robust and sustained. However, they also find that unsuccessful bids have a similar effect. That, in turn, implies that the Olympic trade effect is a consequence not of the Games themselves, which are signal of something deeper, but of the underlying orientations that prompt a country to bid for the Games in the first place.5

Given the importance of tourism in estimating the impact of hosting mega-events, care should be taken to ensure that assessments of the impact consider second-round effects on the tourism sector and not just the number of tourists directly or indirectly attracted to the event. Lee and Taylor (2005) highlight in particular the impact of such factors as diversion of tourism in time (visitors come during the Games who would have come earlier or later6) or space (tourism to other parts of the country declines). This is especially relevant for countries that already have a well-developed tourism industry, and it is even more important when much of the habitual flow of tourism is to the city that hosts the event. Fourie and Santana-Gallego (2011:1) conclude that, “in general, results suggest that mega-events promote tourism but the gain is dependent on the type of mega-event, the participating countries, the host country’s level of development, and whether the event is held during the peak- or off-season.” They find that, other things being equal, non-OECD countries benefit far more from hosting an event than OECD members and that the effects are weaker in countries where tourism demand is already strong. This would suggest that, as an advanced economy with a strong tourism sector, Japan probably has less to gain from the Games now than it did in 1964.

A number of studies point to the importance of idiosyncratic factors in shaping the impact of the Games. Los Angeles (1984), for example, is often cited as a city where the Olympic Games proved to be profitable, but that was the result of the decision to build minimal new infrastructure and to rely almost exclusively on private investment (Appelbaum, 2004). In addition, Los Angeles was the sole bidder that year and was therefore not held to a range of International Olympic Committee requirements. Barcelona (1992) was already on the rise economically, making it difficult to disentangle the magnitude of the Olympic effect, but some assessments certainly suggest that it helped raise the city’s global profile, particularly for tourism. Yet even such a re-branding may not last: Zhang and Zhao (2009) show in the case of Beijing how shallow and short-lived the image boost was. In the cases of Tokyo (1964) and Seoul (1988), the benefits appear to be linked to the status of those countries at the time as rapidly emerging trading powers, in line with the findings of Rose and Spiegel (2009). The Games were a clear signal that the countries were open for business and welcoming to investment at a time when their economies were ripe for rapid growth.

Finally, Kavetsos and Szymanski (2009) find evidence that hosting the event, though not necessarily athletic success, does increase the life satisfaction of host-country residents. However, the effect, while strong, is short lived. In sum, there is little evidence that Olympic Games do much beyond what the country/city could achieve otherwise, other than bring residents pride and happiness. This is especially true in cities with high international profiles and strong tourism sectors even before the Games.

1. See, e.g. Economic Research Associates (1984), Humphrey and Plummer (1995), Oxford Economics (2012), and the study of the London 2012 Games examined by Shapiro (2014), which was commissioned by the government but subjected to no peer review and which came in for sharp criticism on methodological grounds.

2. See, for examples, Madden (2002), Baade and Matheson (2002), Preuss (2004), Giesecke and Madden (2007), Kasimati and Dawson (2009).

3. Blake (2005) argues that the differences between the two approaches can affect the results, with input-output methods tending to produce more positive estimates, owing to the assumption of fixed prices and fixed coefficients in the models. This typically results in much higher estimates than are generated with CGE models. This is particularly important for estimating tourism demand and the impact of construction expenditure on the economy – two areas of particular importance when assessing the impact of the Games.

4. Ecological footprint analysis has the advantage of taking into account global environmental impact and measures that are easily communicated, but is difficult to implement for large-scale events where too many factors need to be considered. The environmental input-output analysis is more transparent and detailed, but relies on a set industry structure that may not be appropriate and is restricted to national analysis. Both approaches are limited by data availability in linking economic and environmental factors.

5. The authors do not control for the self-selection of countries into the sample of countries bidding for the Games.

6. Oxford Economics (2012) notes that international tourist arrivals typically decline in the year prior to an event, as people postpone visits to coincide with it or seek to avoid the disruption caused by construction of the major infrastructure projects. The case of the London Games is instructive: the United Kingdom received about 5% percent fewer foreign visitors in August 2012 than in August 2011. Those who came for the Games did spend more, but the country had already spent billions of dollars to attract them (The Telegraph, 2012).

One of the major difficulties in assessing the balance of costs and benefits lies in the complexity of assessing costs from bid preparation through the handling of legacy assets and liabilities, and identifying all the relevant expenditure sources, which typically include a complicated mix of public funds, sponsorship, ticket sales, International Olympic Committee funds and other sources (Box 3.10). An even greater one is simply that sporting mega-events are relatively rare and have been conducted in such a wide range of different countries and circumstances that it is difficult to generalise. The specific effects and magnitude of cost/benefits may depend on quite idiosyncratic factors. For example, Oxford (2012) concludes that the London 2012 Games benefited from being a clearly counter-cyclical event. While much of the literature points to the impact of Olympic construction and tourism on wages and costs in those two sectors, the preparation of the Games coincided with a period of weakness in the UK and world economies. Construction workers’ wages, in particular, did not increase during the build.

Box 3.10. Costing the Games: London’s experience

The history of the most recent summer Games, held in London in 2012, illustrates how hard it can be to predict Olympic costs. The Games cost the taxpayer vastly more than expected at the time of the bid in 2005. Indeed, as early as March 2007, the government reported that the public budget for preparation of the Olympics had more than tripled, from an initial estimate of just under GBP 2.4 billion at the time of the bid, to GBP 9.3 billion (USD 14.6 billion). Such overruns are not uncommon: Flyvberg and Stewart (2012) examine the 17 Winter and Summer Games held over the 1960-2012 period for which sufficient data are available and find that the average cost overrun was 179% (median 112%). The respective figures for summer Games were 252% and 118%, respectively.

