Click for next page ( 31


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 30
30 One report identifies three user groups impacted by cycling riders on existing facilities and possibly determine induced facilities: road users, non-road users (e.g., occupants of adja- riders on a new facility. However, this is just one benefit and cent properties), and planning/financing agencies (66). The it remains unclear to which categories it applies. One article first group of road users includes all users, cyclists, motorists, focuses exclusively on methods, reviewing the Travel Cost pedestrians, horse riders, and public transport. Alternatively, Method (TCM) to determine economic value and suggesting some studies divide the benefits of non-motorized travel into better alternatives for measurement (73). When it comes to internal versus external benefits. The former include the finan- estimating, many studies "guesstimate" to solve the problem. cial savings, health benefits, increased mobility, and overall Each of the methods and units are different, yielding varied enjoyment for cyclists; the latter include the benefits to others, output that precludes the desired aim of a common unit. such as reduced (a) congestion, (b) road and parking facility Previous work provides the most precise guidance by sug- expenses, (c) motor vehicle crashes, (d) air and noise pollu- gesting a unit by which each characteristic could be mea- tion, and (e) natural resource consumption. sured. These range from simple counts (e.g., reduction of One of the most contentious issues is adequately account- casualties) to decibels to monetary amounts (e.g., vehicle ing for benefits accrued by cyclists of different ages. Adults operating costs) to descriptive measures (e.g., overall con- and children value different types of facilities. Children enjoy venience). More often, general measuring techniques are trails for recreational purposes. Programs such as "Safe Routes offered. For example, it is suggested that hedonic pricing could to Schools" are also becoming important. However, it is dif- be used to measure livability or amenity visual/aesthetic val- ficult to obtain reliable data for an adult cycling population ues; economic input/output models could describe economic much less for a population of children. For this reason, almost development; time could be used to measure transportation all analysis of bicycle facilities, including the research reported savings; and surveys of different kinds (e.g., contingent val- herein, is based on the preferences of an adult population. uation) could be used to capture a host of values or benefits. Which Benefits Apply to the Facility? PROPOSED BENEFITS AND METHODS The range of benefits of cycling facilities include, but are not Past research offers varying perspectives on the bicycle limited to, reduced pollution, congestion, capital investments facility information different audiences require. The central (at least compared with roads and auto use), and increased challenge for urban planners, policy officials, and researchers livability, health, well-being, and quality of life. But anecdo- focuses on the benefits of bicycle facilities that pointedly sat- tally describing such benefits has limited value. Politicians isfy certain criteria. After reviewing existing literature, can- and lobbyists seek reliable and quantifiable estimates. Spe- vassing available data and methods, and consulting a variety cific benefits range from the direct and easy-to-understand to of policy officials, the team has determined that bicycling the difficult to reliably calculate. Counting the number of benefits need to satisfy five criteria: (1) be measured on a cyclists using a new bicycle trail is relatively straightforward municipal or regional scale, (2) be focused on transportation after the fact. The challenge is translating such levels of rid- and urban planning, (3) be estimable via available existing ership into an economic value. data or other survey means, (4) be converted to measures One study suggests seven benefits to consider when esti- comparable with one another, and (5) be measured benefits mating the economic value of walking: livability, accessibil- for both users and non-users (i.e., the community at large). ity and transportation costs, health, external costs, efficient It is also important to describe the range of benefits, to whom land use, economic development, and equity (71). Focusing they apply, and to suggest compelling methods in which they just on greenways, Lindsey (72) articulates six valued bene- could be measured. The list of benefits is guided by previous fits: recreation, health/fitness, transportation, ecological bio- research and includes direct benefits to the user--in the form diversity and services, amenity visual/aesthetic, and economic of mobility, health, and safety benefits--and indirect benefits development. Which benefits are most important? Is it those to society--in the form of decreased externalities, increased that are accrued, those in which the sponsoring agency is pri- livability, and fiscal savings. marily interested, or those for which there is available data? Other benefits certainly exist, but the beneficiaries are not always clear. The aim is not to dismiss their significance What Units and Methods are Used? but merely suggest that practical considerations related to data, methodologies, and measurement often preclude more The last issue involves the units and methods used to calcu- detailed analysis. The benefits mentioned usually have dif- late different benefits. An ideal analysis considers benefits in ferent beneficiaries. These range from society-at-large to a framework using a common unit. But how does an increase individual users (potential and current) to agencies; there is in riders compare with a reduced need for parking spaces? crossover between beneficiaries for each benefit. Consider, How does increased livability compare with decreased health for example, that the most common argument in favor of concerns? With adequate data, it would be possible to count cycling suggests that an increase in facilities will result in

