This chapter provides an overview of recent trends on transport infrastructure development in European and OECD member countries and reviews the main benefits associated with it. It starts with a description of how transport infrastructure investment has changed across countries and regions, both in terms of the magnitude and its composition, paying particular attention to differences between urban and rural regions. Next, it moves onto reviewing the main benefits associated with an efficient transport system, all while making suggestions on how current evaluation tools, such as cost-benefit analysis, could be improved to account for all of them. It emphasises the need for different metrics that can better capture the potential of transport infrastructure improvements and the importance of taking a “more aggregate” view when evaluating investment projects that can account for potential displacement of economic activity. The main benefits are discussed in reference to the type of investment: interregional highway and railway – both traditional and high-speed – development, highway development in the proximity of cities and infrastructure development within cities.
Transport Bridging Divides
1. Transport infrastructure trends and regional development
Abstract
Transport infrastructure was, is and will be important for regional development
Transport infrastructure and economic development
Extensive and efficient transport infrastructure is essential for well-functioning economies and the development of regions and cities. When designed effectively, transport networks can be an engine for productivity and improved quality of life for citizens. “Effective modes of transport – including high-quality roads, railroads, ports, and air transport – enable entrepreneurs to get their goods and services to market in a secure and timely manner and facilitate the movement of workers to the most suitable jobs” (World Economic Forum, 2016, p. 35[1]). For instance, Chile has rolled-out many key investments in its basic infrastructure backbone that are essential for economic development and welfare. This has led to improvements in living standards and Chile’s gross domestic product (GDP) per capita has increased from USD 4 787 in 1990 to USD 22 197 in 2015 (OECD, 2017[2]).
Transport infrastructure investment has always been a fundamental engine of economic development. The facilitating role of transport infrastructure with respect to trade for instance can be traced back in history. A producer in New York in the late 18th century was bound to sell primarily to consumers on the East Coast of the United States. However, less than a century later the same producer had the opportunity to sell to a consumer living in Los Angeles thanks to the railroad network built in the United States between 1790 and 1870.1
Transport infrastructure investment creates economic growth through many different channels. The most basic among them is that transport infrastructure facilitates the exchange of goods. Improved transport infrastructure reduces the cost of trade. Better domestic trade opportunities allow regions to specialise in the sector where they are the most competitive relative to the others. This holds irrespective of the type of regional economy, be it more urban or rural. Beyond the domestic borders, better foreign trade opportunities can unlock not just regional but even countrywide export-driven economic growth. This creates benefits for the exporting firms within a region but also wider benefits if the exported good is integrated into local, regional or global value chains.
Transport infrastructure allows regions and cities to leverage benefits from agglomeration and concentration by expanding commuting opportunities for their workers. This creates benefits for places and for workers who can access better-matching and better-paid jobs without bearing the burden of moving to a different place. Intra-urban and suburban transport infrastructure serves to integrate rural regions into the local labour market of the cities located in their proximity, thereby creating a greater variety in job opportunities and raising the living standards of their inhabitants. For instance, in 1990, the average commute in Korea was less than ten kilometres. Following a round of investment in transport infrastructure, 20 years later, this distance had increased by 30% to 13 kilometres while the time an average Korean worker spent commuting decreased by about one-quarter (OECD, 2016[3]). Similarly, the number of daily commuters between the southern Swedish region of Skåne and Copenhagen in Denmark rose roughly sevenfold to around 20 000 per day after the opening in 2000 of the Øresund bridge (OECD, 2016[4]). Beyond commuting, distances that can be travelled in 2-3 hours allow for business-related day trips that can enhance business interactions, in particular between urban centres. Since 1994, close to 430 million passengers have crossed the Channel tunnel connecting London in the United Kingdom to Paris in France across all transport modes (e.g. the Eurostar train or the Eurotunnel Shuttles) (GETLINK, 2019[5]).2
Transport infrastructure brings firms closer to a larger customer base and a larger pool of workers, which can stimulate hiring and investment by local firms. For instance, a firm that gains access to a broader market thanks to the reduction in transport costs that accompanies improved transport infrastructure might decide to invest more resources to enhance its competitiveness. Alternatively, a firm facing an increase in demand might choose to tap into its unutilised capacity and hire more local workers in order to serve an expanding market. An increase in production will cause an increase in the density of local economic activity further reinforced by productivity spill-overs among neighbouring firms.3
Investing in transport infrastructure
In general, OECD member countries have substantial needs for new investment in transport infrastructure as well as upgrading existing infrastructure. Governments increasingly face tough decisions about where to locate or maintain public investments as resources become scarcer and investment needs multiply. It is increasingly important that service and policy restructuring decisions reflect the diversity of needs and circumstances facing urban and rural communities, and try to maximise the value for money that investment can provide in each context. However, the quality of existing public infrastructure has deteriorated and public infrastructure stock has started to drop in many European countries (CEB, 2017[6]).
Transport infrastructure investment remains one of the key decisions taken by policymakers and accounts for a large fraction of OECD countries’ budgets. In 2016, the average total inland transport infrastructure investment amounted to roughly 1% of GDP across member countries of the OECD International Transport Forum (ITF) (Figure 1.1).4 Transport infrastructure investment peaks during times of major transport infrastructure project execution, this being one of the reasons behind the dispersion in the 2000-16 difference in transport infrastructure investment (as a percentage of GDP). There are other potential reasons, e.g. the change in construction cost. In the United States, for instance, state spending per kilometre rose fivefold between 1960 and 1980 during the construction of Interstate Highways, the most significant infrastructure network in the United States. Increasing demand for transport infrastructure as income levels rose accounts for a large part of that increase (Brooks and Liscow, 2019[7]). Another reason behind the variation in spending as a percentage of GDP is, of course, fluctuation in GDP. For instance, Greece registered a high positive variation in transport infrastructure investment as a percentage of GDP during 2000-16, primarily because its GDP contracted substantially over the same period.
Large infrastructural projects continue in both the developed and the developing world. Some examples are the Trans-European Transport Network in Europe (Box 1.2), to the Golden Quadrilateral in India and Interoceanic Highway in South America. The government of the People’s Republic of China (China hereafter) has been leading a global development strategy, the Belt and Road Initiative, consisting of several transport infrastructure projects with the ambition to connect 65 countries representing over 30% of world GDP (World Bank, 2018[9]).
Transport infrastructure investment is not the only source of spending on transport infrastructure. Countries spend a large amount of money every year on maintaining it. For 11 of the 26 OECD member countries with available data, maintenance spending for inland transport infrastructure was again half the amount spent on infrastructure investment in 2016. For Italy, Latvia and Slovenia, maintenance spending even exceeded investment.5
Large infrastructural projects require buy-in from a large number of national and regional partners. This is the case for the abovementioned Belt and Road Initiative, as well as for the European Transport Network, where a large share of spending takes place at the supranational level. For instance, during the 2014-20 programming period, about 22% of the European Regional and Development Fund (ERDF) and the Cohesion Fund is being invested in network infrastructure in transport and energy.6
Transport infrastructure investment has been severely hit by the Great Recession thus leading to the emergence of significant funding gaps. Transport infrastructure spending still accounts for a significant share of countries’ budgets but the current trajectory points to a shortfall of about USD 350 billion a year (excluding spending due to maintenance backlogs), a gap that triples after considering the extent of transport infrastructure investment needed to meet the United Nations (UN) Sustainable Development Goals (McKinsey & Company, 2016[10]). As countries are looking for strategies to bolster their economies following the ongoing COVID-19 pandemic, transport investment provides an opportunity to move towards greening cities. Investment in electric vehicle charging infrastructure is a key opportunity for recovery packages, both for private vehicles and electrified public transport such as buses (OECD, 2020[11]).
Transport infrastructure, particularly public transport, are delivered and used at the local level. For instance, across Europe, between 2009 and 2015, local and state (regional) governments represented on average 53.6% of total public capital investment (CEB, 2017[6]). The level of investment depends on the type of infrastructure. According to the CEB (2017[6]), subnational governments in Europe tend to invest more in social infrastructure (education, health and community services). In economic infrastructure (transport, amenities and telecommunication), local and state (regional) government contributed 37.1% and central government 62.9% of total investment (CEB, 2017[6]). In OECD countries, subnational governments represented on average 59.3% of total public investment and 3.1% of GDP (Allain-Dupré, Hulbert and Vincent, 2017[12]). There are certainly variations across countries depending on their levels of decentralisation and political system. In particular, across the OECD, 39.2% of total investment by subnational governments is located to economic affairs, accounting for 0.73% of GDP in 2014 where the single most important item by far is transport (Allain-Dupré, Hulbert and Vincent, 2017[12]).
There was a decline in subnational infrastructure investment across European Union (EU) and OECD countries following the global financial crisis of 2007‑08 (Allain-Dupré, Hulbert and Vincent, 2017[12]; CEB, 2017[6]). According to OECD studies, the decline in subnational public investment has been particularly marked in the EU. Public investment conducted by subnational governments across EU countries dropped by almost 18% between 2009 and 2014 (i.e. 5% per year in real terms) (Allain-Dupré, Hulbert and Vincent, 2017[12]). The decline varies across countries and regions, and sectors. In general, between 2010 and 2015, economic affairs (where transport infrastructure is the most important component) has been largely spared by investment cuts as capital expenditures remain almost stable in real terms. However, subnational transport investments declined in Greece, Ireland, Italy, Lithuania and Spain. By contrast, subnational government investment in transport has been rising in Scandinavian countries and in Central European countries that benefit from EU structural and cohesion funds (Allain-Dupré, Hulbert and Vincent, 2017[12]).
Underinvestment in transport infrastructure across regions and cities could limit the possibilities of economic growth and social development. In the United States, for instance, public transport is an important factor in the Chicago Tri-State Metro Region’s attractiveness but investment in the system has not kept up with the needs of a suburbanising population, which has led to road congestion and emissions (OECD, 2012[13]). Governance is an important factor limiting the effectiveness of public transport in the metropolitan area, as fragmentation leads to challenges associated with a lack of interconnectivity, coherence across transit modes, regional freight planning, accountability, and implementation power for regional planning and transport objectives (OECD, 2015[14]). More broadly, investing in the construction and upgrading of transport infrastructure could help to improve the connectivity between rural and urban areas and boost local economies. However, according to some studies, there has been a long-term decline in the availability of infrastructure as the stock of public capital (closely related to infrastructure) as a share of output has fallen over the last three decades across the world (IMF, 2014[15]). The impact of the global financial crisis of 2007-08 has been deep and investment in infrastructure remains below pre-crisis levels in many countries (UNECE, 2016[16]; Allain-Dupré, Hulbert and Vincent, 2017[12]). To support investment in infrastructure, governments may have to prioritise subsectors where infrastructure is the poorest (i.e. the railway sector in Romania and the road sector in Ukraine). Returns on infrastructure investment are higher where current endowments are lower (UNECE, 2016[16]).
Current and future challenges in transport investment
Gains from transport infrastructure investment in some regions might be offset by losses in other regions. The construction of the National Trunk Highway System in China, for example, helped connected urban prefectures to become more attractive, whereas rural prefectures shrank in size (Baum-Snow et al., 2016[17]). From a public policy perspective, this means that any assessment of the overall benefits of transport infrastructure needs to account for trade‑offs and displacement. This section provides a broader description of the benefits of road and rail infrastructure and the potential negative externalities imposed on other regions by transport infrastructure investment in one region.
