Better roads increase emissions

Many people associate higher road standards with lots of benefits, but this does not prevent road improvements from exerting a largely negative effect on the global climate

Written By Arvid Strand, Petter Næss and Aud Tennøy, Institute of Transport Economics, Norway

Transport infrastructure development, including road construction, has always had as its main purpose to facilitate easier transfer of commodities and skilled labor, thereby increasing the efficiency of and contributing to economic growth. Generating as much transport as possible is, however, not an objective in its own right. The challenges and problems associated with transport show that increasing transport volumes bring about undesirable impacts as well as benefits. The influence of road development on the amount of transport and the distribution between transport modes therefore relates to a discussion that has been going on for decades.
    
The study on which this article is based was funded by the Norwegian Road Directorate who wanted us to shed light on the following main question: how does road construction influence on greenhouse gas emissions? We have elucidated this issue by addressing three sub-topics: whether, how and to what extent improved road standard influences greenhouse gas emissions 1) by reducing the emissions per vehicle km driven; 2) by stimulating growth in car traffic and thereby increasing greenhouse gas emissions; and 3) through the greenhouse gas emissions resulting from construction, operation and maintenance of the road network.
    
The study concludes that the question of whether better roads contribute to lower greenhouse gas emissions must be answered mainly negatively. In most cases, the construction of better roads leads to increased greenhouse gas emissions. There are several reasons for this. For one thing, improved road quality facilitates faster driving, often at speed levels where faster driving causes emissions to increase considerably (above 80 km/h). Emissions also increase because people make on average more and longer trips, and because improved conditions for car travelling cause some previous trips by public or non-motorised modes of transport to be replaced by trips by car. Moreover, the road construction itself and the operation and maintenance of the expanded roads require energy use and thereby contribute to increase greenhouse gas emissions. These last components make up an increasing part of the emissions, the lower the traffic volumes are along the new or widened links.

Emissions per vehicle kilometer
The research literature offers somewhat divergent answers regarding the optimal speeds. Seen from the perspective of reducing CO2 emissions per vehicle km, speed levels between 50 and 70 km/h are apparently the most optimal, but the curve is fairly flat over the entire span from 50 to 90 km/h.
    
Estimates produced by the IEA still suggest that considerable amounts of CO2 emissions could be avoided by reducing average speeds from 90 to 70 km/h and from 110 to 90 km/h. This may be important to the discussion on the topic in Norway. Some of the proposed Norwegian road schemes entail a widening of existing two-lane roads where the average speed is somewhat above 80 km/h (legal speed limit 80 km/h) into four-lane motorways with speed limits of 90 or 100 km/h. Other projects are about improving the standard of two-lane roads in such a way that the speed limit may be raised from 70 to 80 km/h or from 80 to 90 km/h. If we consider 50-90 km/h as the optimal speed interval, such road standard improvements will not influence climate gas emissions per vehicle km to any extent worth mentioning. However, if we, in line with IEA, assume that a speed increase from e.g. 70 to 90 km/h causes the fuel consumption of cars to increase by 23 per cent, we have to judge differently. This is even more the case if real speed levels exceed the legal speed limits (which they evidently often do).
    
Our summarising of the literature on speed changes and greenhouse gas emissions can be seen in the table above.
    
We have not found any studies addressing specifically how the road standard influences on the smoothness of driving. The studies focus more on the behavior of drivers or driving style in general. Based on the literature on driving style it is, however, possible to identify the intervals within which smooth driving can contribute to reduced fuel consumption and CO2 emissions. From this, we cautiously estimate that smoother driving facilitated by better road standard can, other things equal, contribute to a reduction in emissions of between 5 and 15 per cent. This of course depends a lot on the kind of road standard improvement and the context in which the change in standard is introduced.

The influence of road standard
The theoretical reasons why road construction contributes to traffic growth by making it faster and/or more comfortable to drive have been well understood for a long time. They are based on the simple economic theory of supply and demand. Better roads make it easier to choose more distant trip destinations as well as to make more trips. Both these changes in travel behavior contribute to increase the amount of transport. In addition, road capacity expansion in congested areas often changes the relative competitiveness of the private car and other travel modes, thus causing additional growth in the number of motor vehicles on the roads.
    