OCR for page 30
31 increased levels of cycling. This assumed increase in cycling case, it would help to understand the type of cycling facili- will be derived from (1) existing cyclists whose current lev- ties residents value. Conjoint analysis enables researchers to els of riding will be heightened (because of more attractive calculate the value that people place on the attributes or fea- facilities) and (2) potential cyclists whose probability for rid- tures of products and services; the aim is to assign specific ing will be increased. Thus, there are potential benefits for values to the options that buyers look for when making a two different populations (current and potential cyclists). But decision to use a good. It is a technique used to explore trade- if any of these heightened levels of cycling result in decreased offs to determine the combinations of attributes that satisfy auto use, then a third beneficiary results--society-at-large-- the consumer. in terms of reduced congestion and resource consumption. In these cases, an individual is provided a choice of alter- Descriptions follow of what each benefit refers to, the pri- natives; for example, the various travel routes by which a mary user group to whom it applies, and a thumbnail explana- particular travel destination can be reached. The choice of a tion of strategies that could be used to measure such benefit. particular mode is assumed to depend on the relative attrac- The proposed method is not to imply there is a single strategy tiveness of the various travel options that the individual for estimating this benefit but merely to provide the reader faces. These methods use experimental procedures to obtain and researcher with an example of how it could be measured. preferences based on the individual's evaluation of the vari- Figure 1 (in the Summary) shows a simplified depiction of ous options given. Typically, these experiments provide potential beneficiaries along with an indication of the primary hypothetical travel scenarios to obtain an individual's pref- benefit. erences (77). Stated preference surveys need to be stratified by audi- ence: current users versus potential users. Current cyclists Mobility could be asked to respond to questions about factors that would provide for a more attractive cycling environment The most directly cited benefits are often from bicycle through different types of environments or facilities. It is facility users. These come in the form of greater satisfaction necessary to have forced trade-offs so that a better environ- of existing cycling (e.g., cyclists would be able to reach their ment might be coupled with higher costs for bicycle storage destination faster, safer, via a more attractive means). How- or a higher travel time. This will allow one to value each ever, existing information by itself (e.g., ridership counts) component of the user's preference. These preferences can cannot reliably shed light on this issue. For this reason, the then be translated to economic benefits using consumers' different transportation benefits for the user are best uncovered surplus measures (78) to determine, for example, the value of through stated preference surveys or experiments. Because an off-road bicycle facility for users of that facility. stated preference methods provide individuals with hypo- For potential users, it is important to create scenarios thetical situations, it becomes feasible to analyze situations based on constructed markets, asking people to attach a value that are qualitatively different from the actual ones seen in to goods or services. This technique quantifies the benefits practice (74). that non-bicycling residents would accrue from a more desir- Because individuals respond to several different hypothet- able bicycling infrastructure. For example, questions could ical choice situations offered to them, the efficiency of data be what mode they would choose for work and non-work collection is improved; enough data is hence available to cal- trips based on the quality of the transportation environment, culate functions describing their preferences or utility. The including travel by auto, walking, transit, and bicycle. It would disadvantage in stated preference methods is that people may query residents about the degree to which they perceive dif- not always do what they say. Individuals' stated preferences ferent bicycling services or how facilities will improve the might not be the preferences they actually show (75). The conditions of their commute, recreational activities, and