Current megatrends – demographic transition, climate change, digitalisation and automation – are creating additional pressures on transport networks (OECD, 2019[18]). The demand for transport infrastructure is expected to rise in response to increasing urbanisation rates.7 Supply of transport infrastructure will need to keep pace if urbanisation is to take place without a slowdown in economic growth or a loss in living standards.8 Even the current COVID-19 pandemic is unlikely to reverse mounting pressures. Public transport networks play a crucial role in helping cities adapt to megatrends. As the lockdowns across OECD member countries relaxed following the first wave of infections, travel resumed quickly but public transport use remained (significantly) below pre-crisis levels, e.g. in London, there was less than half the number of daily riders on the underground on every single day between April and October 2020 than during comparable days in prior years.9 Increased teleworking accounts for part of this decline but a relative shift towards other individual modes of transport is also evident.
Future investment in transport infrastructure must address concerns about environmental sustainability. As countries are starting to commit to net-zero carbon emissions, an increasing number of green and innovative mobility solutions are becoming available and need to be integrated into transport planning. It might even be necessary to completely rethink and redesign mobility systems around accessibility (OECD, 2019[19]).
Technological change implies a redefinition of infrastructure investment needs. On the one hand, digitalisation of society demands investments in broadband infrastructure and the expansion of data centres capacity. Estimates by the European Commission (EC) and the European Investment Bank (EIB) suggest that additional investment worth approximately EUR 55 million a year is required to meet the targets of the EU digital agenda (European Parliament Research Service, 2018[20]). On the other hand, the changing landscape of urban mobility implies the need to convert existing infrastructure, e.g. parking space.
Better data can help steer policy decisions. For example, the use of high capacity vehicles for freight traffic should be encouraged in light of the data that the information and communication technology (ICT) on board these vehicles make it possible to collect. High capacity vehicles are freight trucks that are heavier or longer (or both) than vehicles currently permitted on the general road network. It is estimated that the costs of adapting the existing highway network to the circulation of such vehicles are modest especially when compared to the benefits in terms of reduction of the cost of moving goods and energy demand. Most importantly, these vehicles usually come equipped with in-vehicle sensors that, matched with road-sensors, allow the collection of real-time and fine-grained tracking data that can be used as inputs in traffic management systems seeking to reduce traffic and carbon emissions (ITF, 2019[21]).
The benefits of transport infrastructure investments may increasingly accrue through complementarities and synergies among a cluster of assets rather than stand-alone projects. Such complementarities are easier to find and match at the regional and local levels. With many of the large infrastructures complete, future transport investments will involve extensions to or linkages with other existing infrastructures. Intermodality would need to be further promoted as it could facilitate the movement of goods and people across transport modes and developing the so-called “last mile” infrastructure. This requires better co-ordination at the planning phase and complex interactions at the implementation stage.
To improve the actual access to opportunities that transport infrastructure can provide, it is necessary to increase the investment in transit to increase travel capacity. Transit is a set of technologies (trains, underground, light rail, bus rapid transit and regular buses) that differs from private means of transport. Increasing travel capacity has high fixed costs and they become more advantageous at a high enough population density; thus, investments have to be strategic. For instance, the high costs of the metro systems relative to buses make metros attractive only for the parts of the city with a sufficiently high population density. Moreover, the experience of developed and developing cities shows that fostering accessibility is more than increasing travel capacity and investing in transit technologies. It is a combination of urban policies aimed at easing access to destinations without having necessarily to use public transport and at a low travel cost (Duranton and Guerra, 2016[22]). As the experience of London suggests, every city and every part of the city is unique and requires tailored transport solutions to support growth and socio-economic development (Greater London Authority, 2018[23]).
Solid evidence and evaluation are crucial to tackling the challenges policymakers face when taking transport investment decision. This might require rethinking traditional ex ante evaluation methods such as cost-benefit analysis (CBA), by adopting more comprehensive evaluation criteria to assess the admissibility of potential transport infrastructure projects as a function of their policy objectives (Box 1.1). For instance, a quantitative assessment of the impact of transport infrastructure investment on employment and productivity at different geographical scales is often beyond the scope of standard CBA, which is currently the most widely adopted evaluation tool of transport infrastructure projects.10 The list of costs considered in a CBA can also be expanded depending on the policy objective of the project. For instance, given that transport infrastructure investment can disproportionately benefit high-income households, the increase in inequality should be added to the costs of the project if one of its implicit goals is to ensure that the benefits are more equally distributed among the population.
Box 1.1. Externalities and cost-benefit analysis
Among the channels that tools such as CBA traditionally consider are the “cash in‑flows directly paid by users for the goods or services provided by the operation, such as charges borne directly by users for the use of infrastructure, sale or rent of land or buildings, or payments for services” (EC, 2014[24]). For example, the price charged for a train ticket is among the revenues and therefore benefits associated with rail investment. Benefits typically accounted for in CBA can also be non-financial, such as savings in travel time, the reduced cost of accidents and environmental externalities.
Despite the long list of channels already accounted for, CBA can gain from the inclusion of further benefits. For example, the impact that transport infrastructure investment has on capital investment, employment or productivity in the regions connected to new infrastructures. The upcoming section provides a more detailed discussion of these benefits. CBA should also account for externalities that investment creates beyond directly connected places. This is because transport infrastructure investment does not just have an impact on the regions it crosses but one that often extends onto neighbouring regions. Moreover, positive consequences for one region can also come at the expense of others (Baum-Snow et al., 2016[17]). Hence, in the end, whether there is an aggregate net gain depends on the individual case.
Source: EC (2014[24]), Guide to Cost-Benefit Analysis of Investment Projects, European Commission; Baum-Snow, N. et al. (2016[17]), “Highways, market access, and urban growth in China”.
Investing in quality and maintenance
There are marked differences in the availability and quality of infrastructure across countries but differences are being bridged. For instance, in the EU, the new member states have been catching up with older members (UNECE, 2016[16]). However, the perception of the quality of infrastructure remains very low in many of those countries. According to the Global Competitiveness Report’s Executive Opinion Survey, in OECD member countries in general, the perceived quality of railroad infrastructure (4.4 out of 7) is relatively lower to that of road infrastructure (5 out of 7) (Figure 1.2). This suggests that roads are more reliable than railroads. France and Japan are the two OECD countries ranked among the top three of high-quality perception of both roads and railroad infrastructure. The perception of the quality of road infrastructure in countries such as Austria, Belgium, Mexico, New Zealand and Norway is lower than the OECD average. Meanwhile, Japan and Switzerland have the highest quality perception of railroad infrastructure.
The road and railroad networks are two important subsectors of the transport network. Investment in their expansion and maintenance is critical. The sheer size of many transport networks has made them increasingly costlier to maintain. Many countries have accumulated very large gaps in the required maintenance work to their transport infrastructure network so that maintenance backlogs will absorb a considerable amount of resources invested in transport infrastructure (McKinsey & Company, 2016[10]). When roads do not meet quality standards the average speed decreases and the safety is compromised. For instance, in Ukraine, where road quality perception is rather low, the speed on highways is one-third to one-half of what it is in Western Europe; and fatality rates per 100 000 inhabitants reached 11.3 in 2013, well above the OECD average of 6.8 (OECD, 2018[25]). In Romania, the road network is of major importance for local and regional development. However, although the road network covers the entire network of communes and cities, the road infrastructure is of poor quality and needs extensive maintenance and upgrading. For instance, in the central region, of the 4 696 km of county roads, only 42% have electricity and 33% are cobbled or dirt roads. Moreover, there are over 4 000 km of communal roads that connect the centre and villages. However, only 10% have light duty coating and half of it is outdated; in addition, 80% of communal roads are cobbled or dirt roads.
In many countries, the railway system is the backbone of long-distance freight transport. Underinvestment in railways could represent significant setbacks for economic development. In some East European countries, railway infrastructure requires considerable modernisation. Also, freight shipments use the same rail line, decreasing considerably the average speed of passenger trains; and the average age of locomotives and passenger cars is above 40 years. In Romania, the railway infrastructure suffers from deterioration and the railway fleet is obsolete and inadequate to meet current requirements. For example, at the national level, 4 000 km of railway (30% of the network) have been in need of rehabilitation for more than a decade. This has led to an increase in dangerous points and 44% increase in the length of the railway network affected by speed restrictions, to the detriment of the quality of the services offered. The average travel speed value achieved by passenger trains, taking into account standing at stations, is between 28-58 km/h. For express trains, the average speed is only 20 km/h higher. The low quality of railroad infrastructure means that most of the freight is done by road, which in many cases are secondary congested roads.
Shifting priorities and catching up in transport infrastructure investment
Trends in transport infrastructure investment
Investment in road and rail infrastructure has increased in many countries. In 2016, total inland transport infrastructure investment as a share of GDP was higher than in 2000 in half of the countries for which data are available. For most countries, investment increased to drive this expansion. For instance, the United Kingdom, total inland transport infrastructure investment nearly doubled between 2000 and 2016, despite GDP growth of 33% over the whole period. There are exceptions: in Greece, the increase in investment intensity came through a contraction in GDP between 2000 and 2016.
The composition of total inland transport infrastructure investment changed between 2000 and 2016, shifting towards rail investment. In 2000, the share of total transport infrastructure investment on roads across 18 countries was 62% while, in 2016, this figure was down to 58%. At the same time, the share in railway investment has increased from 28% to 31%. The shift has been quite heterogeneous among countries: while France raised its share of rail investment by 26 percentage points, Poland saw it contracting by 6 percentage points. Overall, the increase in rail investment was driven by Western European countries, while East European ones saw an expansion in total inland infrastructure investment driven primarily by road investment. These differential trends between East and Western European countries should also be interpreted in light of the different starting points, and the need for East European countries to catch up with average road length in the rest of Europe (Figure 1.3).
The shift towards railway spending in the past two decades occurred alongside additional highway investment. The European highway network expanded considerably between 1990 and 2012 (Figure 1.4). The highway network is part of the road network that comprises also motorways, secondary or regional roads. In some countries, the expansion in the highway network was mostly completed well before the 2000s: for example, most of the expansion in the Spanish highway network took place in the early 1990s. In others, especially the East European countries, the development of the highway network started later and it is still underway.
From road to rail
Between 2000 and 2014, the level of per capita CO2 emissions grew less in countries with higher per capita investment in inland transport infrastructure (Figure 1.5). This might seem counterintuitive since transport accounts for 30% of CO2 emissions in OECD member countries (ITF, 2019[27]). Part of the explanation is the fact that countries that invested more resources in inland transport infrastructure and also expanded their share of rail investment between 2000 and 2014. The railway is in fact among the most energy-efficient and lowest-emitting transport modes (IEA, 2019[28]).
The COVID-19 crisis has left rail operators in financial trouble. As lockdown measures restricted interregional and international travel, railway companies saw demand drop dramatically. In Canada, domestic and international rail passengers in April 2020 were less than 2% of the number of passengers in April 2019.11 By July 2020, the number of passengers had risen again but remained at only about 16% of the passenger volume in July 2019. Even in Sweden, one of the countries with the least stringent lockdown restrictions, passenger volumes on trains were 20% or more below the prior year between March and June 2020, and about 10% lower since then.12 To buffer the shock on their rail sectors, OECD countries have taken steps early in the pandemic to ensure companies remain functional, spending billions in bailouts or absorbing losses in their balance sheets (where operators are already public).
Households in countries that invested more in infrastructure do not spend more on transport (Figure 1.6). The type of transport expenditure shifts in countries that invested more in infrastructure. Households in those countries tend to spend an increasing share of their transport expenditures on transport services (i.e. not going to the purchase and operation of personal vehicles). This suggests that in countries that invested more in transport infrastructure – and particularly in railway (Figure 1.3) – consumers shifted their spending away from private transport modes and towards transport services, without changing the overall share of spending going to transport.