In the end, the Games did cost the UK taxpayer just about as much as was projected in 2007 – GBP 9.3 billion – though the structure of that spending was very different to what had been anticipated (operational costs were far higher but the construction of the venues was about GBP 1 billion lower than expected). This was in addition to the London 2012 Organising Committee’s own GBP 2.2 billion operational budget for the day-to-day running of the Games, which came almost entirely from sponsorship, ticket sales, merchandise and from the International Olympic Committee.

Against this total of GBP 11.5 billion in costs, Oxford Economics (2012) estimates the Games’ benefits to the UK economy at around GBP 16.5 billion over 2005-17. That estimate, however, is subject to some of the concerns reviewed in Box 3.5 above – lack of a general equilibrium framework, a lack of clarity with respect to the methods used to derive some estimates and an apparently ad hoc handling of deadweight costs, in particular. It is, however, remarkably thorough in considering expenditure, employment and other factors across sectors and across space.

Sources: Berman, G. (2010), “Financing the London 2012 Olympic Games”, SN/SG/3790, House of Commons Library, London, 27 July; Oxford Economics (2012), The Economic Impact of the London 2012 Olympic & Paralympic Games: A Report by Oxford Economics Commissioned by Lloyds Banking Group, Oxford, July, available at:

…but Japan can do much to hold down costs and maximise the benefits of the Games

Given that Japan is committed to hosting the Games in Tokyo in 2020, the essential question concerns the lessons that can be drawn from the experiences of recent Olympic hosts to make the Games as beneficial as possible. The economic literature surveyed above points to a number of things:

  1. To the extent possible, the authorities should aim to use existing infrastructure and facilities, upgrading where necessary but limiting the amount of new facilities to be built. Tokyo is undoubtedly fortunate with respect to the first of these points. Its existing infrastructure endowments and its position as a past Olympic host should help to limit the costs that it might otherwise incur.

  2. Location matters. In some of the world’s most expensive cities, perhaps the greatest opportunity cost is the loss of scarce and valuable real estate. This point is more challenging: given land pressure and land costs in Tokyo, there is simply no way to avoid tying up some of the most expensive real estate in Japan for the Olympic sites. The alternative would be inaccessible locations on the fringes of the Tokyo Metropolitan Region, which would hardly be desirable.

  3. Precisely because the Olympic venues will occupy some of the most expensive real estate in Japan, it is critical to plan for the future use/disposal of the Olympic venues – many past Olympic hosts have been left with “white elephants” – large sports facilities that are expensive to maintain and for which there is little demand. The authorities also might wish to create an Olympic Legacy company or foundation to take forward the infrastructural legacy of the Games. If so, it should be probably created sooner rather than later and should be involved in the preparation of the venues themselves. Tokyo Metropolitan Government has been working on development plans to use the Olympic legacy sites, especially the Athletes Village, which will become a residential area.

  4. Mobilising private finance is critical, but care must be taken to ensure that this does not result in the Games becoming a rent-seeking opportunity for the architects, construction firms, banks and law firms involved in the process. The pressure to deliver facilities fast in the frenzied run-up to the Games can otherwise lead to spectacular waste and even corruption (Zimbalist, 2012).

London seems to have done rather well so far at managing the post-Games use of many of its Olympic venues: some have been handed over to charitable social enterprises or community groups, some dismantled and relocated, and others adapted to alternative uses. The Olympic Park itself has been redeveloped, renamed and reopened as an important park for East London. However, even London has had difficulties with some sites. The basketball arena was dismantled (it was the largest temporary structure ever erected for an Olympic event) but plans to reuse it in Rio de Janeiro in 2016 came to nothing.28 Plans to lease or sell the stadium have so far been frustrated and it has been taken into public ownership. Stadiums often pose the largest legacy costs: they are expensive to build and difficult to operate profitably, which is why they tend to be built in most places with taxpayer support.29 Thus, even the iconic Bird’s Nest in Beijing largely stands idle.

OECD (2010a) highlights a number of other issues taken in connection with the London 2012 Games that helped strengthen their orientation towards regeneration, growth management and sustainability. These included:

  • Planning public transport investment for the Games with a view to creating a socially and environmentally sustainable Olympics as well as better long-term connectivity for East London.

  • The creation of legacy master plans and a strategic regeneration framework for the area.

  • The introduction of the “CompeteFor” programme to help local small and medium enterprises (SMEs) improve their competitiveness and respond to the opportunity to get involved in Olympic-related supply chains.

The Games represent an important challenge for multi-level governance

The fourth of the lessons cited above points to the importance of co-ordinated governance in ensuring the success of the Games. The experience of previous hosts suggests that the scope and depth of inter-ministerial involvement and co-ordination across levels of government can be important factors in managing service delivery and infrastructure planning for the Olympics:

  • In Sydney, the Olympic Co-ordination Authority (OCA) managed the participation of a wide network of institutions, including the Office of Olympic Co-ordination, in the Department of Premier and Cabinet; the Olympic Construction Authority, housed in the Public Works Department; the Homebush Bay Development Corporation, which was responsible for providing most Olympic venues; part of the Department of Sport and Recreation, which had responsibility for delivering certain new Olympic facilities; and part of the Department of Planning, which was responsible for securing venues for equestrian and mountain bike events (SOCOG, 2001).

  • For the 1988 Seoul Games, each of the government ministries organised special committees to take exclusive charge of projects relating to the Olympic Games. It was incumbent upon all ministries to create a festive mood, encourage public relations for the Games and offer support in manpower and materials. These included the Economic Planning Board and the ministries of finance, justice, sport, agriculture and fisheries, construction, culture and information, and science and technology (SOOC, 1989).