so differences arise because of the systematic bias in survey forth. By measuring how demand might change, one can ascer- responses or because of the difficulty in carrying out the tain the preferences for current non-users, some of whom posed task. would become users if a certain infrastructure package were Two techniques used in stated preference analyses are constructed. contingent valuation and conjoint analysis. Contingent valu- The team's approach to determining user mobility benefits ation is based on the premise that the best way to find out the is described in detail in Appendix D. It quantitatively evalu- value an individual places on something is by asking. Like ates individual preferences for five different cycling envi- other non-market goods, the concept has been applied to ronments. The respondent is asked to trade off a higher travel wilderness, open space, or even more specifically to green- time as a cost incurred to choosing a better facility while ways (76). The second stated preference technique, conjoint allowing the respondent the option of selecting a less attrac- analysis, uses experiments to obtain the preferences of tive facility at a lower travel time. The trade-off of travel time the customer. This market research technique can provide to amenities of a particular facility determines the value important information about new product development, fore- attached to different attributes such as bike lanes, off-road casting market segmentation, and pricing decisions. In this trails, or side street parking. The facilities considered in this

OCR for page 30
32 application are off-road facilities, in-traffic facilities with tors (i.e., increasing physical activity) does result in reduced bike lane and no side street parking, in-traffic facilities with cost. The final step is for researchers to demonstrate that a bike lane and side street parking, in-traffic facilities with no health habits can be changed and that the resultant lower risk bike lane and no side street parking and in-traffic facilities can be maintained over time. As can be seen, the challenges with no bike lane but with street side parking. The results associated with documenting a health financial payback from indicate that respondents are willing to travel up to 20 min a bicycle facility are significant. Looking at the problem opti- more to switch from an unmarked on-road facility with side mistically and from the perspective of needing analytical jus- parking to an off-road bicycle trail, with smaller changes for tification, such exercise is not completely out of the realm of less dramatic improvements. possibility. For this reason, these later steps (three through five) constitute the focus of the following review. The benefits of physical activity in enhancing overall health Health are well established. The task of attaching monetary value to levels of physical activity is a more challenging endeavor. Scientific literature from researchers and practitioners One attempt is offered by Wang et al. (93) who derive cost- from a variety of disciplines show relationships between com- effectiveness measures of bicycle/pedestrian trails by dividing munity design, transportation facilities, and levels of physical the costs of trail development and maintenance by selected activity (79, 80). "Sprawling" land use practices and result- physical activity-related outcomes of the trails (e.g., number of ing auto-dependent travel are themes that now have moved trail users). The average annual cost for persons becoming to the front of the American consciousness; the link to pub- more physically active was found to be $98; the cost was lic health and obesity remains an important component of $142 for persons who are active for general health, and the this discussion (8183). One goal of this research is to learn $884 for persons who are active for weight loss. the extent to which rates of physical inactivity can be linked to Estimating the effect of physical inactivity on direct med- features of the built environment (see Krizek et al. [84]). At a ical costs is a strategy more often employed, though con- regional or neighborhood level, most inquiries focus on land siderably less straightforward. Part of the reason for ambi- use patterns characterized by relatively scattered, single use guity in this research is that the amount of physical activity and low-density development. At a street or facility level, such required to realize certain health benefits is relatively unknown research focuses on access to sidewalks, trails, other non- (i.e., what is the elasticity?) (88, 94, 95). In the field of pub- motorized facilities, and destinations. While recent research lic health, this matter is often approached from the perspec- has linked neighborhood design to travel behavior (85, 86), tive of dose-response relationships. The aim is to learn what little of it has exclusively focused on relationships between change in amount, intensity, or duration of exposure (in this specific facilities, bicycling and walking travel, and levels of case, cycling) is associated with a change in risk of a speci- physical activity. fied outcome (in this case, cost of health care). To establish a health-care, cost-based reason for bicycle Existing literature examining relationships between levels facilities, several types of specific