The Mobility - Productivity Paradox Exploring The Negative Relationships Between Mobility and Economic Productivity

This paper explores a paradox: negative correlations between indicators of mobility (such as VMT) and productivity (such as GDP), and positive correlations between mobility constraints (higher road use prices or traffic congestion) and productivity. These relationships contradict common assumptions that policies and projects that increase vehicle travel (roadway expansions and lower road user prices) increase productivity and support economic development. This paradox can be explained by the following: First, motor vehicle travel is just one of many factors affecting overall accessibility, and planning decisions often involve trade-offs between mobility and other accessibility factors such as the quality of other modes and land use accessibility. Second, many policies that increase mobility violate efficient market principles, which tends to reduce productivity. Third, motor vehicle travel is resource intensive, so increases in such travel increase various costs, including costs borne by industry. Fourth, increased vehicle travel increases the portion of household budgets devoted to vehicles and fuel, expenditures that generate low regional employment and business activity. This paper examines these issues, describes empirical evidence of these impacts, and discusses their implications. Key Word Economic development, Productivity, Economic Efficiency, Transport Policy The Mobility-Productivity Paradox Victoria Transport Policy Institute 2 Introduction Many current policies and planning practices reflect the assumption that constraints on motor vehicle travel (traffic congestion, high fuel prices, road tolls, etc.) reduce economic productivity, and policies which increase vehicle travel (roadway expansions, low road user fees, etc.) increase productivity and support economic development. However, there are reasons to question those assumptions. Certainly, motor vehicle travel is an important input in most economic activities: it delivers raw materials to producers, goods to markets, employees to work, students to schools, and customers to markets. All else being equal, an increase in transport system efficiency should increase productivity. Increased transportation efficiency has contributed significantly to past economic productivity gains. But motor vehicle travel also imposes significant costs. Evidence described in this paper indicates that in regions with high levels of mobility, a significant portion of vehicle travel is economically inefficient: vehicle travel that consumers would forego if they had better options and more efficient pricing, which increases total transportation costs, including costs to businesses. In such circumstances, policies that reduce vehicle travel can increase productivity and support economic development. This paper explores this paradox. It discusses ways that accessibility and mobility affect economic productivity, examines evidence of the relationships between mobility and economic productivity, and discusses their implications. How Accessibility and Mobility Affects Productivity This section discusses various ways that mobility affects economic productivity. Access to Productive Activities Conventional planning tends to evaluate transport system performance based primarily on mobility, using indicators of vehicle travel speed and delay such as roadway level-ofservice and average traffic speed. However, mobility is seldom an end in itself, the ultimate goal of most transportation is access to services and activities (retails, employment, education, recreation, etc.). Several factors affect accessibility (1, 2):  Motor vehicle travel.  The quality of other modes (walking, cycling, ridesharing, public transport, etc.), including mobility substitutes such as telecommunications and delivery services.  Transport network connectivity (the quality of connections between paths, roads, and different modes).  Land use accessibility, which is affected by development density and mix. Planning decisions often involve trade-offs between these. For example, expanding urban roadways tends to improve automobile access but creates a barrier that reduces pedestrian and bicycle access, and therefore public transit access since most transit trips include walking and cycling links. Similarly, urban fringe locations that are easy to access by automobile tend to be difficult to access by other modes. As a result, the benefits of increased mobility are often partly offset by declines in other forms of access, reducing The Mobility-Productivity Paradox Victoria Transport Policy Institute 3 net efficiency gains. A newer planning paradigm evaluates transport system performance based on overall accessibility, not just mobility (3, 4). Recent research improves our understanding of how land use factors affect accessibility:  Kuzmyak found that travelers in more compact neighborhoods experience less congestion than in more sprawled, suburban neighborhoods due to better travel options, more connected streets, and shorter trip distances (5).  