  • In preparing for the 2010 Vancouver Winter Olympics, the government of Canada drew up a multiparty agreement that clarified both its financial commitment and its technical support for the city of Vancouver.30 This agreement addressed such issues as visas and entry procedures, tax and customs relief for the import of goods needed for the Games, financial guarantees equal to the sums allocated by the Province of British Columbia, a federal contribution towards the legacy endowment fund to support the future maintenance and operation of key venues; security for the Games, and allocation and fees for broadcasting. Finally, the government committed to refrain from hosting other important national or international meetings in or near Vancouver in the period immediately before, during or right after the Games (Government of Canada, 2002).

As Olympic planning gathers steam, the Japanese authorities may want to consider the best modalities for organising co-ordination of Olympic efforts across policy sectors and across levels of government from the national to the local.

The Games also offer opportunities to realise important non-monetary benefits

Japan’s position as the first super-ageing society to host the Games (Saito, 2013) also merits specific attention at it prepares for 2020 – not least because it will surely not be the last such host country, and its choices will therefore be relevant to others. While it will, as noted above, be desirable to try to limit the volume of new infrastructure to be built for the Games (Japan does not need another public works-driven stimulus package), the Olympics will still require substantial new construction. Ensuring that new infrastructures and facilities are barrier-free and accessible to all will be an important and tangible legacy of the Games. Japan’s technological prowess already gives it an edge when it comes to aiding people with mobility and visual problems and other special needs.

The Games will also offer Japan a chance to reinvigorate and deepen its economic and social integration into the larger world. As noted in Chapter 1, there is a widespread perception, both within Japan and abroad, that the country has been turning inwards in recent decades. The Games offer an opportunity to change both the perception and the reality underlying it. To do this, Japan must do more than add more English-language signage in public places. It needs to mobilise local communities not merely to volunteer individuals for assigned tasks but to develop and implement initiatives of their own. For example, MEXT (2014) proposes building on the success of the one school-one nation programme created for the 1998 Nagano Winter Games, expanding it to communities: community and neighbourhood groups could host activities around different events or providing volunteers to help at nearby venues – what MEXT calls a “one-district-one-discipline initiative”. Mobilising seniors would be an important element of such initiatives.

Japan will want to leverage the Games to realise other potential environmental and cultural benefits, as well:

  • It is important to recognise the Games as an opportunity to pursue environmental improvements. For example, Germany adopted a Green Games Programme for the 2006 FIFA World Cup. Plans for London 2012 included specific provision to use world class standards in sustainable construction.

  • One of the major benefits of the Athens Games in 2004 was the restoration of ancient sites and buildings as part of preparations for the Games. This could be especially important for Japan, in view of its exceptionally rich cultural heritage and also of criticism that the (sometimes hasty) preparations for some past Games have harmed the hosts’ cultural heritage (Whiting, 2014).

  • It is important not to overlook the potential of major events to contribute to local revitalisation. These were important themes for London 2012 and the Manchester Commonwealth Games of 2002. This is not an automatic consequence of bringing investment and visitors into a place for an event. It requires a degree of intentionality, not least the preparation of measures to train workers and in engaging local firms and populations early and intensively. The Lille City of Culture Programme and the Sydney Olympics both achieved very high levels of local participation, ensuring that local inhabitants shared ownership of the event rather than looking on as perhaps disaffected bystanders (OECD, 2010a).

Social policy for large cities

Better urban policies could raise both fertility and female activation rates

As noted in Chapter 1, the major cities tend to be where fertility is lowest; perhaps surprisingly, the same is true of female labour force participation. The labour force participation rates of prime-age (25-44) married women in Tokyo, Osaka, Hyōgo, Chiba, Kanagawa and Nara prefectures are all 5 percentage points below the rate for Japan as a whole (Figure 3.25). One of major reasons why women in large cities find it hard to combine work and parenthood is the lack of childcare facilities. Since female labour force participation has steadily been increasing, the demand for childcare services has also been increasing even as fertility has fallen. In Tokyo, the demand for childcare services is even greater, since the number of preschool-age children is still growing, in contrast to the rest of Japan (the city’s growth more than offsets its lower fertility rate). While the capacity of childcare services is increasing, in 2013, more than 8 000 children in Tokyo were wait-listed for certified nurseries in their neighbourhoods. The authorities are trying to address this: the Act on Child and Childcare Support enacted in 2012 aims to improve both the quantity and quality of childcare service, by providing childcare services for approximately 400 000 additional children by the end of FY 2017 and by improving staff-child ratios and working conditions for staff. It will also extend support to a wider range of childcare services, to reflect different needs.

Figure 3.25. Share of married women in the workforce, in 2010 (%)
Participation rate of prime-age, married women, by prefecture

Source: Data from the 2010 Japanese census provided directly by the authorities.

Spatial development plans and strategies can also help facilitate the combination of careers and family life. For example, long daily commutes seem to be another obstacle for working parents in large metropolitan areas. The average commuting time for working women aged 25-44 in the cities of Tokyo, Yokohama, Kawasaki, Chiba and Saitama was 80 minutes in 2005, compared to 49 minutes in four Hokuriku prefectures,31 though their average work hours are about the same across the country (Hashimoto and Miyagawa, 2008). Promoting mixed-used urban development in order to make jobs and homes closer at a metropolitan scale is an important long-term urban spatial strategy. Floor area ratio bonuses and other type of incentives could be also used in order to facilitate the provision of childcare facilities at strategic locations in central urban areas.