empirical evidence must of physical activity and health costs varies considerably in be gathered and communicated to interested parties. Using methodology and scope. The majority of existing studies pur- reasoning from Goetzel et al. (87), researchers must first sue a dichotomized approach, separating respondents into two demonstrate relationships between a given feature of the classes: those that satisfy an accepted dose of 30 min per day built environment (e.g., a bicycle facility) and levels of for 5 days and those who do not. In this first group of studies, cycling. This activity is similar to methodologies previously there are at least five statewide reports whose methodology described to measure the demand induced from various facil- and assumptions are relatively general in nature. In most cases, ities. International research on this question is likely to have estimates are derived from an aggregation of medical expen- reliable results that can enhance this line of inquiry in rela- ditures that can in some form be traced back to physical inac- tively short order time (i.e., a couple of years). Second, any tivity. For example, a study commissioned by the Michigan amount of induced cycling that could be "teased" out from a Fitness Foundation (96) concentrated on the economic costs facility would then need to be translated into an average per- to the residents of Michigan. The authors used estimates centage of one's weekly physical activity. For example, the (acknowledged to be conservative) to derive direct costs daily recommended level of physical activity is defined as (e.g., medical care, workers' compensation, lost productivity) 30 min of moderate physical activity on 5 or more days per and indirect costs (e.g., inefficiencies associated with replace- week (88, 89). Cycling 5 mi in 30 min or 4 mi in 15 min ment workers). The final amount totaled $8.9 billion in 2003 would meet these current public health guidelines for physi- ($1,175 per resident). A 2002 report from the Minnesota cal activity (9092).iv Department of Health (97) estimates that in 2000, $495 mil- Third, researchers must then demonstrate that lack of phys- lion was spent treating diseases and conditions that would ical activity--because it is indicative of certain risk factors-- be avoided if all Minnesotans were physically active. This imposes a financial burden to the individual or to society. A amount converts to more than $100 per resident. Additional fourth step would be to show that improving certain risk fac- reports claim that too little physical inactivity was responsi-

OCR for page 30
33 ble for an estimated $84.5 million ($19 per capita) in hospi- Table 25 in Appendix E). Second, some studies are disag- tal charges in Washington State (98), $104 million ($78 per gregate in nature and estimate costs by inpatient, outpatient, capita) in South Carolina (99), and $477 million in hospital and pharmacy claims; others compare average health care charges in Georgia ($79 per capita) (100). expenditures of physically active versus inactive individuals. These reports from various state agencies are complemented Third, some use a dichotomized approach to determine who with more academically oriented research. For example, constitutes a physically active individual while others employ Colditz (101) reviewed literature on the economic costs of inac- a modifiable health risks approach and do so in a relatively tivity and concluded that the direct costs for those individuals continuous scale. The studies are difficult to compare, how- reporting lack of physical activity was estimated to average ever, because some include different conditions, outpatient approximately $128 per person. A separate analysis by Pratt and pharmacy costs, and actual paid amounts rather than et al. (102) analyzed a stratified sample of 35,000 Americans charges. Nonetheless, existing literature provides adequate, from the 1987 national Medical Expenditures Survey. Exam- though developing, methodologies for estimating the public ining the direct medical costs of men and women who reported health impact of bicycle facilities in economic terms. physical activity versus those who did not reveals that the These approaches have recently been made more accessi- mean net annual benefit of physical activity was $330 per per- ble to planners, decisionmakers, and the public through the son in 1987 dollars. An alternative method used a cost-of- Robert Wood Johnson's Active Leadership Program. The illness approach to attribute a proportion of medical and physical inactivity calculator available on the website (103) pharmacy costs for specific diseases to physical inactivity provides an easy-to-use tool to estimate the financial cost of in 2001 (97). The authors first identified medical conditions physically inactive people to a particular community, city, associated with physical inactivity and then collected claims state, or business. It also supplies companion resources and data related to those conditions from approximately 1.6 mil- information needed to re-allocate resources and plan for health- lion patients 16 years old and older from a large, Midwest ier workplaces and communities that are more supportive of health plan. While the resulting conditions from lack of phys- physical activity. ical inactivity include depression, colon cancer, heart