In summary, road capacity increases could be expected to result in generated and induced traffic by influencing:

• route choice
• the proportion who choose to travel in the peak period 
• the amount of transport 
• the distribution of the transport between different modes
• land use (in a longer term)
• the standard of the public transport services (in a longer term)

Among these six effects, the four latter ones contribute to induced traffic (i.e. an increase in the total traffic volume), whereas the two former contribute to generated, but not induced traffic (i.e. traffic diverted from other roads or times of the day, causing congestion to arise anew on the new or expanded road). Changes in the amount of transport (longer and/or more frequent trips) and in the modal split (a higher proportion of traffic by car) occur in a relatively short time after the road capacity expansion and are amplified by long term effects, notably more dispersed patterns of urban development. This contributes in at least three ways to increase car traffic: 1) the urban structure as a whole is more transport-requiring because the distances are longer; 2) the longer distances make it less attractive to walk or go by bike; and 3) it is difficult to provide efficient public transport services in such a dispersed urban structure. The combination of the short-term change from public transport to car and the long-term land use changes weakens the ability of public transport to make revenues from ticket sales. In a longer term, this can reduce the quality of the public transport services.
    
Considerable empirical evidence is now available, showing that the traffic-inducing effect of road construction reducing travel time by car is not only a theoretical construct but a tangible real-life phenomenon. In particular, this is the case for road capacity increases in congested urban areas, where road expansions contribute to increased amounts of transport as well as higher proportions of travel by car. However, in rural areas too, improved road standard will often contribute to traffic growth by stimulating to longer and more frequent trips.
    
A number of investigations in the United Kingdom and the USA indicate that road construction reducing travel time by 10 per cent typically leads to an immediate increase in traffic by 3-5 per cent and a long-term traffic growth of between 5 and 10 per cent. In congested areas, a 10 per cent increase in road capacity (measured in the number of kilometers of driving lanes) seems to result in short- and long-term traffic increases of a similar magnitude as for reduced travel times, i.e. 3-5 per cent in a short term and between 5 and 10 per cent in the long term.

Emissions from construction
The literature on these issues is not impressively extensive. Available Norwegian analyses suggest a total annual emission of 21 tons of CO2 equivalents per lane kilometer due to the construction of a ‘standard’ four-lane road. This estimate is based on an assumed lifetime of 40 years. For a ‘standard’ two-lane road, the corresponding figure is 12 tons of CO2 equivalents per lane kilometer, i.e. only 60 per cent of the emissions resulting from the construction of each lane of a four-lane road.
    
Similarly, the literature suggests that the operation and maintenance of a two-lane road entails on average annual emissions of 33 tons of CO2 equivalents per lane kilometer, with a corresponding figure of 51.5 tons per lane kilometer for four-lane roads. Judged from this, operation and maintenance of two-lane as well as four-lane roads represent considerably higher emissions than those resulting from the construction of these roads.

Conclusions
The results of our literature reviews, as well as supplementary transport model simulations in selected transport corridors, show that the conclusions about the climate impacts of road improvements vary to some extent with the geographical context. We have illustrated our findings for three prototypical situations: local schemes within a large city/metropolitan area; local schemes in smaller towns and villages; and intercity schemes. In none of these example situations is road construction a measure contributing to reducing greenhouse gas emissions.
    
We do not claim that situations cannot be imagined where road construction may have such an emission-reducing effect. This could, for example, be the case in situations where a new road tunnel replaces a steep and winding mountain crossing, thus reducing the driving distance as well as the fuel consumption per kilometer (although the construction of the tunnel would cause substantial emissions). However, we consider the situations where road construction contributes to reduced greenhouse gas emissions as clearly exceptional. In most cases, construction and maintenance of the new or improved roads will, together with speed increase and the direct and indirect impacts of induced traffic, contribute to increase the emissions of greenhouse gases. In metropolitan areas, the increases in emissions can be substantial if the road construction increases the road capacity in congested transport corridors.

Arvid Strand, Petter Næss, Aud Tennøy & Christian Steinsland: Gir bedre veger mindre klimagassutslipp? (Does road improvement decrease greenhouse gas emissions?) TØI rapport 1027/2009