Another option is to develop more childcare facilities at or near railway stations. In 1996, the Japan Railways Co. (JR) opened its first nursery at JR Kokubunji station in Tokyo. Working parents could take their children to the childcare facility at the station and directly take trains to work. JR has since expanded this programme and as of April 2015, it provided 82 childcare facilities of various types at and near stations in the Tokyo metropolitan area, in collaboration with local governments. JR also works closely with childcare facilities near its stations and provides “pick-up” services. In the mornings, parents can drop their children at a station, and JR staff members take them to their childcare facilities. In the evenings, the children are brought back to the station and looked after until their parents collect them. Similar projects are being carried out by all the major railway companies in Tokyo, as well as in Osaka, Kobe and many other cities in Japan.

Some initiatives could help both families with children and the elderly

As noted above, the population ageing challenges that face the large cities differ from those confronting other places. Two particular types of neighbourhoods in large cities call for policy intervention as regards to ageing. The first are decayed and under-invested built-up areas found in urban centres. These neighbourhoods are typically home to large concentrations of older people, who typically live in old wooden buildings that are not earthquake-resilient. Some houses are vacant and poorly maintained. The other places where older people tend to be concentrated in big cities are large-scale housing developments, or “new towns”, in metropolitan suburbs. These tend to be houses constructed in the 1960s and 1970s, when urbanisation was at its fastest. In many instances, the concentration of the elderly in these neighbourhoods also creates a concentration of urban poverty. Long-term urban sustainability will be at risk if urban policies do not address these problems.

Urgent action is needed to roll out comprehensive elderly-care services in these neighbourhoods. Given the speed of increase of the elderly, it is also important to continue promoting elderly-friendly rental housing, with care services for those who stay at home, as alternatives to hospitalisation or long-term care facilities. The central government can play an important role in reorganising regulatory frameworks for land use and construction and promoting the provision of elderly-care (or child-care) facilities in existing residential neighbourhoods. For example, while it requires special permission to build office buildings in most residential areas, small office space for elderly-care services could be permitted as a default use of zoning, considering the magnitude of urban ageing. Converting vacant homes into elderly-care (or child-care) facilities should also be facilitated. Moreover, families with children, no less than older people, can benefit from the creation of walkable environments and more accessible public services in these neighbourhoods. Monitoring service accessibility is important here: for example, Portland’s “20-minute neighbourhood” programme aims to ensure everyday services within 20 minutes (by walking, cycling or mass transit) and to promote the location of services in neighbourhoods in which they are insufficient (OECD, 2012c). In fact, most of age-friendly urban policies are also good for families with children.


Ahlfeldt, G. M. and A. Feddersen (2015), “From periphery to core: Measuring Agglomeration Effects Using High-Speed Rail”, SERC Discussion Paper, No. 172.

Ahrend, R. et al. (2014), “What Makes Cities More Productive? Evidence on the Role of Urban Governance from Five OECD Countries”, OECD Regional Development Working Papers, No. 2014/05, OECD Publishing, Paris,

Appelbaum, B. (2014), “Does Hosting the Olympics Actually Pay Off?”, The New York Times, 5 August.

Australian Department of Transport (2012), Long run economic and land use impacts of major infrastructure projects, SGS Economic & Planning.

Baade, R.A., V. Matheson (2002) “Bidding for the Olympics: Fool’s Gold?”, in C. Barros, M. Ibrahimo and S. Szymanski (eds.), Transatlantic Sport: The Comparative Economics of North America and European Sports, Edward Elgar, London, pp. 127-151.

Berman, G. (2010), “Financing the London 2012 Olympic Games”, SN/SG/3790, House of Commons Library, London, 27 July.

Blake, A. (2005), “The Economic Impact of the London 2012 Olympics,” Christel DeHaan Tourism and Travel Research Institute Working Paper, Nottingham University Business School.

Boulhol, H. and A. de Serres (2010), “Have developed countries escaped the curse of distance?”, Journal of Economic Geography, Vol. 10/1, January, pp. 113-139.

Bowes, Jr., W. K. (2009), “Early Bay Area Venture Capitalists: Shaping the Economic and Business Landscape,” interview conducted by Sally Smith Hughes in 2008, Regional Oral History Office, The Bancroft Library, University of California, Berkeley, (accessed 31 July 2015).

Boyko, C. and R. Cooper (2011), “Clarifying and re-conceptualising density”, Progress in Planning, Vol. 76/1, pp. 1-61, July,

Brezzi, M. and D. Sanchez-Serra (2014), “Breathing the Same Air? Measuring Air Pollution in Cities and Regions”, OECD Regional Development Working Papers, No. 2014/11, OECD Publishing, Paris,

Bridges, B. (2008), “The Seoul Olympics: Economic Miracle Meets the World”, The International Journal of the History of Sport, Vol. 2514, pp. 1 939-1 952.

Cabinet Office (2015a), Basic Plan for Urgent Implementation of Measures for the Tokyo Inland Earthquake (in Japanese), available at:

Cabinet Office (2015b), Brochure on Disaster Management in Japan, available at:

Cairncross, F. (2001), The Death of Distance: How the Communications Revolution Is Changing Our Lives, Harvard Business School Press, Cambridge, MA.

Chen, C. L. and P. Hall (2012), “The Wider Spatial-Economic Impacts of High-Speed Trains: A Comparative Case study of Manchester and Lille Sub-regions, Journal of Transport Geography, Vol. 24, pp. 89-110.

Chen, Z. and K. Haynes (2013), “Spatial Impact of Transportation Infrastructure: A Spatial Econometric CGE Approach”, in P. Nijkamp, A. Rose and K. Kourtit (eds.), Regional Science Matters, Springer International Publishing, New York, pp. 163-186.