Levine, et al. found that changes in development density affect the number of jobs and services available within a given travel time about ten times more than proportional changes in traffic speed (6).  A study that measured the number of jobs accessible by automobile within certain time periods for the 51 largest US metropolitan areas found that the five cities with the most intense congestion (highest Travel Time Index ratings) are among the best for automobile employment access because their lower traffic speeds are more than offset by higher employment densities which reduce commute distances (7).  Cortright found that roadway expansions that stimulate sprawl can increase total travel times because higher traffic speeds are more than offset by longer travel distances (8). These studies indicate that transport system changes intended to increase vehicle traffic speeds often reduce overall accessibility thereby reducing the efficiency of other modes and stimulating more dispersed development. Certain types of accessibility most directly affect productivity, including commercial deliveries (freight, service vehicles, etc.), business travel, and commuting to work and school. Reducing the resource (time, vehicle, fuel) costs of such travel tends to increase productivity. The magnitude of these impacts varies depending on the type of industry and conditions. For example, interregional shipping is a major portion of resource and bulk retail industry costs. Local services, such as plumbers and utilities, are affected by local travel conditions, including traffic speeds, congestion, and land use accessibility. Commuting is a major input in service industries (retail, restaurants, hotels, etc.) and therefore businesses’ ability to attract and retain suitable employees. Changes in these transport costs can affect those industries’ productivity. Although most high-value freight is transported by truck, most local services are distributed by motor vehicle, and most commuting is by automobile, alternative modes, more accessible land use patterns, and demand management strategies are sometimes the most cost effective way to improve accessibility, in which case they can provide the greatest productivity gains. For example, road pricing that gives priority to commercial and high occupant vehicles on congested roads, more compact and mixed development, and commute trip reduction programs that shift travel from automobiles to higher occupant vehicles, can improve accessibility while reducing total vehicle travel. The Mobility-Productivity Paradox Victoria Transport Policy Institute 4 Economic Efficiency There are two basic requirements for economic efficiency: 1. Consumer sovereignty, which means that consumers can choose the goods they demand (for which they are willing to pay marginal costs). As a result, efficiency can increase with transport system diversity if it allows users to choose the combination of modes, services and qualities that best meet their needs. There is little reason to maintain options with minimal demand (for example, cycling facilities or transit services that attract few users), but transport system efficiency is likely to increase if alternative modes receive at least as much support as automobile transport, and often more for equity sake (to provide basic mobility for non-drivers and affordable mobility for lower-income people), and to help achieve strategic objectives (such as preserving openspace and increasing public health). For example, if society spends $5.00 on roads and parking facilities to accommodate an automobile commute it should be willing to devote at least that much for other commute modes, and often more for equity sake and to achieve other objectives. 2. Efficient pricing, which means that prices (direct costs to users for consuming a good) reflect the full marginal costs of producing that good, unless a subsidy is specifically justified. This tests users’ willingness-to-pay for the goods they consume so society does not spend $2 on a good (including roads and parking facilities) that users only value at $1. As a result, economic efficiency tends to increase if travelers are charged for the costs they impose, including congestion, road and parking facilities, accidents, and pollution. People sometimes assume that policies that make vehicle travel cheaper increase productivity, but this is only true of true resource savings; economic transfers that externalize costs tend to reduce productivity. For example, if roads are financed through general taxes rather than user fees, savings to motorists will be offset by higher costs elsewhere in the economy, and cheaper vehicle costs are likely to induce additional vehicle travel that increases total transport costs, including externalities such as traffic congestion, parking subsidies, accidents, pollution damages, and sprawl-related costs. Table 1 summarizes various transportation market requirements, distortions, reforms and their travel impacts. Although these distortions may individually seem modest and justified, their impacts are cumulative and synergistic (total impacts