A growing literature in economics aims to quantify the contribution of highway infrastructure to economic activity and welfare (Alder 2017, Allen and Arkolakis 2014, 2019, Asturias et al. 2019, Baum-Snow et al. 2017, Cosar and Demir 2016, Cosar et al. 2020). Because a substantial portion of international trade takes place on the domestic highways of importing and exporting countries, it is important that policymakers understand the potential interaction between domestic transportation networks, market integration, and globalisation (e.g. Braconier and Pisu 2014 on the effect of roads on integration within Europe). This interaction is important for understanding the value of domestic infrastructure investment and weighing these against the substantial costs of building and maintaining domestic roads depicted in Figure 1.
Figure 1 Spending on road infrastructure by country and year
a) Road infrastructure spending by country
b) Road infrastructure spending by year
In a new paper (Jaworski et al. 2020), we use data on the US highway system, domestic production, consumption, ports, and domestic and international trade to study how the Interstate Highway System (IHS) mediates exposure to domestic and international trade for almost 3,000 US counties. In particular, we consider the entire highway network of the US, which comprises over four million miles. The key parts of the network are shown in Figure 2a and include nearly 50,000 miles of the IHS, over 130,000 miles of additional federal-aid highways, and 210,000 miles of state highways. These roads include the most important parts of the system for moving goods across states as well as to and from ports for subsequent international shipments. On many highway segments, users face significant congestion levels that significantly raise both domestic and international trade costs. Figure 1B shows congestion levels on the IHS measured by the Level of Service (LOS), which reflects the quality of traffic flow on each highway segment, where higher values of LOS are associated with higher congestion levels. The figure indicates that for nearly 18,000 miles or 40% of the IHS traffic congestion acts as a significant impediment to trade (and commuting).
Figure 2 Highway network and congestion levels
a) Components of the US highway network
b) Congestion on US highways
We use an endogenous specification of domestic and international trade costs that takes into account the availability of the road network and congestion levels. On the one hand, given the available network, producers choose the least costly route to transport goods from point A to point B. On the other hand, transporting goods between any pair of locations generates additional traffic and increases congestion on the relevant segments, which subsequently affects trade costs. Given these two sources of endogeneity of trade costs, quantifying the value of highways or any other domestic infrastructure presents several challenges. In particular, changes in the configuration of the transportation network (1) affect all domestic and foreign trading partners due to the geographic pattern of specialisation; (2) generate spillovers across industries and locations that are connected through the input-output structure of the economy; and (3) alter route choice and hence the level of congestion. We provide a tractable framework to address these challenges, which we then use to carry out a quantitative assessment of the value and spatial effects of US highways.
Specifically, based on a version of the model with domestic and international trade, many locations and sectors, an input-output structure of production, and costly migration, we find that the total value of the entire IHS was $619 billion in 2012 dollars. This accounts for 3.9% of US aggregate GDP in 2012. There is substantial heterogeneity in the effects of removing the IHS across US counties. The largest losses are concentrated in the Northeast and West as illustrated in Figure 3a. The results in Figure 3b show that the IHS plays an important role in shaping comparative advantage of the most remote counties. As a result, removing the IHS would significantly reshape production patterns in those counties and lead to substantial decreases in the value of total output. The decomposition of the total effects reveals that on average about 75% and 25% can be attributed to changes in domestic and international market access, respectively. The relative importance of the two effects differs significantly across counties such that the international component is relatively more important for US coastal regions.
Figure 3 Estimates of the value of the entire HIS
a) Contribution of all trade costs to the reduction in real GDP
b) Reduction of real GDP and all trade costs
In addition, we can also use the model to quantify the effects of congestion. In particular, when we remove congestion as a mechanism, the losses from removing the entire IHS fall from $619.1 to $564.4 billion. In other words, this suggests that the effects of trade-related congestion are nearly $55 billion, or 9% of the total effect of removing the IHS.