City of Kyoto (2015), “Projects to Create Low-Carbon City of Kyoto”, Presentation at the 6th High Level Seminar on Environmentally Sustainable Cities, 9-10 February 2015, Johor Bahru, Malaysia.

Collins, A., C. Jones and M. Munday (2009), “Assessing the Environmental Impacts of Mega Sporting Events: Two Options?”, Tourism Management, Vol. 30, pp. 828-837.

Decoville, A. et al. (2010), “Spatial Integration in European Cross-Border Metropolitan Regions: A Comparative Approach”, Working Paper, No. 2010-40, CEPS/INSEAD, Esch-sur-Alzette.

Demetriades, P. and T. Mamuneas (2000), “Intertemporal Output and Employment Effects of Public Infrastructure Capital: Evidence from 12 OECD Economies”, Economic Journal 110, pp. 687-712.

Demographia (2015), “Population Density, Traffic Density and Nitrogen Oxides (NOx) Emission Air Pollution Density in Major Metropolitan Areas of the United States”, available at:

Dierkes, J. and N. Akimoto (2014), “How Japan Could Truly Benefit from the 2020 Olympics”, The Diplomat, 23 March.

Disdier, A.-C. and K. Head (2008) “The Puzzling Persistence of the Distance Effect on Bilateral Trade”, The Review of Economics and Statistics, Vol. 90/1, February, pp. 37-48.

Duranton, G. and D. Puga (2004), “Micro-foundations of Urban Agglomeration Economies”, in: J.V. Henderson, J.F. Thisse (eds.) Handbook of Urban and Regional Economics, Vol. 4., North Holland, Amsterdam.

Duranton, G. and D. Puga (2002), “From Sectoral to Functional Urban Specialization”, NBER Working Paper, No. 911, August.

Duranton, G. and M. Storper (2008), “Rising Trade Costs? Agglomeration and Trade with endogenous Transaction Costs”, Canadian Journal of Economics, Vol. 41/1, February, pp. 292-319.

ECMT (1992), Transport and Spatial Distribution of Activities: Report of the Eighty-Fifth Round Table on Transport Economics Held in Paris on 5-6 April 1990, ECMT Round Tables, No. 85, OECD Publishing, Paris,

Economics Research Associates (1984), Community Economic Impact of the 1984 Olympic Games in Los Angeles and Southern California, Los Angeles Olympic Organizing Committee, Los Angeles.

Feigenbaum, B. (2013), “High-speed rail in Europe and Asia: Lessons for the United States”, Policy Study 418, Reason Foundation, available at:

Ferber, U. and T. Preuss (2006), Flächenkreislaufwirtschaft: Erprobung und Umsetzung, Deutsches Institut für Urbanistik, Berlin.

Fire and Disaster Management Agency (2015), Result of the Survey on Earthquake-Resistant Public Facilities for Disaster Management, Press release on 18 February 2015 (in Japanese), available at:

Flyvberg, B. and A. Stewart (2012), “Olympic Proportions: Cost and Cost Overrun at the Olympics 1960-2012”, Said Business School Working Papers, University of Oxford, June.

Fourie, J. and M. Santana-Gallego (2011), “The Impact of Mega-Sports Events on Tourist Arrivals”, Tourism Management, Vol. 32, pp. 1 364-70.

Friedman, T. (2005), The World is Flat: A Brief History of the Twenty-First Century, Farrar, Straus and Giroux, New York.

Fundación Idea/Cámara de Senadores/SIMO Consulting (2014), “México compacto: Las condiciones para la densificación urbana inteligente en México”, Fundación Idea, Cámara de Senadores and SIMO Consulting, Federal District, Mexico, available at:

Gaigné, C., S. Riou and J.-F. Thisse (2012), “Are compact cities environmentally friendly?”, Journal of Urban Economics, Vol. 72, No. 2-3, pp. 123-136, September-November,

Gibbons, S. (2015), “Planes, Trains and Automobiles: The Economic Impact of Transport Infrastructure”, SERC Policy Paper, No. 13.

Giesecke, J. and J. Madden (2007) “The Sydney Olympics, Seven Years On: An Ex-post Dynamic CGE Assessment”, Centre of Policy Studies, Monash University Working Paper, G-168.

Government of Canada (2002), “Annex E. Covenant of the Government of Canada: Vancouver’s Candidacy for the 2010 Olympic and Paralympic Winter Games” in “Multiparty Agreement for the 2010 Olympic and Paralympic Games”, 14 November, (accessed 15 July 2015).

Hashimoto, Y. and S. Miyagawa (2008), “Why is Women’s Labor-Force Participation Rate Low in Metropolitan Areas?”, RIETI Discussion Paper Series, 08-J-043 (in Japanese), available at:

Hay, A., K. Meredith and R. Vickerman (2004), “The Impact of the Channel Tunnel on Kent and Relationships with Nord-Pas de Calais”, Centre for European Regional and Transport Economics, University of Kent, Canterbury.

Hixson, M. et al. (2012), “Resolving the Interactions between Population Density and Air Pollution Emissions Controls in the San Joaquin Valley, USA”, Journal of the Air and Waste Management Association, Vol. 62/5, pp. 566-575.

Humphreys, J. and M. Plummer (1995), “The Economic Impact on the State of Georgia of Hosting the 1996 Olympic Games”, Selig Center for Economic Growth, Atlanta.

Independent Transport Commission (2014), “Ambitions and Opportunities: Understanding the Spatial Impact of High Speed Rail”, London, November.

Kang, H.-W. (2007), The Impacts of Beijing 2008 Olympic Games on China’s Economy, Department of Economics, National Chengchi University, Taipei, July.