are greater than the sum of their individual impacts). For example, planning practices that undervalue active transport (walking and cycling), by ignoring the parking cost savings and health benefits they provide, can lead to underinvestment in sidewalks and bike paths, which not only reduces walking and cycling access, it also reduces public transit access, since most transit trips include links by these modes. Similarly, underpricing road use (for example, by financing roads through general taxes rather than user fees) not only increases traffic congestion and roadway costs, by inducing additional vehicle travel it increases parking costs, accidents and pollution emissions. Conversely, underpricing parking facility use, by including them as building costs instead of charging users directly not only increases the number of parking spaces needed, by inducing additional vehicle travel it also increases traffic congestion, accidents and pollution costs. The Mobility-Productivity Paradox Victoria Transport Policy Institute 5 Table 1 Market Principles, Distortions, Reforms and Travel Impacts (9) Market Requirements Common Distortions Potential Reforms Travel Impacts Optimal transport planning. Planning practices should be comprehensive and multi-modal, considering all impacts and options, and investing in the most cost-effective option overall. Conventional planning evaluates transport system performance based primarily on automobile travel conditions, and tends to overlook benefits of other modes (parking savings, affordability, basic mobility for non-drivers, environmental benefits). This favors automobile-oriented transport improvements. A major portion of transport funds are dedicated to roads and parking facilities and cannot be used for alternative solutions, even if they most cost effective. Evaluate transport system performance using multimodal level-of-service ratings and other indicators of overall accessibility. Comprehensive and multimodal planning which considers all impacts, modes and improvement options, including demand management strategies. Least-cost planning invests in the most cost effective options, considering all impacts (benefits and costs). More comprehensive and neutral planning could significantly improve transport options (walking, cycling, public transit, carsharing, etc.) and support demand management, reducing automobile travel 1020%. Integrated planning. Integrate planning by different agencies and jurisdictions to help optimize overall accessibility. Current land use planning practices, such as land density limits and generous minimum parking requirements, result in dispersed, automobiledependent communities. Integrate transport and land use planning to create more accessible, multimodal communities. Residents of more accessible, multi-modal communities tend to drive 10-40% less than they would in automobile dependent areas Efficient pricing. Prices should reflect marginal costs unless a subsidy is specifically justified. Motor vehicle travel is significantly underpriced. Many costs are either fixed or external. Cost-based pricing of roads, parking, insurance and vehicle fuel. Efficient pricing is likely to reduce automobile travel 20-40%. Efficient markets reflect certain principles. Current transport planning and pricing often violate these principles in ways that stimulate mobility. Transport market reforms, such as more neutral and multimodal transport planning and more efficient pricing, tend to reduce motor vehicle travel. If these reforms were fully implemented, per capita vehicle travel would probably decline 35-50%. This suggests that a significant portion of current vehicle travel is economically inefficient; it results from market distortions that reduce transport options, disperse development, and underprice vehicle travel. With more efficient planning and pricing, travelers would choose to drive less, rely more on alternative modes, and be better off overall as a result (10). Other researchers reach similar conclusions, although they consider a smaller set of distortions and reforms (11, 12). Economically inefficient vehicle travel tends to reduce productivity. For example, underinvestment in alternative modes tends to increase traffic and parking congestion, and reduces non-drivers’ access to schools and jobs, and therefore the labor pool available to businesses, particularly service industries that require lower-wage workers). Transport underpricing imposes costs on other economic sectors, as discussed in the following section. The Mobility-Productivity Paradox Victoria Transport Policy Institute 6 Cost Burdens Motor vehicle travel is relatively costly, including costs for vehicles and fuel, roads and parking facilities (and therefore land), accident and environmental damages. Other transport modes also impose costs, but less per passenger-mile. For example, automobile passengers typically require an order of magnitude more road space than walking, cycling and public transit, plus space for parking, which increases congestion and facility costs. Figure 1 compares estimated costs of car, bus, bicycle and