Investment in metro and high-speed rail has grown substantially
Contrary to the evolution of conventional rail tracks, the length of metro and high-speed rail has grown substantially. Conventional rail tracks (as opposed to metros, trams or high-speed rail) make up 94% of all rail track-kilometres in Europe, North America, China and Japan, but their length has grown very little during 1995-2016 in most regions (IEA, 2019[28]). The most rapid growth has been in the length of the metro networks. In China, India, Japan, Russia, Europe and North America, the length of the metro rail network increased by 4 800 km between 2000 and 2017. China and India have played a determinant role, with 32 out of the 43 cities where a new metro system opened between 2010 and 2017 being located in Asia. The length of the high-speed rail network – the largest share of which is located in China, Japan, Korea and Europe – expanded from approximately 30 000 km to about 70 000 km between 2010 and 2017, with China being responsible for most of the expansion (IEA, 2019[28]). The share of high-speed tracks located in China in 2010 was about 36%. By 2017, this share had risen to 63%. Some countries have seen neither an expansion nor an uptake of the rail network. In Canada and the United States, for instance, the uptake – as measured by the number of passengers per km – was mostly flat between 1995 and 2016, while it nearly doubled worldwide during this period (IEA, 2019[28]).
Rail investment supports the low-carbon transition
Transport investment should shift decisively towards rail, being among the most efficient and lowest-emitting modes of transport when compared to road, maritime and air transport. Rail accounts for only 2% of energy use in the transport sector in spite of carrying 8% of motorised passenger transport and 7% of freight transport (IEA, 2019[28]). The rail share is projected to grow substantially in some of the scenarios elaborated by the International Energy Agency (IEA). For example, in the “high rail scenario”, a combination of changes in consumer habits coupled with a shift of transport investment towards rail transport is expected to be matched by an increase in passenger and freight activity by about 6 trillion passenger-kilometres and 3 trillion tonne-kilometres respectively relative to the baseline scenario.13
High-speed (HS) rail and urban transport are playing a pivotal role in the changing composition of transport investment needs. HS rail penetration varies widely across countries also because the profitability of this investment hinges on its ability to connect large metropolitan areas. On longer distances, the evidence from existing HS rail lines suggests that around 2 hours’ trip time (roughly 3 hours door-to-door), travellers start shifting towards air travel (ITF, 2014[29]). There is less consensus on the necessary population size in the catchment area of HS rail. The profitability of investment depends on a range of demand factors, including price, the connection between metropolitan areas and the wealth of the connected places. Even if there is an expected profitable business case, the high upfront cost of investment can make countries hesitant to shift a large share of their total annual investment into a small number of HS rail lines.
Box 1.2. The Trans-European Transport Network
In Europe, transport infrastructure investment has been a stronghold of EU investment policy since the early 1990s. Since the signing of the Maastricht Treaty in 1993, the Trans-European Transport Network (TEN-T) has been the tool of infrastructure policy in Europe and an important instrument to achieve cohesion and growth of European regions.
Under the current 2014-20 plan, transport infrastructure projects are divided into a core and a comprehensive network based on their priority, with priority projects typically being long-implementation, high-cost and cross-border. The core network is foreseen to be completed by 2030, the second by 2050. The core network is organised into nine main corridors (Figure 1.7) relying on transport via rail, airport and port. Rail transport occupies the centre stage with respect to each of these corridors, with some of the tracts being built ex novo and some others being renovated or upgraded. The estimated magnitude of funding required to complete the core network during 2021-30 amounts to EUR 500 billion, a figure that rises to EUR 1.5 trillion if the comprehensive network is also considered (EC, 2017[30]).
Rail investment occupies the largest share of TEN-T funding. In 2015, rails absorbed as much as 51.5% of TEN-T EU-level expenditures, followed by roads (30.6%), ports and motorways (9.2%), airports (5.5%), multimodal infrastructure (2.1%) and inland waterways (1.1%) (EC, 2017[31]). The situation was different in 1997 when roads and rail represented similar shares of TEN-T funding (Puga, 2002[32]). The increase in the share of budget absorbed by rail investment is partly explained by the shift towards high-speed (HS) rail. Despite the aggregate increase in HS rail investment, HS rail penetration was in 2016 still very heterogeneous among the European countries for which data are available (Figure 1.8). Among those European countries, France was the one with the highest share in either dedicated or upgraded railroad, close to 50%, followed by Italy and Spain.
Unlike traditional rail, HS rail – which is typically not suited for the transport of goods – benefits mostly passenger transport. HS rail investments can favour the economic activity of larger cities at the disadvantage of smaller urban centres (Puga, 2002[32]). For instance, multi-establishment firms may find convenient to relocate their headquarters to larger cities where they can benefit from the proximity to the service-producing sector since the reduction in travelling distance makes arm’s length relationships no more necessary. The resulting increase in agglomeration in larger cities tends to benefit especially the service-producing sector therein concentrated.
While the benefits may seem more pronounced for the service sector, the manufacturing sector can also benefit. HS rail indeed reduces the cost of search of suppliers and buyers for individual firms. Hence, firms located in the proximity of HS rail stations can better optimise the size and composition of their network of suppliers and buyers and be more productive (Bernard, Moxnes and Saito, 2019[35]).
Source: EC (2017[30]), Delivering TEN-T Facts and Figures, European Commission; EC (2017[31]), Progress Report on the Implementation of the TEN-T Network in 2014 and 2015, European Commission; Puga, D. (2002[32]), “Progress report on the implementation of the TEN-T Network in 2014 and 2015”, Journal of Economic Geography, Vol. 2, pp. 373-406; Bernard, A., A. Moxnes and Y. Saito (2019[35]), Production Networks, Geography and Firm Performance.
The still relatively scarce empirical evidence quantifying the benefits of HS rail holds back investment in the expansion of the network.14 For example, one significant challenge to HS rail network expansion is posed by the fact that, as the network expands, the set of new tracks that is profitable to develop shrinks or, in other words, the expansion of the network does not raise the profitability of other potential tracks sufficiently (ITF, 2014[29]). Traffic volume is an important driver of HS rail investment profitability, which means that, in each country, there exists just a limited set of potentially profitable connecting lines, typically those connecting large urban centres.15 Moreover, due to the substitutability between air travel and HS rail travel, the distance between endpoints capable of maximising revenues is likely to be between 500 km and 1 000 km, thus further reducing the set of potentially profitable connecting lines (ITF, 2014[29]).
Complementary actions can enhance the profitability of HS rail investment. Urban transport is an important complement to interregional HS rail. Strengthening local public transport systems improves access to HS rail stations. Often, inadequate feeding connections outweigh the timesaving benefits of HS rail (ITF, 2014[29]). For instance, in France, the HS rail network converges on Paris, whose multiple train stations each tend to serve a distinct axis of the network. Hence, without the well-developed metro system in the city centre of Paris, allowing fast travel from one station to another, the gains from the rollout of HS infrastructure in France might have been more modest. The acknowledgement of the complementarity of urban transport to HS rail investment is underscored by the Grand Paris Express project, a 38 billion-worth urban transport investment project in Île-de-France aiming at connecting strategic sites situated in the commuting zone of Paris metropolitan area, ranging from airports and TGV (Train à grande vitesse, HS train) stations to employment centres (Institut d'Aménagement et d'Urbanisme, 2013[36]).
Urban transport demands more investment. Despite the share of urban rail in Europe being relatively large (15% compared to 5% in China); further investment is required in European cities to: i) close the gaps in accessibility highlighted in Chapter 2; and ii) adapt urban transport to the needs of a low-carbon transition. Urban transport indeed represents a large share of passenger transport that in turn contributes to 50% of total transport-related emissions (OECD, 2019[37]). Yet, only 35% of the countries that signed the Paris Agreement in 2015 have included public transport in their climate action plans.
Investments favouring the local diffusion of knowledge can increase the return to investments in local connectivity. Investments in the local transport infrastructure help better connect cities with the surrounding rural regions. By reducing the physical distance between workers and firms located in the same region, these investments facilitate the exchange of knowledge and ideas. Local governments can maximise the returns from investments into local transport infrastructure by promoting policies that favour the creation and exchange of ideas. A similar mix of policy interventions was adopted in the West Sweden region that gravitates around the main city of Gothenburg. Regional authorities undertook investments into innovation and research infrastructure, through the creation of, for example, science parks and at the same time strengthened the local transport network, which magnified their returns on both investments (OECD, 2018[38]).
Regional differences in highway investment reflect narrowing infrastructure gaps
In Portugal and Spain but also France and the United Kingdom for example, the highway network has grown considerably between 1990 and 2012. In other countries, the growth has been minimal, either because the highway network was already well developed in 1990 (e.g. Germany or Italy) or because these countries experienced very modest changes in the spatial distribution of economic activity during their period considered. For example, in Scandinavian countries, areas that were little populated in 1990 continued to be so also in 2012, so that the need to build new infrastructure in these countries did not materialise. Finally, in East European countries, such as the Czech Republic Hungary or Poland, there has been little improvement in accessibility for people and firms during the past 20 years and considerable additional effort is needed.
The major expansion of the European highway network shifted from the centre and the south to the east between 1990 and 2012. Since 2000, countries like the Czech Republic started to catch up with the growth rate of the length of the highway network in the region of Severovýchod, located in the north-eastern part of the Czech Republic, was 400% compared to national median growth slightly less than 100%. At the same time, the region of Liguria, Italy, experienced zero growth, as opposed to Umbria, which instead saw the length of the transport network grow by nearly 200%. These very high growth rates can be explained in terms of the nearly absent highway network at the beginning of the sample in the corresponding regions.
Countries where predominantly urban regions grew faster than predominantly rural or intermediate regions expanded their highway network less during 2000-12. Academic studies have established a causal and positive relationship between investment into highways departing from the city centre and the share of people in the suburbs of metropolitan areas.16 Across OECD countries with available data, there is some evidence that countries with greater expansion of their highway network experienced a greater shift away from predominantly urban areas to predominantly rural or intermediate regions (Figure 1.10). Ireland and Poland in particular stand out.
Box 1.3. Highway expansion in Europe in urban and rural regions
The type of regions with the largest expansion in the highway network between 2000 and 2012, depends on the country (Figure 1.9). In respectively 6 out of 17 countries, the network expanded most in predominantly urban or predominantly rural regions, with the highway network expanding strongest in intermediate regions in the remaining 5 countries. While in East European countries highway investment during this period focused on priority projects, such as endowing the main cities with a highway network, in many Western European countries, such as France or Spain, highway investment aimed at expanding highway penetration in areas where it had been scarce up to that moment, i.e. predominantly rural areas. It follows that in Western European countries, the length of the highway network grew by 52% (41%) in predominantly rural (urban) regions, while the pattern is reversed in East European regions, where highway length grew by 70% (160%) in predominantly rural (urban) regions.
European regions are not the only ones that witnessed a steep increase in investment in roads. Korea devoted a total budget of KRW 15.8 trillion (USD 14.4 billion) or 1.1% of 2013 GDP road and rail infrastructure investment. The expansion in the road network has been in particular significant, with paved roads in 1951 having a total length of 580 km compared to over 87 000 in 2013, including more than 4 100 km of HS expressways (OECD, 2016[3]).
Maintenance cost account for an increasing share of infrastructure spending
As the expansion of the highway network slows the cost of maintaining existing infrastructure rises. In the United States, for instance, the share of maintenance spending exceeds the one of investment. Real – that is, adjusted for the cost of raw materials – spending on operation and maintenance jumped from USD 243.3 billion to USD 266.5 billion; meanwhile, real spending on capital projects plummeted 16%, from USD 207.1 billion to USD 174.0 billion between 2007 and 2017 (Brookings Institution, 2019[39]). As the length of the road network expands and as the network matures, the relative importance of maintenance costs relative to new investment will increase. Existing studies document that not only required maintenance costs are large but also that large shortfalls materialise with respect to this expenditure category (McKinsey & Company, 2013[40]), with negative consequences for traffic and road safety.