Kasimati, E. and P. Dawson (2009) “Assessing the Impact of the 2004 Olympic Games on the Greek Economy: A Small Macroeconometric Model”, Economic Modelling, Vol. 26/1, pp. 139-146.

Katsuda, T. (2014), “Local Employment Development: Japan’s Experience”, Employment Security Bureau, Ministry of Health, Labour and Welfare, Tokyo, 14 November.

Kavetsos, G. and S. Szymanski (2009), “National Well-Being and International Sports Events”, Journal of Economic Psychology, No. 31, pp. 158-171.

Kundu, A. (2014), “Urbanisation and Economic Development in Asia”, in OECD (2014a), OECD Regional Outlook: Regions and Cities: Where Policies and People Meet, OECD Publishing, Paris,

Kveiborg, O. (2013), Economic Effects of Large-Scale Infrastructure Projects, BeltTrade, Copenhagen.

Lee, C. and T. Taylor (2005), “Critical reflections on the Economic Impact Assessment of a Mega-Event: The Case of the 2002 FIFA World Cup”, Tourism Management, No. 26, pp. 595-603.

Leff, S. and B. Petersen (2015), Beyond the Scorecard: Understanding Global City Rankings, Chicago Council on Global Affairs, Chicago, May.

Lippert, S. (2013), “Japan’s Mega-project to Link Tokyo and Osaka with 505 km/h Train”, World Review, 16 December, (accessed 15 July 2015).

Lutz, W. and Q. Ren (2002), “Determinants of Human Population Growth”, Philosophical Transactions of the Royal Society B: Biological Sciences, No. 357, pp. 1 197-1 210.

Lutz, W., M. Testa and D. Penn (2006), “Population Density is a Key Factor in Declining Human Fertility”, Population and Environment, No. 28 (January), pp. 69-81.

Lützeler, R. (2011), “Left Behind in the Global City: Spaces and Places of Ageing and Shrinking in the Tokyo Metropolitan Area”, in F. Coulmas and R. Lützeler (eds.), Imploding Populations in Japan and Germany: A Comparison, Koninklijke Brill, Leiden.

Madden, J. R. (2006), “Economic and fiscal impacts of mega sporting events: A general equilibrium assessment”, Public Finance and Management, Vol. 6/3, pp. 246-394.

Madden, J. R. (2002), “The economic consequences of the Sydney Olympics: The CREA/Arthur Andersen Study”, Current Issues in Tourism, Vol. 5/1, pp. 7-21.

McCann, P. (2007), “Globalization, Trade and the Changing Geography of New Zealand: Urban versus Rural or Centre versus Periphery?”, PowerPoint Presentation, 51st Conference of the Australian Agricultural and Resource Economics Society, 13-16 February, Queenstown, New Zealand.

Ministry of Education, Culture, Sports, Science and Technology (2015), “Result of the Survey on Earthquake-Resistant Renovation of Public School Buildings”, press release, Ministry of Education, Culture, Sports, Science and Technology, 2 June (in Japanese), (accessed 6 August 2015).

Ministry of Education, Culture, Sports, Science and Technology (2014), “White Paper on Science and Technology 2014”, Ministry of Education, Culture, Sports, Science and Technology, Tokyo, June.

Ministry of Health, Labour and Welfare (n.d.), “Basic Survey on Wage Structure”, (accessed 15 July 2015).

Minami, S. and Yagi, S. (2013), “Rough Road Ahead for Linear Shinkansen”, Yomiuri Shimbun, 20 September.

Mizuho (2014), “The Economic Impact of the 2020 Tokyo Olympic Games”, Mizuho Research Institute, 17 October.

Ministry of Land, Infrastructure, Transport and Tourism (2015), Brochure of the revised Regional Public Transport Revitalisation Act, 8th edition, Ministry of Land, Infrastructure, Transport and Tourism, July 2015, available at:

Moretti, E (2012), The New Geography of Jobs, Houghton Mifflin Harcourt.

Mori Memorial Foundation (2014), “Global Power City Index 2014”, Institute for Urban Strategies, Mori Memorial Foundation, October, available at:

Morichi, S. and T. Shimizu (2002), “An Analysis of the Effect of High-Speed Railway on Inter-Regional Migration and Traffic Flow in Japan”, Proceedings of International Conference on Inter-city Transportation, Vol. 1, pp. 131-139.

Nagata, K. (2015), “Tokyo to Open ‘One-Stop’ Office to Help Foreigners Launch Businesses”, Japan Times, 6 March 2015, (accessed 1 August 2015).

OECD (2015a), Governing the Metropolitan City of Venice, OECD Publishing, Paris,

OECD (2015b), “Metropolitan areas”, OECD Regional Statistics (database), (accessed 10 September 2015).

OECD (2015c), Ageing in Cities, OECD Publishing, Paris,

OECD (2015d), The Metropolitan Century: Understanding Urbanisation and Its Consequences, OECD Publishing, Paris,

OECD (2015e), Governing the City, OECD Publishing, Paris,

OECD (2015f), OECD Economic Surveys: Japan, OECD Publishing, Paris,

OECD (2014a), Major Transport Infrastructure Projects and Economic Development, ITF Round Tables, No. 154, OECD Publishing, Paris,

OECD (2014b), The Economics of Investment in High-Speed Rail, ITF Round Tables, No. 155, OECD Publishing, Paris,

OECD (2013a), Green Growth in Kitakyushu, Japan, OECD Green Growth Studies, OECD Publishing, Paris,

OECD (2013b), Investing Together: Working Effectively across Levels of Government, OECD Publishing, Paris,