dis- ease, osteoporosis, and stroke, the results from this study conclude that the costs of claims to the health plan attribut- Safety able to physical inactivity translates to $57 per member. One challenge of these analyses is the decision whether to include Increased cyclist safety is an often assumed, poorly under- diseases causally related to obesity. stood, and highly controversial benefit of bicycle facilities. A different approach than the dichotomized strategy esti- The task of establishing a safety derived, cost-based justifi- mates the impact of different modifiable health risk behaviors cation for bicycle facilities is similar to the process described and measures their impact on health care expenditures. After in the previous section for estimating public health benefits, gathering information from more than 61,500 employees of albeit with different data. Researchers must first demonstrate 6 employers gathered over a 5-year study period, Goetzel et al. relationships between a given cycling facility and safety out- (87) focused on a cohort of slightly more than 46,000 employ- comes. Then they need to demonstrate that the measured out- ees. The analysis found that a "risk-free" individual incurred come of conditions with decreased safety imposes a financial approximately $1,166 in average annual medical expenditures burden to the individual or to society. while those with poor health habits had average annual med- In general, the literature about the safety dimensions of ical expenditures of more than $3,800. Thus they estimated bicycling manifests itself in three primary aspects: (1) helmet the per-capita annual impact of poor exercise habits to be use, (2) safety programs, and (3) number of crashes or per- approximately $172. Pronk et al. (89) also identify the rela- ceived level of safety that can be ascribed to facility design. tionship between modifiable health risks and short-term The last category is most germane to the construction of health care charges. This research surveyed a random sam- facilities and is the center of the following discussion. One ple of 5,689 adults 40 years old or older enrolled in a Min- issue is how to combine data about safety (e.g., crashes or nesota health plan. Multivariate analysis on the modifiable perceived comfort) with different attributes of cycling facil- health risks (diabetes, heart disease, body mass index, phys- ities. The team's aims to understand the degree to which dif- ical activity, and smoking status) concluded that an addi- ferent cycling facilities lead to an incremental safety benefit, tional day of physical activity (above zero) would yield a measured in terms of decreased crashes or medical costs. 4.7% reduction in charges (or a $27.99 reduction). The over- Existing literature measures safety in one of three ways: arching result of the study is that obesity costs approximately (1) number of fatalities, (2) number of crashes, and (3) per- $135 per member, per year and those with low fitness (inac- ceived levels of comfort for the cyclist. Key explanatory vari- tivity) cost approximately $176 per member per year. ables behind these outcome measures are myriad and complex Several matters should be noted when determining values to identify. For example, the overwhelming majority of bicy- for health benefits. First, annual per capita cost savings vary cle crashes resulting in fatalities are caused by collisions between $19 and $1,175 with a median value of $128 (see with motor vehicles (104). Less severe crashes tend to occur

OCR for page 30
34 at intersections or at locations where motor vehicles and absence of a bicycle lane, but the width of the outside lane bicycles come in contact with each other (105); it is further variable did include the bicycle lane were it present. The suggested that crashes are caused by differing expectations research by Krizek and Roland (119) analyzed the severity of between auto drivers and bicyclists (106). However, there instances where existing bicycle lanes terminated and their is evidence to suggest that some bicycle crashes do not corresponding physical characteristics. The findings suggest involve any other party (107, 108); this is especially true for that bicycle lane discontinuations ending on the left side of children (109). the street, increased distance at intersection crossings, park- The prevailing argument is that enhanced facilities--bike ing after a discontinuation, and width of the curb lane are sta- lanes, bikeways, and special intersection modifications-- tistically significant elements that contribute to higher levels improve cyclist safety (83). This claim, however, is con- of discomfort for the cyclist. troversial and a source of debate between Forester (57) and The degree to which perception of safety translates into Pucher (58). One of the issues concerns differences between actual increased safety, however, is still debated. It is diffi- what cyclists state they prefer (i.e., their perception) and what cult to translate perceived measures of safety into quantifiable studies with collision data actually reveal. or economic estimates. It is widely acknowledged that increased perception of There is evidence to support the notion that collision-type safety is important to encourage cycling as a means of trans- crashes are lower on off-road paths (120). Using