walking. According to this analysis, total costs per passenger-mile (including infrastructure, vehicle, crash and pollution costs) are approximately 63¢ for cars, 61¢ for bus passengers (this is relatively high because most bus travel occurs under urban conditions; car travel costs more than $1.00 per passenger-mile under such conditions), 19¢ for cycling and 14¢ for walking. Figure 1 Estimated Average Costs for Car, Bus, Bicycle and Walk (13) $0.00 $0.05 $0.10 $0.15 $0.20 $0.25 Vehicle Ownership Vehicle Operation Crash Costs Parking Costs Congestion Road Facilities Traffic Services Air Pollution Noise D o ll a rs P e r P a s s e n g e rM il e Automobile Transit Bicycle Walk Automobile travel tends to have higher total costs (infrastructure, vehicles, crash and pollution costs) than most other modes. Many of these are external costs imposed on other industries, which reduce productivity. For example, businesses bear a significant portion of general taxes that finance roads, they bear substantial parking subsidy costs, traffic congestion delays and accident damages. Some industries (tourism, farming, fishing) are harmed by pollution emissions. Consumer Expenditures on Imported Goods Vehicles and fuel are capital intensive, most of the inputs are imported from other regions and countries. As a result, expenditures on these goods tend to generate less employment and business activity than most other goods. For example, a million dollars shifted from fuel expenditures to a general bundle of consumer goods adds 4.5 jobs to the U.S. economy, and each million dollars shifted from vehicle fuels to public transit operation generates 18.5 jobs (14). As a result, policies that increase motor vehicle travel, and therefore the portion of household budgets devoted to vehicles and fuel, tend to reduce regional and domestic employment and productivity. The Mobility-Productivity Paradox Victoria Transport Policy Institute 7 Productivity Impact Summary Table 2 summarizes how four types of transportation policies affect mobility and the four categories of productivity impacts just described. Table 2 Comparing Transportation Improvement Strategies Unpriced Road Expansions Improve Alt. Modes Eff. Transport Pricing Smart Growth Policies Mobility impacts Expanding unpriced roads increases mobility. Improving alternative modes reduces mobility. More efficient pricing reduces mobility. More compact, multimodal development reduces mobility. Accessibility for productive activities Increases automobile access but can reduce other forms of access. Can increase all types of access, including access for non-drivers. Tends to increase higher-value access (such as freight and service vehicles). Tends to increase overall accessibility. Economic efficiency Generally economically inefficient (would not be justified if users paid the incremental costs). If demand exists for alternative modes, improving them tends to increase efficiency. Increases economic efficiency (assuming price reforms reflect market principles). If demand exists for more accessible development, smart growth policies increase efficiency. Cost burdens By increasing total vehicle travel it tends to increase total transport costs, including many external costs borne by industries. By reducing total vehicle travel it tends to reduce total transport costs, including many external costs borne by industries. By reducing total vehicle travel it tends to reduce total transport costs, including many external costs borne by industries By reducing total vehicle travel it tends to reduce total transport costs, including many external costs borne by industries. Consumer expenditures on imported goods Tends to increase total vehicle and fuel expenditures. Tends to reduce total vehicle and fuel expenditures. Tends to reduce total vehicle and fuel expenditures. Tends to reduce total vehicle and fuel expenditures. Unpriced (or underpriced) roadway expansions tend to increase mobility, which increases automobile access (at least in the short-run) but tends to reduce other forms of access, is generally economically inefficient, tends to increase total transportation costs (including many costs to industry) and increases consumer expenditures on imported goods. The Mobility-Productivity Paradox Victoria Transport Policy Institute 8 Empirical Evidence This section discusses empirical evidence of the relationships between mobility and economic productivity. Evidence of Positive Relationships Between Mobility and Productivity There is some evidence of a positive relationship between mobility and economic productivity. For example, the Highway Users Alliance claims that the graph below proves that, because VMT and GDP have historically been correlated, efforts to reduce vehicle travel must reduce economic productivity. Figure 2 US VMT and GDP Trends (15) The Highway Users Alliance claims that this graph proves that reductions in vehicle travel will reduce economic productivity, but correlation does not prove causation. Similarly, economist Randall Pozdena claims that the correlation between income and energy use shown in Figure 3, and because