OECD (2013a), How’s Life? 2013 – Measuring Well-Being, OECD Publishing, Paris,

OECD (2012a), Redefining “Urban”: A New Way to Measure Metropolitan Areas, OECD Publishing, Paris,

OECD (2012b), OECD Territorial Reviews: Skåne, Sweden 2012, OECD Publishing, Paris,

OECD (2012c), Compact City Policies: A Comparative Assessment, OECD Green Growth Studies, OECD Publishing, Paris,

OECD (2011), Demand-Side Innovation Policies, OECD Publishing, Paris,

OECD (2010a), “Local Development Benefits from Staging Global Events: Achieving the Local Development Legacy from 2012 - A Peer Review of the Olympic and Paralympic Legacy for East London”, OECD Local Economic and Employment Development (LEED) Working Papers, No. 2011/01, OECD Publishing, Paris,

OECD (2009a), OECD Economic Outlook, OECD Publishing, Paris,

OECD (2009b), How Regions Grow: Trends and Analysis, OECD Publishing, Paris,

OECD (2008), OECD Economic Surveys: Japan 2008, OECD Publishing, Paris,

Official Statistics of Japan (2015), 2013 Housing and Land Survey, (accessed 15 July 2015)

Okata, J. and A. Murayama (2011), “Tokyo’s Urban Growth, Urban Form and Sustainability”, in A. Sorensen and J. Okata (eds.), Megacities: Urban Form, Governance and Sustainability, Springer, New York.

Overman, H. (2011), “HS2: Assessing the Cost and Benefit”, CentrePiece, Vol. Winter 2011/12, pp. 18-20.

Overman, H. et al. (2009), Strengthening Economic Linkages between Leeds and Manchester: Feasibility and Implications: Full Report, The Northern Way, Newcastle upon Tyne, available at:

Oxford Economics (2012), The Economic Impact of the London 2012 Olympic & Paralympic Games: A Report by Oxford Economics Commissioned by Lloyds Banking Group, Oxford, July, available at:

Preuss, H. (2004), The Economics of Staging the Olympics: A Comparison of the Games 1972–2008, Edward Elgar, London.

Puga, D. (2012), “European regional policies in light of recent location theories”, Journal of Economic Geography, Vol. 2, pp. 373-406.

Quintini, G. (2011), “Right for the Job: Over-Qualified or Under-Skilled?”, OECD Social, Employment and Migration Working Papers, No. 120, OECD Publishing, Paris,

Rathman, G.B. (1999), “‘Biotechnology Startups’ in Springham”, in D.G. and Moses, V. and Cape, R.E. (eds.) Biotechnology - The Science and the Business, Taylor & Francis.

Renn, A. (2012), “What is a Global City”, New Geography, 12 July, (accessed date 15 July 2015).

Rose, A. and M. Spiegel (2009), “The Olympic Effect”, National Bureau of Economic Research Working Paper, No. 14 854, Cambridge, April, available at:

Shapiro, A. (2014), “Did London Get an Economic Boost from the 2012 Olympics?”, Parallels: NPR, 3 February, (last accessed 15 July 2015).

Saito, J. (2013), “The Olympic Games to Be Held in a Super Aged Society”, Japan Center for Economic Research, Tokyo, 3 October.

Sassen, S. (2005), “The Global City: Introducing a Concept”, Brown Journal of World Affairs, Vol. IX, No. 2, Winter/Spring.

Sassen, S. (2001), The Global City: New York, London, Tokyo, 2nd edition, Princeton University Press, Princeton.

SOCOG (2001), Sydney Organising Committee for the Olympic Games, (accessed 15 December 2012).

SOOC (1989), “Seoul Olympic Organizing Committee Official Report”, Korea Textbook Co., Ltd, Seoul, available at:

Staton, T. (2015), “The top 15 pharma companies by 2014 revenue”, FiercePharma, No. 18 March, (accessed on 1 August 2015).

Storper, M. (2009), “Local economic development: Some cities develop more than others: Specialization, Human Capital and Institutions”, CAF Working Paper, 2009/08.

Storper, M., T. Kemeny and N. Makarem (2015), The Rise and Fall of Urban Economies: Lessons from San Francisco and Los Angeles, Stanford University Press.

Stott, I. et al. (2015), “Land Sparing is Crucial for Urban Ecosystem Services”, Frontiers in Ecology and the Environment, Vol. 13/7, pp. 387-393,

Sutherland, D. et al. (2009), “Infrastructure Investment: Links to Growth and the Role of Public Policies”, OECD Economics Department Working Paper, No. 686, OECD Publishing, Paris,

Telegraph (2012), “Tourist Spending Spree at London 2012 Olympics Boosts UK Economy”, The Telegraph, 12 October, (accessed 15 July 2015).

Tian, X. (2011), “The Causes and Consequences of Venture Capital Stage Financing” Journal of Financial Economics, Vol. 101/1, pp. 132-159,

UNDESA (2014), World Urbanization Prospects: The 2014 Revision, Population Division, United Nations Department of Economic and Social Affairs, June.

Vickerman, R. and A. Ulied (2009), “Indirect and Wider Economic Impacts of High Speed Rail”, in G. de Rus (ed.), Economic Analysis of High Speed Rail in Europe, Fundación BBVA, Bilbao, pp. 89-118.

Whiting, R. (2014), “Negative Impact of 1964 Olympics Profound”, Japan Times, 24 October.

World Bank Group’s website (n.d.), (accessed on 1 August 2015).

Yglesias, M. (2014), “Manhattan is less dense today than it was in 1910”,, 23 September,, (accessed 15 July 2015).

Zhang, L. and S. Zhao (2009), “City Branding and the Olympic effect: A Case Study of Beijing”, Cities, No. 26, pp. 245-254.