The highway networks in most developed countries require substantial maintenance spending. Most western countries, built them in the first decades after World War II. As budgets tightened, the state of disrepair of these highway networks has become evident after the Great Recession. Many European countries have been forced to put in place more stringent budget limitations that have contributed to the slack in the recovery of public investments, especially in the European periphery (OECD, 2016[41]).
Maximising the lifespan of roads needs therefore to be a guiding principle of new investments into transport infrastructure (ITF, 2018[42]). From a public finance point of view, it is important to minimise the maintenance cost of each new kilometre added to the existing transport network in order for total maintenance costs not to scale up as rapidly as the rate at which the transport network expands. Big data can be used to forecast growth in transport demand with better accuracy and plan transport infrastructure updates more effectively. Traffic management systems can also benefit from more in-depth modelling made possible by the use of big data. For instance, local policymakers could use these big data to devise a system that makes it possible for deliveries to take place in off-peak hours, thus reducing the traffic strain of roads (ITF, 2018[43]). The establishment of a good co‑operation level with the transport and logistics sector can further help local policymakers improve the distribution of storage facilities and collection points on the territory, thus re-equilibrating the traffic strain of certain roads.
The circulation of automated vehicles will likely enhance the efficient use of roads. The major advantages of having automated vehicles circulating on highways will be in terms of safety and drivers’ well-being. The slow diffusion that these technologies have had so far is mostly imputable to normative issues and the lag with which regulatory bodies are catching up, from setting the standards for competition in this sector to the rules for data security (ITF, 2018[44]).
Benefits from transport infrastructure investment
Transport infrastructure creates economic benefits through different channels. Importantly, these channels also differ between infrastructures connecting regions and those connecting people within a city. Interregional transport infrastructure facilitates mostly the movement of goods, while within-city transport infrastructure the movement of people. Given that the investments take place at different scales, the competent authorities as well as governance arrangements are different. A companion report on Improving Transport Planning for Accessible Cities (OECD, 2020[45]) provides an in-depth discussion of planning and governance arrangements for better outcomes from infrastructure investment. The review of the benefits of transport infrastructure investment provided in the upcoming section deals with each type of infrastructure investment separately.
Evaluating the contribution of different channels is complicated by challenges in measuring the economic benefits of transport infrastructure investment (Box 1.3). The location choice of infrastructure is not random but follows a specific rational (e.g. economic, social, political, etc.). If transport infrastructure is constructed to connect economic hubs with high growth potential in a country (or across countries), any estimate of the economic benefits of the investment risks confounding the contribution of the new infrastructure with the contribution of other factors that led to higher growth potential. Economists have developed a range of strategies to address this challenge that help identify the true “causal” effect of investment.17 The strategies rely on historic transport networks, geographical characteristics of the area or policy choices that are unrelated to economic considerations, for example, to try to estimate what the economic development in a place would have been in absence of the transport investment. Such a counterfactual approach does come at a price as the question that can be answered “causally” is often not the question that is the most pressing. For example, a road that connects two major metropolitan areas also creates access for less densely populated areas along the way. Finding a causal estimate for the impact of investment on the two major metropolitan areas is very difficult but many studies have developed strategies to identify the impact on the less densely populated areas along its way.
Box 1.4. Challenges in the measurement of infrastructure investment’s economic impact
The location of transport infrastructure investment is not random but carefully chosen to optimise economic and social objectives. A comparison of the economic development following the realisation of the new infrastructure between places where investment occurred and others will therefore result in misleading estimates of the realised value for connected places. If investment is, for example, targeted towards areas expected to grow, it is difficult to disentangle how much of the resulting growth can be attributed to the investment itself and how much is due to other factors that promoted growth even before investment took place. What is missing is data on a “counterfactual” state of the world, i.e. what would have been the growth in a region or city where investment took place if that investment had not taken place.
Source: Lee, N. and A. Lembcke (2020[46]), “The economic benefits of accessibility: A survey”, Unpublished Manuscript.
Benefits from interregional road and rail infrastructure
While interregional transport infrastructure builds on several transport modes, this section focuses on the benefits associated with road (and in particular highways) and rail infrastructure investment. These 2 modes together accounted in 2015 for 93% of worldwide surface freight transport (ITF, 2019[27]) and 73% of worldwide non-urban passenger transport (ITF, 2017[47]).
Highways and railways have a long history in most OECD countries. Railways construction started during the second half of the 19th century in many countries (e.g. Europe, Australia, Canada, Japan, United States, etc.). For example, in 1840, total length in kilometres of European rail was close to zero. In 1910, it was greater than 200 000 km, more than what it was in 2010 (Martí-Henneberg, 2013[48]). The first modern highways were built from the 1940s onwards driven by the advent of mass production of automobiles and the highway network continues to expand today. The upcoming section draws lessons from a recent set of studies assessing the economic benefits associated with transport infrastructure investments in OECD member countries.18 The results highlight that interregional transport infrastructure investment is likely to benefit some regions but can create adverse economic effects in regions with worse access to infrastructures or, in some cases, in regions where access improves. Careful evaluation is required to ensure that infrastructures provide a positive aggregate net impact.
Interregional transport infrastructure allows for deeper trade integration
Transport infrastructure reduces the cost of shipping goods from one region to another. The costs associated with the shipment of goods typically fall under the umbrella of trade costs, which includes two broad types of costs. The first type of costs is those that arise due to the physical distance between trading partners. These generalised transport costs include both non-pecuniary costs, such as the opportunity cost of travel time required to ship goods, and pecuniary costs, such as fuel costs, taxes and fees, and other operating costs (Combes and Lafourcade, 2005[49]). The second type is costs stemming from the countries’ trading partners reciprocal trade policies (e.g. tariffs). As infrastructure investment tends to be an important tool for regional development, it is necessary to consider the wider range of policies and regulatory barriers that facilitate or hinder trade, in particular for major infrastructure that connects across country borders.19
Better transport infrastructure reduces the cost of shipping goods and magnifies the impact of other cost-reducing measures. For example, in France, the development of the highway network between 1978 and 1998 led to a 3.2% reduction in transport costs. Compared against a generalised reduction of 38% percentage points, transport infrastructure accounts, therefore, for one-tenth of the overall cost reduction. Other measures had a larger impact, the decline in fuel price and fuel consumption accounted for a 6% reduction in transport costs and lower labour costs (partly due to higher productivity and partly due to lower per capita wages) for a further 5.5%. While improvements in the transport network did not contribute as much in absolute terms to the overall decline, they shaped the geography of the gains resulting from other cost-reducing measures. Areas with better transport infrastructure had stronger economic gains than those with worse infrastructure (Combes and Lafourcade, 2005[49]).20
Lower transport costs reduce prices for intermediates and consumer goods produced by firms located farther away. Consumers gain unequivocally through lower average prices and access to a wider variety of goods that they can purchase. For the United States, for example, an estimate suggests that the value of the expanded variety of imported goods between 1972 and 2001 amounted to 2.6% of GDP for US consumers (Broda and Weinstein, 2006[50]). Firms also gain better access to intermediate inputs produced by firms located farther away and sold at a lower price. Some firms, however, do not benefit from better infrastructure connection. Firms that produce with less cost efficiency are at a higher risk of being outcompeted by more distant and competitive firms, in particular when they produce highly standardised products. Conversely, firms that produce more complex or tailored inputs, as well as those that produce more cost efficiently have the opportunity to seize a larger market share. An example is a change in domestic supply links in Japan after the Chinese accession to the World Trade Organization (WTO). Estimates based on 4.5 million buyer-seller links show that local production networks were strengthened (in particular for non‑standardised inputs) and that offshoring and sourcing from outside the country reduced the attractiveness of domestic suppliers that produce standardised inputs and that are located further away from the sourcing firm (Furusawa et al., 2017[51]).
Benefits of new transport infrastructure depend on the accessibility it provides
The gains for a given region depend on how many new customers can be reached via the improved transport network. Simply building a highway is not sufficient for economic growth to materialise in a given region. The crucial question is what improvements in accessibility the new infrastructure provides. The average income level – proxying for expenditure capacity – of the new markets accessible by the region is one way to capture accessibility (see Box 1.4). Accessibility and overall gains also depend on the wider transport network. If the additional infrastructure connects to an existing network with good accessibility or improves access to existing logistic centres, ports or airports, it is more likely to support significant benefits in terms of freight traffic volumes. The Megaregion of Western Scandinavia – covering the regions along the western coast from Oslo via Gothenburg to Malmö and Øresund – is a major freight hub for Norway and Sweden. It hosts Norway’s largest airport, Sweden’s two largest ports and connecting road infrastructure that has been improved in recent years. In contrast, there is a lack of fast rail infrastructure crossing from Norway into Sweden and between the second- and third-largest cities in Sweden. Trains have gradually lost competitiveness to roads, particularly on the Oslo-Gothenburg route (OECD, 2018[38]).
Additional highways tend to create better market access but gains vary across regions. Roads are essentially a means to fulfil a certain need, such as the exchange of goods between two places but the construction of a road in itself does not necessarily generate such need, therefore potentially failing to create a market where it does not exist. In support of this argument, the correlation between the evolution of market access and the evolution of highway length across TL3 regions during 2000-12 is far from perfect (Figure 1.11). In about 40% of the 743 TL3 regions for which data are available, the growth rate in market access between 2000 and 2012 was lower than the one expected given the growth rate in highway length during the same period.21 In Spain, for example, between 1980 and 2000, market access improved in almost all municipalities because of the highway network construction but the improvements were more pronounced in the most peripheral regions (Holl, 2007[52]). Due to the non-automatic relation between road length and market access, the use of changes in road length as a proxy for the opportunities unlocked by transport infrastructure investment can be quite misleading and metrics based on market access should be preferred.
Accessibility increases by building new infrastructure and through growth in already connected regions. An important feature of infrastructure is that once a road is in place, continuous economic improvements in one place along its path can create positive spill-overs to other connected places. Transport connections can therefore create lasting stimulus by benefitting from the success of connected places. In a study for 28 European countries, road infrastructure construction is the main determinant for accessibility improvements when fewer mobile factors are considered. For accessibility to population, 66% of the increase in accessibility between 1990 and 2012 comes from improved road infrastructure. In contrast, for more mobile factors (such as employment and GDP), growth in connected areas becomes more important. For GDP, only 18% of the accessibility improvements can be attributed to new or better roads, the remaining 82% stem from economic growth in already connected places and in case of access to employment the two aspects are roughly equally important (Adler et al., 2020[53]).
Box 1.5. An accessibility-based measure of potential economic gains
The notion of “market access” – how the term “accessibility” is translated by trade economists – is central to the field of New Economic Geography (NEG). At the core of NEG models is a rigorous treatment of the various determinants of the location decision of workers and firms. When there are transport costs, the decision on where to locate depends both on the size of the market reachable given the available transport network and on the unit cost of production.
The potential economic gains from transport infrastructure investment for a region are proportional to changes in market potential, where one way to measure market potential is (Adler et al., 2020[53]):
Since expenditure in a given region m depends on its income, one way to construct a measure of market potential is by taking the sum of all regions in a country’s GDP after having discounted them by the cost of shipping goods from region r to region m, or . The parameter d captures the extent to which goods produced in one region are substitutable with goods produced in another region. The more substitutable – the higher d – the goods, the larger the market potential offered by close-by regions as opposed to regions further away.
Production in one region can increase because, thanks to improved transport infrastructure, local producers can now start selling to richer regions located farther away. However, changes in market potential depend not only on changes in the geography of transport infrastructure but also on changes in the geography of economic activity. Firms located in a small region can in principle benefit from the construction of a new highway connecting to a larger one. However, if firms located in the larger region are more competitive in a larger share of industries, consumers located in the small region will want to purchase their products. A labour supply shortage in the larger region will drive local wages up, and eventually be filled either via commuting or via actual immigration from the nearby small region. In either case, market potential will dry up in the small region, thereby reinforcing the relocation of firms towards the larger one.
