The transportation sector and low-carbon growth pathways: modelling urban, infrastructure, and spatial determinants of mobility

Abstract

This article contributes to the controversial debate over the effect of spatial organization on CO₂ emissions by investigating the potential of infrastructure measures that favour lower mobility in achieving the transition to a low-carbon economy. The energy–economy–environment (E3) IMACLIM-R model is used to provide a detailed representation of passenger and freight transportation. Unlike many of the E3 models used to simulate mitigation options, IMACLIM-R represents both the technological and behavioural determinants of mobility. By comparing business-as-usual, carbon price only, and carbon price combined with transport policy scenarios, it is demonstrated that the measures that foster a modal shift towards low-carbon modes and a decoupling of mobility needs from economic activity significantly modify the sectoral distribution of mitigation efforts and reduce the level of carbon tax necessary to reach a given climate target relative to a ‘carbon price only’ policy.

Policy relevance

Curbing carbon emissions from transport activities is necessary in order to reach mitigation targets, but it poses a challenge for policy makers. The transport sector has two peculiarities: a weak ability to react to standard pricing measures (which encourages richer policy interventions) and a dependence on long-lived infrastructure (which imposes a delay between policy interventions and effective action). To address these problems, a framework is proposed for analysing the role of transport-specific measures adopted complementarily to carbon pricing in the context of international climate policies. Consideration is given to alternative approaches such as infrastructure measures designed to control mobility through less mobility-intensive denser agglomerations, investment reorientation towards public mode, and logistics reorganization towards less mobility-dependent production processes. Such measures can significantly reduce transport emissions in the long term and hence would moderate an increase in the carbon price and reduce its more important detrimental impacts on the economy.

Keywords:

• infrastructure
• mitigation policy
• spatial organization
• transport

Notes

See Shalizi and Lecocq (2009) for a description for the case of the US Interstate Highway on the influence of the induced demand effect on CO₂ emissions.

Recent programmes have estimated the contribution of these behavioural changes (e.g. ITF, 2007). Barkenbus (2010) has estimated that they can contribute to a 10% reduction of fuel demand.

Two recent studies that go some way to rectifying this are Anable et al. (2012) and Brand, Tran, and Anable (2012).

Bottom-up technology-rich models rely on exogenous trends of transportation demands, and therefore have no endogenous evolution of modal choices or mobility volumes. Top-down macroeconomic models generally represent the transportation sector in nested constant elasticity of substitution production functions, so that demand changes are exclusively price-induced.

The IMACLIM-R model used in this article divides the economy into 12 regions (the US, Canada, Europe, OECD Pacific, the former Soviet Union, China, India, Brazil, the Middle East, Africa, the Rest of Asia, the Rest of Latin America), and 12 production sectors (coal, crude oil, natural gas, refined products, electricity, construction, agriculture and related industries, energy-intensive industries, air transport, sea transport, other transports, other industries and services). In addition, IMACLIM-R includes transportation with personal vehicles and non-motorized transport.

The model does not differentiate between inter- and intra-city trips, so ‘public transport’ includes both urban public transports (e.g. buses, metros) and inter-city trains.

The functional form chosen for the relation between the marginal speed, v_j , in mode j and the utilization of the transportation infrastructure capacity was . Parameters values were calibrated such that (1) v ₀ equals 700, 80, and 50 km/h for air transportation, cars, and public transport respectively, (2) v_j (1) = v ₁, where v ₁ equals 5 km/hour for all modes, and (3) the households’ maximization programme results in observed data on mobility and budget shares per mode for the calibration year 2001.

‘Electric vehicles’ implicitly represents all types of vehicles that use electricity as a service provider (including fuel cells and hydrogen vehicles).

Note that the simplified representation of climate policies used in the analysis ignores some of the factors that may actually affect the cost of climate policies, such as the inter-temporal flexibility for allocating emissions reductions (‘when flexibility’), the international redistribution of carbon tax revenues, and internal recycling towards a reduction of labour taxes (for an overview, see IPCC, 2007).

This assumption is consistent with the relatively constant percentage of commuter Vehicle Miles of Travel in the US, from 1969 to 2007, that comprise total Vehicle Miles of Travel (approximately 30%; see Santos, McGickin, Nakamoton Gray, & Liss, 2011).

For example, a land planning measure may affect the price of land, which may in turn affect the purchasing power and location decisions of households.