before and portation and recreation (51, 110). Subsequently, providing after analysis, Garder's research (121) found raised bicycle separated bicycle facilities along roadways is mentioned as a crossings to be more appealing and safer for cyclists than at- key to increased perception of safety according to the litera- grade crossings. However, there exists an equal, if not greater ture related to bicycle-related stress factors (111); bicycle body of research suggesting no relationship or a relationship interaction hazard scores (112), relative danger index (113), in the opposite direction. Research examining conflicts at compatibility indexes (114). approaching intersections on bike lane and wide curb lane The goal of these works is to determine and predict condi- segments determined that both facilities improve riding con- tions for safe bicycling based on different cyclists' percep- ditions for bicyclists, but that the two facilities themselves tions of safety. The culmination of these works can best be are not different in safety (122). Smith and Walsh analyzed described under the banner of level of service (LOS) models, before and after crash data for two bike lanes in Madison, originally developed in 1987 in Davis, California (115, 116). Wisconsin, finding no statistically significant difference (123). The participants of these studies were of diverse demographic Also, Hunter's analysis of a bike box in Eugene, Oregon, and skill backgrounds and cycled 30 roadway segments. showed that the rate of conflicts between bicycles and motor Including the variables of traffic volume per lane, posted vehicles changed little in the before and after periods (124). speed limit weighted with the percentage of heavy vehicles, No conflicts took place while the box was used as intended. adjoining land use, width of outside through lane, and pave- Hunter also evaluated colored (blue) pavement and accom- ment conditions, the researchers were able to explain almost panying signing used in weaving areas at or near intersec- 75% of the variation of perceived safe conditions. The model tions in Portland, Oregon (125). The colored rectangular area consists of four basic factors--pavement conditions, traffic within the bike land came to be known as "blue bike lanes," speed, lane width, and traffic volume per lane which aim to even though only the weaving area was colored. Although serve as a tool for predicting perceived safety and comfort conflicts were rare, the rate of conflict per 100 entering bicy- along roadways between automobiles and bicycles. clists decreased from 0.95 in the before period to 0.59 in the The bulk of the existing literature on bicycle LOS and per- after period. In addition, significantly more motorists yielded ceived safety focuses primarily on through travel on mid-block to bicyclists in the after period. roadway segments. Previous research has rarely considered There appears to be good reason for the existing debate bicycle lanes separately from other shared use conditions over the safety benefits of bicycle facilities. While there is (wide curb lanes or paved shoulders) and rarely considered considerable literature suggesting cyclists perceive greater the role of intersections. While stretches of roadways are safety with facilities--and advocates certainly argue for important, often the most significant and complex design and such--the bottom line is that there is little conclusive evi- safety challenges occur at street intersections (117). Two dence to suggest this. One theory suggests that if a particular recent research papers focused on this matter (118, 119). setting is deemed unsafe, a cyclist will likely be vigilant and Landis' recent work (118) derived a model to evaluate the avoid an incident. As a result, the number of incidents may perceived hazard of bicyclists riding through intersections. be no greater with an unsafe condition than a safe condition. Again, with a highly varied demographic and cyclist ability However, such argument does not support the conclusion sample, this study produced a model with a high degree of that both conditions are equally safe. In the less safe condi- explanatory power (R2 = 0.83) for bicycle intersection LOS. tion, the cyclist will either avoid it or endure a cost of stress Significant variables included motor vehicle volume, width to use it. of the outside lane, and the crossing distance of the intersec- Yet an alternative theory, not directly applicable to spe- tion. In this study, there was no control for the presence or cific facilities, is derived from the concept of safety in num-

OCR for page 30