The economic development of cities in the People’s Republic of China following the construction of the highway network starting in 1988 exemplifies the potential trade‑offs inherent in infrastructure development. The network was conceived with the idea of promoting economic activity in hinterland prefectures but had the opposite result. “Primate” cities – the largest prefectures located within one day’s drive from hinterland prefectures – specialised in manufacturing for which they enjoyed a comparative advantage. Manufacturing created jobs that produced more value-added and therefore paid higher wages, which attracted workers to primate cities. They grew in employment and population at the expense of nearby hinterland prefectures that shrank in size and specialise in agriculture (Baum-Snow et al., 2016[17]).
Source: Adler, M. et al. (2020[53]), “Roads, market access and regional economic development”, OECD Regional Development Working Papers, OECD Publishing, Paris; Baum-Snow, N. et al. (2016[17]), “Highways, market access, and urban growth in China”.
Market access improvements promote regional economic development. Adler et al. (2020[53]) estimate the impact of accessibility improvements that the expansion of the road network and the growth of Europe created between 1990 and 2012 on GDP, employment and population. A 10% increase in market access (in terms of GDP) in other regions increases the GDP in a region by 2% on average. To put this into perspective, Haute-Garonne, the French département where Toulouse, France’s fifth-largest metropolitan area, is located, benefitted from a significant improvement in the highway network in the southwest of France and in Spain. The result was an increase in accessibility in terms of GDP of nearly 40% between 1990 and 2012 with about 60% of that improvement during the first 10 years of the period. Madrid, Spain’s capital city, became the centre of the country’s rapidly expanding highway network during the period but the more peripheral location within Europe meant that the accessibility gains were smaller than in Toulouse, about 30% in terms of GDP accessibility, with half of the gains already realised between 1990 and 2000. But even this smaller improvement translates to 6% additional GDP over the 22-year period due to improvements in accessibility. For other outcomes, the impact is even stronger than for GDP. Employment in regions, where accessibility to employment outside the region improved by 10%, grew on average by 7%. For the population, the impact is of 6% in a similar order of magnitude. Combined, these results show that better accessibility helps regions grow but also that new opportunities support mainly activities that are in lower value-added and lower productivity sectors, as GDP grows less than employment or population.
The benefits of new infrastructure depend on the existing network with the strongest gains in places closing infrastructure gaps (Adler et al., 2020[53]). Estimates for regions in Eastern Europe for the 1990-2012 period show benefits from market access improvements in terms of GDP and per capita GDP. This is in line with an expansion of economic activity that was stronger for sectors with high value-added, capturing the expansion of manufacturing activities and the integration of regions in new EU member countries in European (or global) value chains. In contrast, market access increased employment in Southern European regions but these did not translate into GDP gains, i.e. employment improvements appear to have mainly fostered growth in low- and medium-income jobs as GDP gains and per capita GDP gains are statistically insignificant. The benefits from improved accessibility arise mainly through the integration with the European-wide road network. When estimates are limited to access improvements within three hours of driving, roughly equivalent to the distance of a business daytrip, the results remain positive but are reduced in magnitude by a factor of four. The market access benefits from European integration and trade are substantially larger than those at the regional level and those for smaller clusters of regions.
Trade and productivity along highways
Improvements in road accessibility create trade-driven growth opportunities for both existing firms and new entrepreneurs. These opportunities materialise either via direct links with the new markets or via indirect links operating through supply chains. Most studies analysing the economic impact of highway construction find evidence of a rise in the total number of active firms in the regions benefitting from highway investment. For example, following the dramatic improvements in the British road network between 1960 and 1990, a 10% increase in accessibility in a neighbourhood (electoral ward with an average population of 6 000 inhabitants) led to a 3%-4% increase in the number of firms (Gibbons et al., 2019[54]). A few studies are also able to differentiate whether the rise in the number of firms is driven by fewer firm deaths or more firm births, finding evidence of an increase in the birth rate of firms (Holl, 2004[55]; 2004[56]). These studies recognise that new firms can be displaced from other regions. In Spain, following the construction of the motorway network, the birth rate of firms in municipalities between 10 km and 20 km away from the new motorways dropped by 13.5 percentage points between 1980 and 1994 relative to the birth rate in municipalities within 10 km from the new motorways (Holl, 2004[55]).
Opportunities for trade increase labour demand. Employment increases both at the aggregate and the plant levels in the regions receiving a highway connection. For example, following the dramatic improvements in the British road network between 1960 and 1990, a 10% increase in accessibility led to a 3%-4% increase in local employment (Gibbons et al., 2019[54]).22 Moreover, the effects are larger for large and manufacturing-oriented firms, i.e. those that are more likely to trade over longer distances within the country or export their goods (Audretsch, Dohse and Pereira dos Santos, 2017[57]).
Wages can go up in regions receiving a highway connection (Sanchis-Guarner, 2012[58]; Gibbons et al., 2019[54]; Chandra and Thompson, 2000[59]). For example, following the expansion of the British road network between 1960 and 1990, a 10% increase in accessibility results in an increase in average wages of around 2.5%-3% (Gibbons et al., 2019[54]). The increase in wages can be a consequence of rising labour shortages, i.e. the increase in labour demand outpaces the supply of labour. However, continued increases in wages are unlikely to be driven by labour shortages. In the case of longer term increases in wages the likely channel is an increase in productivity in local firms. For example, Gibbons et al. (2019[54]) find the firm-level productivity increase following the improvement in accessibility in the UK to be of the same order of magnitude as the increase in average wages. In India, following the construction of the Golden Quadrilateral, manufacturing productivity growth during 2000-09 in districts located less than 10 km away from the new highways was 2 percentage points higher than in districts located more than 50 km away from the new highways (Ghani, Goswami and Kerr, 2016[60]). Moreover, aggregate productivity can also increase thanks to a more efficient allocation of resources, i.e. workers taking jobs in firms that are more productive and investment flowing into those firms. This is what happened for example in India following the construction of the Golden Quadrilateral: industries with a larger share of employment located close to the new highways gained more in terms of allocative efficiency, i.e. in terms of the likelihood of more productive plants being also of larger size (Ghani, Goswami and Kerr, 2016[60]).
More efficient transport infrastructure can facilitate the spread of ideas and boost local innovation. Agrawal, Galasso and Oettl (Agrawal, Galasso and Oettl, 2017[61]) find that in the United States a 10% increase in the stock of highways in 1983 at the metropolitan level translated into 1.7% more regional patenting during 1983-88. While patents are only an imperfect measure of innovation, it is reasonable to believe that competition among firms increases after an increase in connectedness. Hence, to the extent that product market competition is a solid driver of innovative activity (Aghion and Akcigit, 2015[62]), a reduction in transport costs should also be expected to lead to an increase in innovative activity.
Trade and productivity along railways
Historically, access to the railway network is an important driver of long-term growth. Studies on the economic impact of railway expansion tend to consider population growth as their outcome of interest.23 The unequivocal finding from studies focusing on the early years of the railway expansion is that access puts cities on a path of steadily higher population growth compared to other cities (Hornung, 2015[63]; Atack et al., 2010[64]; Berger and Enflo, 2015[65]; Büchel and Kyburz, 2016[66]; Donaldson and Hornbeck, 2015[67]). For example, Hornung (2015[63]) finds that the population grew in cities that benefitted from railway access 2.1 percentage points more compared to cities that did not during 1849–71. As these studies focus on cities in the late 19th century, they show the importance of initial connections while the transport network expanded. The effect comes from the reduction in transport cost that raises the value of production and land (in these historic studies primarily the value of agricultural land). The strength of effects today is likely weaker as highways, trucks and cars provide suitable substitutes for many trips.
High-speed (HS) rail investment provides economic benefits if major economic hubs anchor the routes. Ahlfeldt and Feddersen (2018[68]) find that 10 percentage point faster growth in market access following the construction of a HS railway between Cologne and Frankfurt leads to 2-3 percentage point higher GDP per growth during 1992-2006. Similar results are found by Carbo et al. (2019[69]) for the HS rail corridor connecting Madrid to Barcelona. The main difference between traditional and HS railway investment is that while traditional railway also serves the purpose of shipping goods, HS railway eases passenger traffic and business trips. Therefore, while the first-order impact of traditional railway investment is likely to be on the manufacturing sector, the most likely consequences of HS railway investment are trade in services and a reorganisation of production in multi-establishment firms. A cautionary note is that the cost of HS railway investment is high, both for the initial development and for the subsequent operation of the lines. Most currently existing lines connect major cities with substantial passenger flows between them. The costs and required subsidies vary substantially even between these major lines (Albalate and Bel, 2012[70]), which means that any development requires careful evaluation of the potential demand.
The effect of shutting down a railway is negative. Most academic studies focus on the development of new infrastructure, few studies analyse the impact of disinvestments, partly due to the rarity in the occurrence of disinvestment projects. The economic interest for this type of projects should be high given that current infrastructure faces the risk of becoming obsolete in face of newer technologies. Gibbons, Heblich and Pinchbeck (2019[71]) find that the national railway disinvestment of unprofitable lines that took place in the United Kingdom during the 1950s through the 1970s contributed to the decline in rural population. For local communities, a 10% decrease in accessibility due to the dismantlement of the railway network between 1951 and 1981 is associated with a population 3 percentage points lower in 1981 than in a district where market access did not change. The population shifted towards cities and rural areas in the North West and Central England because of the rail cuts in areas of Scotland, South West England and Wales. The major benefactors were the largest UK cities. Without the disinvestment, London and its commuting area would have had 9% less population in 2001. Other major cities such as Birmingham, Glasgow and Manchester would also have had a lower population.
The limitations of transport infrastructure benefits
The potential gains from infrastructure improvements are finite. A counterfactual exercise for EU countries considers providing “perfect” accessibility in all regions by upgrading the existing secondary road network to highways (the roads with the fastest speeds). The results of Adler et al. (2020[53]) suggest that predicted gains in terms of GDP are, on average, between 17% and 26%, with highest remaining gains for regions in Eastern Europe. Predictions for employment and population gains are slightly larger. Overall, the already well‑develop highway network provides a high degree of accessibility in many parts of continental Europe but there are still places remaining where benefits can be substantial, especially in Eastern Europe. Additional interventions should, of course, follow best practices and be subject to careful CBA in line with best practices in OECD member countries.
The gains in one region can come at the expense of another region. An example is the response by firms to the construction of the Portuguese motorway system between 1986 and 1997. New firms chose to locate in municipalities located in a 10-kilometre corridor from new motorways, at the expense of the municipalities located outside the corridors (Holl, 2004[55]). In the United States, earnings in counties that became connected to the Interstate Highway System between 1969 and 1993 rose whereas earnings declined in those without access (Chandra and Thompson, 2000[59]) relative to the evolution of the average wage in each state and year. The concentration of economic activity comes also at a cost as agglomeration benefits are balanced by agglomeration costs (Box 1.5).
Aggregate gains depend on whether a new economic activity is created and on the extent to which workers and firms relocate to take advantage of new opportunities. Aggregate gains are maximised when new infrastructure gives rise to new economic activity (gains from additionality) possibly in a country’s more productive regions (gains from improved allocative efficiency) without subtracting resources from the other regions (losses from displacement). The demand for new hires and investments generated in a region by new infrastructure is met by tapping into both local resources and those initially located in other regions. Since there are no restrictions to the mobility of workers, local job seekers as well as job seekers living in other regions and willing to relocate may try to take advantage of the new employment opportunities associated with the demand for new economic activity. Combes, Gobillon and Lafourcade (2019[72]) show that the aggregate gains for France from the Grand Paris Express project would be maximised if the new jobs created by the project were filled by individuals that never lived in France, as opposed to individuals moving from other regions of France to Paris.24 Indeed, in the most common case in which the demand for new economic activity in the region receiving the new infrastructure is also met by tapping into resources initially located in other regions, the aggregate gains will be given by the gains for this region minus the losses in terms of economic activity experienced by other regions.