Finally, we calculate the value of 20 longest individual highways in the IHS by counterfactually removing each highway and quantifying the associated economic impact. We find that the most valuable segments of the IHS are I-5, I-10, I-80, I-95, and I-40. The most valuable highway (I-5) runs from north to south on the west coast and accounts for $55 billion in total value and for $39.7 million on a per mile basis in 2012 dollars. There is also substantial heterogeneity across the 20 highway segments we consider such that the aggregate losses are between $2.7 and $55 billion and losses per mile are between $2.5 and $39.7 million as reported in Table 1.
Table 1 Estimates of the value of individual IHS segments
Notes: The table shows results from counterfactual exercises removing the twenty longest individual segments (in miles) of the Interstate Highway System. Column 1 shows the total number of miles. Columns 2 and 3 show the total and per-mile reduction in real GDP, respectively. Columns 4 and 5 show the portion of the total reduction attributed to the domestic and international components of trade costs, respectively.
The results in Table 1 suggest that each highway segment plays an important role in providing better domestic and international market access. On average, the international component accounts for 15% of the total effect, and for certain highways it is higher than 25%. This suggests that a large part of the total value of the domestic infrastructure is due to international linkages and globalization.
Every year policymakers around the world decide how much to invest in new construction, improvements, and maintenance of domestic infrastructure. Our results suggest that these decisions must be made in conjunction with considering how improvements in the domestic transportation networks would affect domestic and international trade as well as distributional consequences for different locations within a country. In turn, the gains from international trade policy may also crucially depend on the extent of domestic market integration (Atkin and Donaldson 2015, Ramondo et al. 2016). The framework described in this column can be applied to new investment decisions and can help policymakers understand the potential gains of additional highway segments, the sources of those gains in domestic and international markets, and the aggregate and regional effects on domestic production and employment. This may be useful for quantifying the value of proposals for entirely new or substantially upgraded transportation systems in developing countries (de Soyres et al. 2020, Reed and Trubetskoy 2019) as well as updates to existing highway networks in developed countries (Gibbons et al. 2017).
Alder, S (2017), “Chinese Roads in India: The Effect of Transport Infrastructure on Economic Development,” Working Paper.
Allen, T and C Arkolakis (2014), “Trade and Topography of the Spatial Economy,” Quarterly Journal of Economics 129(3): 1085–1140.
Allen, T and C Arkolakis (2019), “The Welfare Effects of Transportation Infrastructure Improvements,” NBER Working Paper 25487.
Atkin, D, and D Donaldson (2015), “Who’s Getting Globalized? The Size and Implications of Intra-National Trade Costs,” NBER Working Paper 21439.
Asturias, J, M Garcia-Santana, and R Ramos (2019), “Competition and the Welfare Gains from Transportation Infrastructure: Evidence from the Golden Quadrilateral of India,” Journal of the European Economic Association 17(6): 1881–1940.
Baum-Snow, N, L Brandt, J V Henderson, M A Turner, and Q Zhang (2017), “Roads, Railroads, and Decentralization of Chinese Cities,” Review of Economics and Statistics 99(3): 435–448.
Braconier, H and M Pisu (2014), “Roads to Deeper European Integration,” VoxEU.org, 20 February.
Cosar, A K and B Demir (2016), “Domestic Road Infrastructure and International Trade: Evidence from Turkey,” Journal of Development Economics 118: 232–244.
Cosar, A K, B Demir, D Ghose, and N Young (2020), “Road Capacity, Domestic Trade and Regional Outcomes,” Working Paper.
de Soyres, F, A Mulabdic, and M Ruta (2020), "Common Transport Infrastructure: A Quantitative Model and Estimates from the Belt and Road Initiative," Journal of Development Economics 143.
Gibbons, S, H Overman, T Lyytikainen, and R Sanchis-Guarner (2017), "New Road Infrastructure: The Effects on Firms," VoxEU.org, 27 July.
Jaworski, T, C T Kitchens, and S Nigai (2020), “Highways and Globalization,” NBER Working Paper 27938.
Ramondo, N, A Rodriguez-Clare, and M Saborio-Rodriguez (2016), “Trade, Domestic Frictions, and Scale Effects,” American Economic Review 106 (10): 3159–84.
Reed, T and A Trubetskoy (2019), "Assessing the Value of Market Access fromBelt and Road Projects", World Bank Policy Research Working Paper No. 8815.