recessions often follow petroleum price spikes, efforts to reduce per capita vehicle travel reduce economic productivity. He concludes that, “a one percent change in VMT/capita causes a 0.9 percent change in GDP in the short run (2 years) and a 0.46 percent in the long run (20 years).” Certainly energy use, vehicle travel and GDP tend to increase together, but much of this effect is the result of increased wealth allowing consumers to purchase more vehicles and fuel, so increased VMT is an productivity output rather than an input. The log-log format in Figure 3 exaggerates the relationships between energy and economic development. For example, although the U.S. and Norway appear close together, Norwegians actually consume about half as much fuel per capita as U.S. residents. The graph includes countries with very different levels of industrialization. An increase in per capita vehicle travel in low income countries such as Zimbabwe or Laos has a very different productivity impacts than in wealthy, countries like the U.S. or Norway. The Mobility-Productivity Paradox Victoria Transport Policy Institute 9 Figure 3 Per Capita GDP Versus Barrels of Oil (16) Pozdena claims this graph proves that increased energy consumption increases economic productivity. A loglog graph such as this exaggerates such relationships. Evidence Of Negative Relationships Between Mobility and Productivity Many researchers find weak or negative relationships between personal vehicle travel and economic productivity (17, 18, 19). Increasing from very low to moderate levels of mobility tends to increase productivity, since motor vehicles are used for high-value trips, but as mobility increases marginal productivity benefits are likely to decline, and may become negative as external costs and inefficiencies increase (20, 21). For example, farmers and tradesmen (carpenters, bricklayers, etc.) can significantly increase productivity by using truck transport rather than headloading or animal carts, but there is often no increase in productivity when workers shift from commuting by bicycle or public transit to driving; such shifts can provide consumer benefits (workers have a wider range of housing options), but only if automobile travel significantly increases employers employment pool is it likely to increase productivity. On the other hand, shifting from bicycle or transit to automobile commutes tends to increase some costs – costs of roads and parking facilities, congestion, accident and pollution – which is likely to reduce productivity. Among wealthy countries there is considerable variation in per capita vehicle travel. Although per capita VMT grew during most of the last century, it has become saturated in most wealthy countries and the level at which this saturation occurs varies depending on transport and land use policies (22). The U.S. averages more than twice the per capita vehicle travel as most other OECD countries, as indicated in Figure 4. Of particular interest is Norway, which produces petroleum but maintains high fuel prices and has other policies to discourage vehicle travel and support alternative modes. These policies minimized domestic fuel consumption, leaving more oil to export. As a result, Norway has one of the world’s highest incomes, a competitive and expanding economy, a positive trade balance, and the world’s largest legacy fund. The Mobility-Productivity Paradox Victoria Transport Policy Institute 10 Figure 4 Per Capita Annual Vehicle Travel By Country (23) 0 5,000 10,000 15,000 20,000 25,000 Un ite d S tat es Ita ly Ca na da De nm ark Fr an ce Fin lan d Sw itz erl an d No rw ay Sw ed en Un ite d K ing do m Ge rm an y Ne the rla nd s Sp ain Ja pa n Po lan d A n n u a l V e h ic le -K m s P e r C a p it a Per capita vehicle mileage is significantly higher in the U.S. than in other industrialized countries. Residents of wealthy countries such as Switzerland, Norway and Sweden drive about half as much as in the U.S. due to policies and planning practices that increase transport system efficiency. Similarly, annual per capita vehicle mileage varies significantly among U.S. cities, from fewer than 5,000 average annual vehicle-miles per capita to more than 15,000. Although many factors influence these differences, they result, in part, from transport and land use policies that affect the travel options available, travel incentives, and land use patterns. There is no evidence that lower VMT cities such as New York, Sacramento, Chicago and Portland, are less economically successful than higher VMT cities such as Atlanta, Houston or Birmingham; in fact, the lower VMT cities tend to have higher per capita GDP, as indicated later in this paper. The amount of vehicle travel and energy required per unit of GDP varies widely. Virtually all developed countries are increasing GDP per unit of energy and mobility called decoupling (24, 25). Some extract far more productivity (material wealth and income) per unit of mobility and energy than others (Figure 5). Figure 5 GDP per Passenger-Kilometer for Various Countries (26) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 197