Benefits from infrastructure investment are clearly outweighed by indirect costs in two extreme cases. The first is when the new infrastructure merely triggers a redistribution of economic activity across initially similar regions. In this case, one region merely gains resources (e.g. workers) at the expenses of another region with similar characteristics. The second is when capital investment that follows infrastructure investment would have taken place in any case absent of infrastructure. In both cases, gains from additionality and gains from improved allocative efficiency fail to materialise.
Aside from these two extreme cases, the aggregate effect tends to be positive. Whether benefits outweigh adverse effects depends on the balance between the gains from additionality and the gains (losses) from improved (worsened) allocative efficiency. For instance, the new infrastructure can trigger in the recipient region other forms of capital investment that would have taken place however in other parts of the country absent of the infrastructure. Gains from additionality fail to materialise in this case as well. However, the project can still have an aggregate net positive value if it redistributes resources towards more productive regions. New infrastructure can trigger in the recipient region other forms of capital investment that would not have taken place in any other parts of the country absent of the infrastructure. In this case, the project presents gains from additionality. Nevertheless, the aggregate net positive value is reduced if resources are pulled away from more productive regions in order to satisfy the new demand for labour created by the infrastructure. These potential losses from worsened allocative efficiency can be averted if the project succeeds in attracting unutilised local resources into production, for instance by tapping into a situation of high unemployment or low labour force participation (Venables, Laird and Overman, 2014[73]).
Agglomeration shadows can outweigh the benefits of better access. In a seminal study, Vickerman, Spiekermann and Wegener (1999[74]) argue that the promotion of HS transport networks across Europe would benefit already well-connected core areas over peripheral areas that the investment in transport connection aims to support. Better connections may help smaller cities benefit from agglomeration economies in large cities but may also lead to agglomeration shadows where firms in the core start serving the market in the surrounding area at the expense of local businesses. Transport infrastructure investment may therefore amplify differences among ex ante asymmetric regions.
One of the reasons for which pre-existing gaps may be amplified is the existence of agglomeration economies, i.e. productivity advantages from co-location (Box 1.5). As accessibility in a region improves, firms can serve a wider market. The larger market potential attracts new firms thanks to economies of scale, i.e. the fact that the firm’s average cost declines as output increases. In choosing where to locate, firms can also reap the benefits from locating in already dense areas, next to already existing firms and close to a larger pool of workers, i.e. they can also reap “agglomeration benefits”. The productivity gains from co-location mean that firms are more productive, which allows them to pay better wages and thereby attract more workers, and in turn attract more firms. Estimates for Chinese rural prefectures following the construction of the highway network started in 1988 show that this is the case as “primate cities” connected by new highways drew in workers from nearby rural prefectures. The rural prefectures lost population and manufacturing employment following the construction of the highway network (Baum-Snow et al., 2016[17]).
Inequality within regions can increase in response to infrastructure investment. Efforts to improve accessibility may also differentially affect different groups, with skilled workers often benefitting at the expense of the less well qualified (Fretz, Parchet and Robert-Nicoud, 2017[75]). Michaels (2008[76]) finds that labour demand for skilled workers increased during 1959-75 more in counties traversed by the US interstate highway system compared to the rest. Fretz et al. (2017[75]) find that the share of top taxpayers increased by 24% between 1947 and 2010 in areas located within 10 km of access to the new highway network in Switzerland. They argue that this is because high-income workers can benefit more than proportionately from the opportunities offered by tighter trade integration.
Box 1.6. Agglomeration economies: Benefits and costs
Metropolitan areas and dynamic medium-sized cities have enormous potential for job creation and innovation as they are hubs and gateways for global networks such as trade or transport. In many OECD member countries, labour productivity (measured in terms of GDP per worker) and wages can be observed to increase with city size.
Stronger productivity levels are a reflection of a bonus intrinsic to being in a city, known as the agglomeration benefit. On average, a worker’s wage increases with the size of the city where he/she works, even after controlling for worker attributes such as education level. OECD estimates suggest that agglomerations benefit in the form of a wage premium rises by 2%-5% for a doubling of population size (Ahrend et al., 2014[77]), which is in line with comparable studies for individual countries (Combes et al., 2012[78]). However, agglomeration benefits do not accrue homogeneously across cities and they show sizeable variations within countries.
Higher productivity is due in part to the quality of the workforce and the industrial mix. Larger cities on average have a more educated population, with the shares of both very high-skilled and low-skilled workers increasing with city size. A 10 percentage point increase in the share of university-educated workers in a city raises the productivity of other workers in that city by 3%-4% (Ahrend et al., 2014[77]). Larger cities typically have a higher proportion of sectors with higher productivity, such as consulting, legal or financial services, etc. They are also more likely to be hubs or service centres through which trade flows and financial and other flows are channelled. These flows typically require the provision of high value-added services.
Living in large cities does provide benefits but it also has disadvantages. While productivity, wages and the availability of many amenities generally increase with city size, so do what is generally referred to as agglomeration costs. Some agglomeration costs are financial: for example, housing prices/rents and, more generally, price levels are typically higher in larger cities. In addition, a number of non-pecuniary costs, such as pollution, congestion, inequality and crime, typically also increase with city size, while trust and similar measures of social capital often decline. Survey data from European cities confirm that citizens in larger cities – despite valuing the increased amenities – are generally less satisfied with the other aspects mentioned, notably air pollution.
To some extent, city size is the outcome of a trade-off between these benefits and costs. Mobility across and within cities implies that – at least in the medium to long term – wage levels, commuting costs and other urban (dis)amenities are reflected in land prices, and more generally in a city’s cost of living. This is supported by findings suggesting that for increasing population size, these agglomeration benefits and costs go up at a broadly comparable pace (see Combes et al. (2012[78]) for evidence on France and Gibbons et al. (2011[79]) for the United Kingdom). A similar picture emerges when looking directly at cities’ productivity and price levels. Evidence from Germany shows that, on average, increases in a city’s productivity, and hence wages, are matched by similar increases in local price levels.
Source: OECD (2015[80]), OECD Urban Policy Reviews: Mexico 2015: Transforming Urban Policy and Housing Finance, https://doi.org/10.1787/9789264227293-en.
Benefits from urban transport infrastructure
Economic benefits
Improvements of the public transport network in cities raise firm productivity and the value of land and housing around the stops. Firm productivity responds to the densification around stops and the associated economies of agglomeration (Box 1.6). House prices in the areas affected by the construction or refurbishment of the public transport network tend to rise in response. As areas become more accessible, residents can gain better access to opportunities, whether services or jobs, and therefore housing demand in these areas increases.
An increase in house prices signals that a given transport infrastructure investment succeeded in the goal of improving accessibility and welfare. Gibbons and Machin (2004[81]) evaluate the impact on house prices of the construction of new stations under improvements made to the London Underground and Docklands Light Railway in South East London in the late 1990s. They find that house prices in the areas affected by the transport investment increased by 9.3 percentage points more than the other areas during 1997-2001. Based on the same transport innovation, Pogonyi et al. (2019[82]) find that areas located in the proximity of the new metro line saw their productivity and the total number of plants increase, while areas located farther away saw the total number of plants decrease but no change in productivity. Billings (2011[83]) estimates that light rail transit in Charlotte, North Carolina, led to an increase in the price of single-family properties of 4% and 11.3% for condominiums sold within 1 mile of light rail transit stations. Ahlfeldt et al. (2015[84]) find that city blocks located in the proximity of the Berlin wall and farther away from 1936 subway stations experienced a smaller loss in the price of floor space as a consequence of the city division. Finally, Mayer and Trevien (2017[85]) find that places that received an RER connection in Paris witnessed higher employment growth between 1970 and 1990 but also experienced higher rates of gentrification. Evidence that considers the benefits of Seoul’s bus rapid transit system finally also finds strong positive effects on densification and land prices around the system’s stops (Cervero and Kang, 2011[86]), with land price premiums of up to 10% estimated for residences within 300 m of bus rapid transit stops.
Broader socio-economic benefits
Transport infrastructure investment matters also for a broader range of socio-economic outcomes at the neighbourhood level. For instance, the expansion of light rail in 16 major US cities during 1970-2000 had very heterogeneous impacts on segregation patterns. “Walk and Ride” stations attracted skilled residents, while “Park and Ride” suburban stations often experienced an increase in poverty (Kahn, 2007[87]). Moreover, the improvement of public transport, especially in city centres, coincided with a trend of increasing concentration of poor households in central cities, while skilled residents who are less public transport-dependent could migrate to suburban areas (Glaeser, Kahn and Rappaport, 2008[88]). Based on data on the full Facebook social network in the United States, Bailey et al. (2019[89]) find that the shape of the transport network is a more effective predictor of social connectedness between any two places than the mere physical distance. These findings stress the importance of the transport policy when it comes to achieving certain policy objectives such as social inclusiveness.
Studies analysing the impact of transport infrastructure investment in cities would benefit from the adoption of an accessibility-based measure. Currently, most studies use the density of the transport network as a proxy for the exposure to potential gains from transport infrastructure investment. This choice is motivated partly by the lack of detailed data on the distribution of economic activity within cities. However, this is also a potentially important oversight since, as already discussed, transport infrastructure investment is not important per se but rather because of its role at connecting people to jobs, producers to consumers, etc. Accessibility measures help capture this aspect (see Chapter 2 for more on accessibility measures).
Effects of transport infrastructure on the reorganisation of economic activity
Urban highways lead to suburbanisation in metropolitan areas and job growth. Radial segments of the highway network departing from the city centre cause a move of people into the suburbs.25 Looking at different types of transport infrastructure, Baum-Snow et al. (2017[90]) find that radial highways displaced 4% of the central city population in Chinese cities to surrounding regions and ring roads displaced about an additional 20%. Duranton and Turner (2012[91]) assess the overall impact of broadly defined road density on city growth between 1983 and 2003 and find that a 10% increase in the stock of highway causes a 1.5% rise in local employment in US cities. This finding is confirmed in some OECD member countries. Brandily and Rauch (2018[92]) show that Sub-Saharan cities that experienced a weaker increase in road density in the city centre also saw a population growing at a lower rate.
Transport infrastructure investment not only causes suburbanisation in cities but also triggers a reorganisation of economic activity in metropolitan areas. For example, transport infrastructure investment in China displaced economic activity away from central cities towards the suburbs, not just population, and that more specifically, each radial railroad reduced central city industrial GDP by about 20%, with ring roads displacing an additional 50% (Baum-Snow et al., 2017[90]). Garcia-López, Hémet and Viladecans-Marsal (Garcia-López, Hémet and Viladecans-Marsal, 2017[93]) show that the construction of rail station in Paris between 1968 and 2010 reinforced the polycentric nature of the city, leading to the emergence of several sub-centres in the proximity of the new railway stations. These effects are however not universal.
The reorganisation of economic activity depends on the sectors. Baum-Snow et al. (2017[90]) provide evidence that radial highways decentralise service sector activity in Chinese cities, radial railroads decentralise industrial activity and ring roads decentralise both. This decentralisation pattern is similar to the one observed for European and US cities from the 1940s to the 1960s, during which time manufacturing plants originally located in central cities benefitted from the development of transport network and moved to the urban periphery to find lower labour and land costs (Mayer and Trevien, 2017[85]).