35 bers where the likelihood of bicycle-automobile crashes inter- estimate an upper bound of all mileage that is substituted and act in a nonlinear manner (the exponent for growth in injuries the overall social costs being saved. However, this does not is roughly 0.4) for an entire metropolitan area. Applying this account for the possibility that bicycle trips may be substi- concept, one would need to calculate the total bicyclists at the tuting for modes other than driving. Furthermore, it says lit- metropolitan level (X). Then one could compute 1.X exp 0.4. tle about how many additional trips from potential cyclists This number provides the additional bicycle safety cost of could be induced. Such information would be most reliably adding bicyclists. The cost of car crashes could even be obtained by estimating a mode-choice model for different reduced by the proportional reduction in cars on the road. types of cycling trips and calculating the likelihood of sub- Then, each additional bicyclist increases the total number stitution rates in that manner. The latter strategy is subject to and thus the total cost of bicycle crashes (though not the per elaborate modeling schemes and survey data. unit cost). In low-volume portions of the nonlinear relation, It is important to recognize, however, that any reduced the decrease in fatality rate outpaces the increase in volume congestion benefit to society needs to be tempered by an so that, even with more cyclists, the number (not just the induced demand phenomenon which may obviate congestion rate) of fatalities decreases. Values for the benefits and costs or pollution reductions due to diversion (127). This implies of such crashes could be obtained from third party sources that reduced traffic congestion that may result from the con- (e.g., http://www.oim.dot.state.mn.us/EASS/) which typically struction of an additional bike lane may largely (though not summarize the cost per injury of car crashes per type. entirely) be consumed by other drivers making additional trips, drivers lengthening trips, and additional development. This suggests that any reduction in congestion (and subse- Reduced Auto Use quently pollution and energy benefits) will be small at best. Nevertheless, the additional opportunities for drivers to pur- The most common assumption is that cycling trips substi- sue activities that previously had been too expensive prior to tute for auto trips, yielding transportation benefits to society- the capacity expansion (of roads or bike lanes) engender some at-large such as decreased congestion, improved air quality, benefits for new drivers. and decreased use of non-renewable energy sources. While the substitution element may hold true for some cyclists it is extremely difficult to reliably determine the trips that would Livability otherwise be made by car. The nature and magnitude of any substitution is important Another benefit refers to social attributes accrued by indi- to determine and could be estimated via a variety of means. viduals who receive benefits of such facilities, either directly In some instances, a bike trip may replace a car commute; in or indirectly. One of the reasons people pay a premium to many cases, however, bicycle trips are likely made in addi- live in desirable areas is that they are paying for the option to tion to trips that would otherwise occur (126) or for a differ- use specific facilities, whether or not they actually do. For ent reason (e.g., recreation). Assuming a fixed demand of instance people may pay a premium to live near a bike path overall travel, a best-case scenario for bicycle substitution despite not cycling themselves because they might want to in stems from an assumption well known in the field of travel the future. In this respect, such proximity would be valued behavior modeling referred to as Independence from Irrelevant by current and potential users. These benefits are revealed Alternatives (IIA). That is, bicycles draw from other modes in through preferences that represent an elusive phenomenon to proportion to their current mode shares. For instance, bicycles which an economic value can be attached. A compelling strat- would draw 85% from current drive alone trips, 5% from auto egy to measure these non-market goods analyzes the choices passenger trips, 5% from transit trips, and 5% from walk trips. people reveal in their purchase of home locations in efforts This of course is unlikely to be strictly true, so an important to understand how they implicitly or explicitly evaluate the part of the benefit analysis would be to determine which of desirability of a certain good. A revealed preference approach these groups are more likely to switch to bicycling and fur- would measure individuals' actual behavior, and this can be thermore, which socio-economic characteristics could be tar- done through hedonic modeling to learn if and how much geted to result in higher rates of cycling. residents value accessibility to bicycle facilities. Assuming bicycling facilities can help bicyclists travel Discerning the relative value of non-market goods using faster, more safely, in a better environment or for shorter dis- hedonic modeling techniques is a method that has been tances, its utility compared with other modes will increase. employed for years since its first applications by Lancaster There may be an estimable effect in terms of substitution, and (128) and Rosen (129). An extensive review of this literature there are different approaches for measuring this phenome- (130) contains nearly 200 applications that have examined non. At a crude level, one could estimate the number of bicy- home purchases to estimate values of several home attributes cle miles of travel and auto miles of travel. Assuming a fixed including structural features (e.g., lot size, a home's finished rate of substitution (i.e., 60% of all cycling trips are utilitar- square feet, and number of bedrooms), internal and external ian in nature and are substituting for a car trip), one could features (e.g., fireplaces, air conditioning, garage spaces, and