Zimbalist, A. (2012), International Handbook on the Economics of Mega Sporting Events, Edward Elgar, London.


← 1. The comparison here refers to the cities themselves, not to the functional urban areas (FUAs), which have far lower densities. Indeed, only 4 FUAs in the OECD area have population densities above 4 000 for the entire FUA. These are Seoul/Inchon, Busan and Changwon in Korea, as well as Tokyo (Japan).

← 2. The reverse might be argued and often has been that the urban mega-region in central Honshū – and Tokyo, in particular – tends to suck money and talent out of the rest of the economy. However, the fact that concentration is not accompanied by rising wage disparities and a growing productivity gap suggests that rural-to-urban migration is driven by push factors in the regions more than the pull of the capital.

← 3. This proxy is valid under the assumption that the local labour market is competitive, if this is the case, employee’s wages reflect their productivity.

← 4. The analysis here covers only those countries with metro-level patent data available and more than 2 functional urban areas with populations above 500 000.

← 5. Kantō, based on Tokyo; Keihanshin, around Osaka and Kyoto; Chūkyō, centred on Nagoya.

← 6. See Lutz and Ren (2002) for an overview of the literature.

← 7. Female literacy is by far the biggest factor depressing fertility, as it simultaneously offers women new opportunities outside the home and makes it easier for them to learn how to manage their reproductive careers.

← 8. It is noteworthy that if one excludes the most prominent outlier, Okinawa, from the figures, the slope of the line steepens and the r-squared reaches almost 40%.

← 9. Estimates here refer to the FUAs, as defined in OECD (2012a), rather than to the cities defined as administrative units.

← 10. PricewaterhouseCoopers’ “Cities of Opportunity”; A.T. Kearney’s “Global Cities Index”; the EIU’s “Hot Spots 2025”; Mori Memorial Foundation’s “Global Power Cities Index”; Z/Yen Group’s “Global Financial Centres Index 17”; fDi Intelligence’s “Global Cities of the Future”; Grosvenor’s “Resilient Cities”; Arcadis’ “Sustainable Cities Index”; and the Reputation Institute’s “CityRepTrak”.

← 11. Hong Kong, China; London, New York, Paris, Singapore and Tokyo turn up near the top of almost all of the ratings. A few other OECD cities (Chicago, Seoul, Stockholm, Sydney and Toronto) turn up more sporadically. There are 90 “slots” available in the top-ten rankings reviewed, but only 37 cities are ranked; 6 are found in at least 5 top-ten lists. In all, 50% of the 90 slots were held by just 8 cities. Beijing and Shanghai appear, twice and once, respectively. Singapore and Hong Kong appear in the top 10 on 7 of the 9 lists, but other cities outside the OECD area occupy only 5 of the 90 slots.

← 12. The main exception being the Mori Memorial Foundation.

← 13. The zones have been created in the framework of the 2013 Japan Revitalisation Strategy; they are discussed in greater detail in Chapter 4 below.

← 14. Building standards are particularly important: the 1980 building code addresses the seismicity of buildings and the evidence suggests that it makes a big difference – during the 1995 Kobe earthquake, it was older buildings that tended to collapse.

← 15. The old-age share of the population in public rental housing is far higher than the average in Tokyo, having risen from 18.6 to 32.2% over 1995-2005.

← 16. The theory does not encompass the housing market, but such outcomes are consistent with it, if not predicted by it.

← 17. The 2013 figure was close to 100 000, with 15-29 year-olds accounting for over 80% of the inflow.

← 18. The Central Circular Route, the Tokyo Outer Ring Road and the Metropolitan Intercity Expressway.

← 19. In April 2015, a maglev train carrying 49 Japan Railways employees reached a world-record speed of 603 km/h, travelling 1.8 km in 11 seconds. The new record came less than a week after the train had reached 590 km/h, breaking its own 2003 record of 581 km/h.

← 20. The corresponding figures for Osaka and Nagoya were, respectively, 3.7 m metres2 (+52%) and 2.1 m metres2 (+79%).

← 21. See, for example, the 2014 Global Power City Index produced by the Mori Memorial Foundation: Tokyo

← 22. At present, there are two commercial maglev systems are in operation: the high-speed Transrapid system in the Pudong district of Shanghai, China, and the low-speed “Linimo” line in Nagoya, built for the 2005 World Expo. China and Korea are both building low-speed maglev lines of their own design, in Beijing and at Seoul’s Incheon airport, respectively.

← 23. On a visit to the United States in 2013, Prime Minister Shinzō Abe observed that the technology could connect the centres of Washington, DC, and New York City in less than an hour (Lippert, 2013).

← 24. That said, studies point to the importance of adequate investment in the maintenance of public infrastructure. Such investment may be less visible – and thus less attractive to elected politicians – but it can yield important returns.

← 25. See also Feigenbaum (2013) on the difficulty of making a “purely commercial” case for HSR.

← 26. This addresses two drawbacks of previous models. Most could not address the spatial interaction that exists within regional transport systems, while others took spatial interactions into account using a partial equilibrium framework rather than a general one.

← 27. McCann (2007) observes that IT interactions tend to be most intensive among co-workers who also have the most frequent face-to-face interactions.

← 28. The idea of using it for the Glasgow Commonwealth Games in 2014 was also floated.

← 29. If stadiums were good business, the private sector would supply them.

← 30. This agreement was negotiated and signed by Canada, the province of British Columbia, the city of Vancouver, the Resort Municipality of Whistler, the Canadian Olympic Authority, the Canadian Paralympic Committee, and the Vancouver 2010 Bid Corporation.

← 31. MIC’s social and life basic survey H18 – update.