An important misconception is that transport infrastructure investment relieves congestion. The amount of vehicle-travelled kilometres in US cities increases proportionally to roadway lane kilometres for interstate highways (Duranton and Turner, 2011[94]). The results point to a “fundamental law of road congestion”, i.e. more supply of roads leads to more people driving and the same level of congestion as before. Policymakers that aim at expanding the transport infrastructure network in order to reduce road congestion should keep in mind that as the network expands, more workers and firms will take advantage of it for their purposes, whether it is to commute to work or to ship goods, therefore leaving road congestion overall unchanged.
References
[53] Adler, M. et al. (2020), “Roads, market access and regional economic development”, OECD Regional Development Working Papers, OECD Publishing, Paris.
[62] Aghion, P. and U. Akcigit (2015), Innovation and Growth: The Schumpeterian Perspective.
[61] Agrawal, A., A. Galasso and A. Oettl (2017), “Roads and innovation”, The Review of Economics and Statistics, Vol. 99/3, pp. 417-434.
[68] Ahlfeldt, G. and A. Feddersen (2018), “From periphery to core: Measuring agglomeration effects using high-speed rail”, Journal of Economic Geography, Vol. 18/2, pp. 355-390.
[84] Ahlfeldt, G. et al. (2015), “The economics of density: Evidence from the Berlin Wall”, Econometrica, Vol. 83/6, pp. 2127-2189.
[77] Ahrend, R. et al. (2014), “What Makes Cities More Productive? Evidence on the Role of Urban Governance from Five OECD Countries”, OECD Productivity Working Papers, No. 6, OECD Publishing, Paris, https://doi.org/10.1787/2ce4b893-en.
[70] Albalate, D. and G. Bel (2012), “High-speed rail: Lessons for policy makers from experiences abroad”, Public Administration Review, Vol. 72/3, pp. 336-349.
[12] Allain-Dupré, D., C. Hulbert and M. Vincent (2017), “Subnational Infrastructure Investment in OECD Countries: Trends and Key Governance Levers”, OECD Regional Development Working Papers, No. 2017/05, OECD Publishing, Paris, https://dx.doi.org/10.1787/e9077df7-en.
[64] Atack, J. et al. (2010), “Did railroads induce or follow economic growth?: Urbanization and population growth in the American Midwest 1850-1860”, Social Science History, Vol. 34/2, pp. 171–197.
[57] Audretsch, D., D. Dohse and J. Pereira dos Santos (2017), Do Toll-free Highways Foster Firm Formation and Employment Growth? Results from a Quasi-natural Experiment.
[89] Bailey, M. et al. (2019), Social Connectedness in Urban Areas.
[102] Baum-Snow, N. (2007), “Did highways cause suburbanization?”, The Quarterly Journal of Economics, Vol. 122/2, pp. 775-805.
[90] Baum-Snow, N. et al. (2017), “Roads, railroads, and decentralization of Chinese cities”, Review of Economics and Statistics, Vol. 99/3, pp. 435-448.
[17] Baum-Snow, N. et al. (2016), “Highways, market access, and urban growth in China”.
[65] Berger, T. and K. Enflo (2015), “Locomotives of local growth: The short-and long-term impact of railroads in Sweden”, Journal of Urban Economics, Vol. 98, pp. 124-138.
[35] Bernard, A., A. Moxnes and Y. Saito (2019), Production Networks, Geography and Firm Performance.
[83] Billings, S. (2011), “Estimating the value of a new transit option”, Regional Science and Urban Economics, Vol. 41/6, pp. 525-536.
[92] Brandily, P. and F. Rauch (2018), Roads and Urban Growth.
[50] Broda, C. and D. Weinstein (2006), “Globalization and the gains from variety”, Quarterly Journal of Economics, Vol. 121/2, pp. 541–585.
[39] Brookings Institution (2019), “Shifting into an era of repair: US infrastructure spending trends”, https://www.brookings.edu/research/shifting-into-an-era-of-repair-us-infrastructure-spending-trends/.
[7] Brooks, L. and Z. Liscow (2019), “Infrastructure costs”, Hutchins Center Working Paper, No. 54, Brookings.
[66] Büchel, K. and S. Kyburz (2016), Fast Track to Growth? The Impact of Railway Access on Regional Economic Development in 19th Century Switzerland.
[69] Carbo, J. et al. (2019), “Evaluating the causal economic impacts of transport investments: Evidence from the Madrid-Barcelona high speed rail corridor”, Journal of Applied Statistics, Vol. 46/9, pp. 1714-1723.
[6] CEB (2017), Investing in Public Infrastructure in Europe - A Local Economy Perspective, Council of Europe Development Bank, https://coebank.org/media/documents/Investing_in_Public_Infrastructure_in_Europe_27dc1Pg.pdf.
[86] Cervero, R. and C. Kang (2011), “Bus rapid transit impacts on land uses and land values in Seoul, Korea”, Transport Policy, Vol. 18/1, pp. 102-116.
[59] Chandra, A. and E. Thompson (2000), “Does public infrastructure affect economic activity? Evidence from the rural interstate highway system”.
[78] Combes, P. et al. (2012), “The productivity advantages of large cities: Distinguishing agglomeration from firm selection”, Econometrica, Vol. 80/6, pp. 2543-2594.
[72] Combes, P., L. Gobillon and M. Lafourcade (2019), Labour Productivity Gains from the “Greater Paris Express” Rail Project.
[49] Combes, P. and M. Lafourcade (2005), “Transport costs: Measures, determinants, and regional policy implications for France”, Journal of Economic Geography, Vol. 5, pp. 319-349.
[103] Crescenzi, R. and A. Rodríguez-Pose (2008), “Infrastructure endowment and investment as determinants of regional growth in the European Union: The economics of regional transport investment”, European Investment Bank Papers, Vol. 13/2, pp. 62-101.
[67] Donaldson, D. and R. Hornbeck (2015), “Railroads and American economic growth: A market access approach”.
[22] Duranton, G. and E. Guerra (2016), “Urban accessibility: Balancing land use and transportation”, University of Pennsylvania, https://faculty.wharton.upenn.edu/wp-content/uploads/2017/05/Urban-accessibility-Balancing-land-use-and-transportation-Gilles.pdf (accessed on 3 April 2018).
[97] Duranton, G. et al. (2015), Handbook of Regional and Urban Economics, Elsevier.
[96] Duranton, G., P. Morrow and M. Turner (2014), “Roads and trade: Evidence from the US”, The Review of Economic Studies, Vol. 81/2, pp. 681-724.
[91] Duranton, G. and M. Turner (2012), “Urban growth and transportation”.
[94] Duranton, G. and M. Turner (2011), “The fundamental law of road congestion: Evidence from US cities”, American Economic Review, Vol. 101/6, pp. 2616-2652.
[30] EC (2017), Delivering TEN-T Facts and Figures, European Commission.
[31] EC (2017), Progress Report on the Implementation of the TEN-T Network in 2014 and 2015, European Commission.
[24] EC (2014), Guide to Cost-Benefit Analysis of Investment Projects, European Commission.
[33] EC (n.d.), Infrastructure and Investment, European Commission, https://ec.europa.eu/transport/themes/infrastructure_en.
[20] European Parliament Research Service (2018), Investment in Infrastructure in the EU: Gaps, Challenges, and Opportunities.
[34] Eurostat (2020), Railway Transport – Length of Lines, by Maximum Speed (rail_if_line_sp) (database).
[95] Fernald, J. (1999), “Roads to prosperity? Assessing the link between public capital and productivity”, The American Economic Review, Vol. 89/3, pp. 619–638.
[75] Fretz, S., R. Parchet and F. Robert-Nicoud (2017), Highways, Market Access and Spatial Sorting.
[51] Furusawa, T. et al. (2017), “Global sourcing and domestic production networks”, CESifo Working Papers, No. 6658, CESifo.
[93] Garcia-López, M., C. Hémet and E. Viladecans-Marsal (2017), “Next train to the polycentric city: The effect of railroads on subcenter formation”, Regional Science and Urban Economics, Vol. 67, pp. 50-63.
[104] Garcia-López, M., A. Holl and E. Viladecans-Marsal (2015), “Suburbanization and highways in Spain when the Romans and the Bourbons still shape its cities”, Journal of Urban Economics, Vol. 85, pp. 52-67.
[5] GETLINK (2019), Traffic Figures, https://www.getlinkgroup.com/uk/eurotunnel-group/operations/traffic-figures/.
[60] Ghani, E., A. Goswami and W. Kerr (2016), “Highway to success: The impact of the Golden Quadrilateral Project for the location and performance of Indian manufacturing”, The Economic Journal, Vol. 126/591, pp. 317-357.
[71] Gibbons, S., S. Heblich and E. Pinchbeck (2019), The Spatial Impacts of a Massive Rail Disinvestment Program: The Beeching Axe.
[54] Gibbons, S. et al. (2019), “New road infrastructure: The effects on firms”, Journal of Urban Economics, Vol. 110, pp. 35-50.
[81] Gibbons, S. and S. Machin (2004), Valuing Rail Access Using Transport Innovations.
[79] Gibbons, S., H. Overman and G. Resend (2011), “Real earnings disparities in Britain”.
[88] Glaeser, E., M. Kahn and J. Rappaport (2008), “Why do the poor live in cities? The role of public transportation”, Journal of Urban Economics, Vol. 63/1, pp. 1-24.
[99] Government of Canada (2020), Transportation Activity Indicators, Transport Canada, https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2310026901&cubeTimeFrame.startMonth=01&cubeTimeFrame.startYear=2019&cubeTimeFrame.endMonth=07&cubeTimeFrame.endYear=2020&referencePeriods=20190101%2C20200701 (accessed on 4 November 2020).
[23] Greater London Authority (2018), Mayor’s Transport Strategy, http://www.london.gov.uk (accessed on 15 July 2019).
[52] Holl, A. (2007), “Twenty years of accessibility improvements. The case of the Spanish motorway building programme”, Journal of Transport Geography, Vol. 15/4, pp. 286-297.
[55] Holl, A. (2004), “Manufacturing location and impacts of road transport infrastructure: Empirical evidence from Spain”, Regional Science and Urban Economics, Vol. 34/3, pp. 341-363.
[56] Holl, A. (2004), “Transport infrastructure, agglomeration economies and firm birth: Empirical evidence from Portugal”, Journal of Regional Science, Vol. 44/4, pp. 693-712.
[63] Hornung, E. (2015), “Railroads and growth in Prussia”, Journal of the European Economic Association, Vol. 13/4, pp. 699-736.
[28] IEA (2019), The Future of Rail, International Energy Agency, Paris, https://www.iea.org/reports/the-future-of-rail (accessed on 3 June 2020).
[15] IMF (2014), IMF World Economic Outlook (WEO), Legacies, Clouds, Uncertainties, International Monetary Fund, https://www.imf.org/external/pubs/ft/weo/2014/02/ (accessed on 5 September 2019).
[36] Institut d’Aménagement et d’Urbanisme (2013), Perspectives d’accès ferroviaire aux pôles franciliens depuis le Bassin Parisien.
[21] ITF (2019), High Capacity Transport: Towards Efficient, Safe and Sustainable Road Freight, International Transport Forum, OECD, Paris.
[8] ITF (2019), ITF Transport Infrastructure Database, International Transport Forum, OECD, Paris, https://doi.org/10.1787/trsprt-data-en (accessed on 29 July 2019).
[27] ITF (2019), ITF Transport Outlook 2019, OECD Publishing, Paris, https://doi.org/10.1787/transp_outlook-en-2019-en.
[42] ITF (2018), Policies to Extend the Life of Road Assets, International Transport Forum, OECD, Paris.
[44] ITF (2018), Safer Roads with Automated Vehicles?, International Transport Forum, OECD, Paris.
[43] ITF (2018), Towards Road Freight Decarbonisation: Trends, Measure and Policies, International Transport Forum, OECD, Paris.
[47] ITF (2017), ITF Transport Outlook 2017, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789282108000-en.
[29] ITF (2014), The Economics of Investment in High-Speed Rail, ITF Round Tables, No. 155, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789282107751-en.
[87] Kahn, M. (2007), “Gentrification trends in new transit-oriented communities: Evidence from 14 cities that expanded and built rail transit systems”, Real Estate Economics, Vol. 35/2, pp. 155-182.
[46] Lee, N. and A. Lembcke (2020), “The economic benefits of accessibility: A survey”, Unpublished Manuscript.
[48] Martí-Henneberg, J. (2013), “European integration and national models for railway networks (1840-2010)”, Journal of Transport Geography.
[85] Mayer, T. and C. Trevien (2017), “The impact of urban public transportation evidence from the Paris region”, Journal of Urban Economics, Vol. 102, pp. 1-21.
[10] McKinsey & Company (2016), Bridging Global Infrastructure Gaps.
[40] McKinsey & Company (2013), “Infrastructure productivity: How to save $1 trillion a year”.
[76] Michaels, G. (2008), “The effect of trade on the demand for skill: Evidence from the interstate highway system”.
[11] OECD (2020), “Building back better: A sustainable, resilient recovery after COVID-19”, Tackling Coronavirus (COVID-19): Contributing to a Global Effort, OECD, Paris, http://www.oecd.org/coronavirus/policy-responses/building-back-better-a-sustainable-resilient-recovery-after-covid-19-52b869f5/ (accessed on 10 June 2020).
[45] OECD (2020), Improving Transport Planning for Accessible Cities, OECD Urban Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/fcb2eae0-en.
[37] OECD (2019), Accelerating Climate Action: Refocusing Policies Through a Well-Being Lens - Highlights, OECD, Paris.
[19] OECD (2019), Accelerating Climate Action: Refocusing Policies through a Well-being Lens, OECD Publishing, Paris, https://dx.doi.org/10.1787/2f4c8c9a-en.
[18] OECD (2019), Regional Outlook 2019: Leveraging Megatrends for Cities and Rural Areas, OECD Publishing, Paris, https://doi.org/10.1787/9789264312838-en.
[25] OECD (2018), Maintaining the Momentum of Decentralisation in Ukraine, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264301436-en.
[38] OECD (2018), OECD Territorial Reviews: The Megaregion of Western Scandinavia, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264290679-en.
[2] OECD (2017), Gaps and Governance Standards of Public Infrastructure in Chile: Infrastructure Governance Review, OECD Publishing, Paris, https://doi.org/10.1787/9789264278875-en.
[41] OECD (2016), OECD Regional Outlook 2016: Productive Regions for Inclusive Societies, OECD Publishing, Paris, https://doi.org/10.1787/9789264260245-en.
[4] OECD (2016), OECD Territorial Reviews: Japan 2016, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264250543-en.
[3] OECD (2016), Road Infrastructure, Inclusive Development and Traffic Safety in Korea, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264255517-en.
[14] OECD (2015), “Chicago, United States”, in Governing the City, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264226500-9-en.
[80] OECD (2015), OECD Urban Policy Reviews: Mexico 2015: Transforming Urban Policy and Housing Finance, OECD Urban Policy Reviews, OECD Publishing, Paris, https://doi.org/10.1787/9789264227293-en.
[13] OECD (2012), OECD Territorial Reviews: The Chicago Tri-State Metropolitan Area, United States 2012, OECD Territorial Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264170315-en.
[101] OECD/European Commission (2020), Cities in the World: A New Perspective on Urbanisation, OECD Urban Studies, OECD Publishing, Paris, https://doi.org/10.1787/d0efcbda-en.
[82] Pogonyi, C., D. Graham and J. Carbo (2019), “Metros, agglomeration and firm productivity. Evidence from London”, SSRN Electronic Journal.
[32] Puga, D. (2002), “Progress report on the implementation of the TEN-T Network in 2014 and 2015”, Journal of Economic Geography, Vol. 2, pp. 373-406.
[105] Redding, S. and M. Turner (2015), “Transportation costs and the spatial organization of economic activity”, in Handbook of Regional and Urban Economics, Elsevier B.V.
[58] Sanchis-Guarner, R. (2012), Driving up Wages: The Effects of Road Construction in Great Britain.
[26] Stelder, D. (2016), “Regional accessibility trends in Europe: Road infrastructure, 1957-2012”, Regional Studies, Vol. 50/6, pp. 983-995.
[106] Storeygard, A. (2016), “Farther on down the road: Transport costs, trade and urban growth in Sub-Saharan Africa”, The Review of Economic Studies, Vol. 83/3, pp. 1263-1295.
[100] Transport Analysis (n.d.), “Transport indicators week 45”, https://www.trafa.se/en/pages/transport-indicators/ (accessed on 4 November 2020).
[98] UK Government (2020), Transport Use during the Coronavirus (COVID-19) Pandemic, https://www.gov.uk/government/statistics/transport-use-during-the-coronavirus-covid-19-pandemic (accessed on 3 November 2020).
[16] UNECE (2016), “Infrastructure and growth”, Sustainable Development Brief, No. 3, http://www.unece.org/?i (accessed on 5 September 2019).
[73] Venables, A., J. Laird and H. Overman (2014), “Transport investment and economic performance: Implications for project appraisal”.
[74] Vickerman, R., K. Spiekermann and M. Wegener (1999), “Accessibility and economic development in Europe”, Regional Studies, Vol. 33/1, pp. 1-15, http://dx.doi.org/10.1080/00343409950118878.
[9] World Bank (2018), Belt and Road Initiative, https://www.worldbank.org/en/topic/regional-integration/brief/belt-and-road-initiative.
[1] World Economic Forum (2016), The Global Competitiveness Report 2016-2017, World Economic Forum, Geneva, http://www3.weforum.org/docs/GCR2016-2017/05FullReport/TheGlobalCompetitivenessReport2016-2017_FINAL.pdf (accessed on 5 September 2019).
Annex 1.A. Literature
Annex Table 1.A.1. List of studies on the economic impact of interregional transport infrastructure investment
Interregional transport |
Type of infrastructure investment |
Period |
|
---|---|---|---|
France |
Commuter train |
To be realised |
|
Germany |
Railway |
1840-71 |
|
Germany |
High-speed railway |
||
Portugal |
Highway |
1986-97 |
|
Portugal |
Highway |
2010-11 |
|
Spain |
High-speed railway |
||
Spain |
Highway |
1980-94 |
|
Sweden |
Railway |
1855-70 |
|
Switzerland |
Highway |
1960-2010 |
|
Switzerland |
Railway |
19th century |
|
United Kingdom |
Highway |
2002-08 |
|
United Kingdom |
Highway |
1997-2008 |
|
United States |
Highway |
1953-89 |
|
United States |
Highway |
1969-93 |
|
United States |
Highway |
1959-75 |
|
United States |
Railway |
1850-60 |
|
United States |
Highway |
2007 |
|
United States |
Railway |
1870-90 |
Note: See Duranton, G. et al. (2015[97]), Handbook of Regional and Urban Economics, Elsevier, for a broader survey of the literature that includes a larger geographical scope.
Notes
← 1. As a consequence of the increasing opportunities for trade and pressure on scarce resources, the construction of the railroad network in the United States led to a 60% increase in the value of agricultural land (Donaldson and Hornbeck, 2015[67]).
← 2. Other than passenger traffic, the Channel tunnel has also eased freight traffic. More than one‑quarter of total trade in goods between the UK and continental Europe goes through the Channel tunnel, which therefore functions as one of the main arteries for the exchange of goods in Europe.
← 3. The term productivity spill-overs refers to the diffusion of partially non-excludable knowledge among neighbouring firms that interact by means of supply agreements, participation in formal and informal business community gatherings, etc. See further the section on benefits from road and rail infrastructure for a broader description of the positive externalities resulting from agglomeration.
← 4. The analysis presented in this report on interregional transport infrastructure focuses on inland transport, in great part accounted for by rail and road transport. It therefore excludes transport via maritime ports and airports.
← 5. Total inland infrastructure investment and maintenance spending includes road and rail transport. Calculations based on ITF (2019[8]).
← 6. The European Regional Development Fund (ERDF) and the Cohesion Fund (CF) are the two main arms of EU Regional Policy. The ERDF aims to strengthen economic and social cohesion in the European Union by correcting imbalances between its regions and therefore involves all EU regions. On the other hand, the CF, whose goal is also to reduce economic and social disparities and promote sustainable development, is aimed at member states whose gross national income (GNI) per inhabitant is less than 90% of the EU average (https://ec.europa.eu/regional_policy/en/policy/what/investment-policy/). For information on how the funds are allocated, see https://cohesiondata.ec.europa.eu/overview.
← 7. The proportion of the world population living in cities of at least 50 000 inhabitants has increased from 37% to 48% during 1975-2015 (OECD/European Commission, 2020[101]).
← 8. This is also evident in the common global ambition set out in the Sustainable Development Goals that are part of the 2030 Agenda for Sustainable Development. Seven out of the 17 goals are transport-related (ITF, 2019[27]).
← 9. Based on UK Government (2020[98]).
← 10. The UK Department of Transport provides appraisals guidelines that recognise the importance of additional benefits accruing from the reorganisation of economic activity (Venables, Laird and Overman, 2014[73]), thus positioning itself as world leader in terms of the innovativeness of transport project appraisals.
← 11. Based on Government of Canada (2020[99]).
← 12. Based on Transport Analysis (n.d.[100]).
← 13. Three assumptions guarantee the minimisation of the impact on total expenditure: i) costs preventing higher-capacity utilisation of the rail network are minimised; ii) revenues from rail system are maximised (e.g. via land value capture systems); and iii) an effective system of taxes addressing environmental externalities is in place.
← 14. The empirical evidence is relatively scarce due primarily to the young age of many HS rail investment projects. See OECD (2018[38]) for a review of the existing evidence as part of the discussion on how HS rail investment could contribute to the integration of the megaregion of Western Scandinavia.
← 15. The other important factors on which HS rail investment profitability typically rests in cost-benefit analyses are savings in construction costs and savings in travel time.
← 16. See, for instance, Baum-Snow (2007[102]).
← 17. Several recent papers summarise this literature (Redding and Turner, 2015[105]; Gibbons et al., 2019[54]).
← 18. The studies include historic and contemporaneous infrastructure investments and were chosen for their application of empirical methods that allow the identification of causal effects. List of relevant studies in Annex Table 1.A.1.
← 19. The slowdown in international trade in 2019-20 due to rising tariffs highlights this point. Rising bilateral tensions could reduce the global level of output by over 0.5% in 2020 (ITF, 2019[27]) relative to baseline OECD projections.
← 20. As another example of the complementarity between transport infrastructure and other cost-reducing measures, looking at a set of 15 coastal Sub-Saharan countries, the 300% oil price increase between 2002 and 2008 induced the income of cities near the capital and port city to increase by 7% relative to otherwise identical cities 500 kilometres farther away (Storeygard, 2016[106]).
← 21. The expected growth rate in market access given the growth rate in highway length refers to the growth rate in market access returned by a linear regression on the growth rate in highway length as unique regressor.
← 22. Other studies confirming the positive impact on employment are (Audretsch, Dohse and Pereira dos Santos, 2017[57]; Duranton and Turner, 2012[91]).
← 23. Data on population growth tend to be more easily available given the early period (second half of the 19th century) when most of these projects took place in developed countries.
← 24. The Grand Paris Express is a group of rapid transit lines being built in the Île-de-France region, consisting of about 200 km and connecting 68 train stations.
← 25. Baum-Snow (2007[102]) shows that each new highway caused central city population in US cities to decline by about 18% between 1950 and 1990. Garcia-López et al. (2015[104]) confirm the same findings for Spain, Baum-Snow et al. (2017[90]) for China and Fretz et al. (2017[75]) for Switzerland.