In addition to the above examination of why some respondents had not used trails at all, the study also explored whether people were able to use trails as frequently as they liked. All interviewees were asked if they had wanted to use trails more than they had during the past 12 months. Regardless of their actual level of use, 22% reported that their trail use was constrained in some way (i.e., they wanted to use trails more than they had during the past 12 months, but were unable to for some reason). By comparing responses to this question with the one regarding whether respondents had used trails at all during the past 12 months, it was possible to divide the sample into the following three distinct groups: trail users, nonusers (nonusers who did not want to use trails), and "reluctant nonusers" (nonusers who wanted to use trails but were unable to for some reason). A total of 54% of respondents indicated that they were either users or wanted to be users of trails and greenways in NC during the past 12 months (Figure 11). Note that these are the percents used in computing the weighted means and weighted percents reported in many of the tables.

Figure 11. Proportion of North Carolinians Who Used or Wanted to Use

Trails in Last 12 Months.

Those who reported using trails less than they had wanted to were provided with a list of 12 potential reasons and asked to indicate which ones were ever problems that kept them from using trails in NC during the past 12 months. Of these, lack of time was by far the most common constraint. Three-quarters indicated that lack of time was a reason for not using trails as often as they would have liked (Table 6). Not enough information about trails and no trails close to home were listed as constraints by at least one third of the respondents. The remainder of the potential constraints were identified as problems by 19% or fewer of the constrained respondents.

Table 6.

Reasons Respondents did Not Use Trails as Often as Desired During the Past 12 Months (n = 886)

Respondents' awareness of the existing types of trails in the state varied widely. The vast majority were aware of surfaced and unsurfaced trails and over three-quarters knew of national scenic, historic or recreation trails. Only about half were aware of trails for mountain bicycles, however. The least recognized type of trail was rail-trails (trails built on abandoned railroad corridors) (Table 7).

Table 7.

North Carolinians obtained their trail information from many sources, but friends were the source for the majority (Table 8). Over 60% indicated that their friends were one of their sources. Nearly 40% used signs at trail access points, magazine articles, or news articles. The internet was the source listed by the fewest respondents. When asked which sources they preferred to use to obtain information about trails in NC the most common responses were trail brochures, the internet/WWW, and park and recreation offices (Table 9). Overall, respondents reported that they would have to travel a considerable distance to access the nearest trails. Average one-way distances to the nearest surfaced and unsurfaced trails were 21 and 23 miles, respectively (Table 10).

Table 8.

Table 9.

Table 10.

A series of questions were asked to assess trail conditions across the state and respondents' experiences with them. In order to do this they were asked to consider only their most recent trail trip in North Carolina rather than their trail experiences in general. Because the sample was drawn randomly, the results obtained are considered reasonably representative of trails across North Carolina. Those who had never visited a trail in NC were asked to skip these questions.

For three-quarters of the respondents, the last trail visited had been unsurfaced. The remaining quarter had most recently used a surfaced trail (Table 11). The majority had traveled 20 miles or less to access the trail (Table 12). Over 12%, however, had traveled over 100 miles one-way. Similarly, it took most people 30 minutes or less to get to the trail while a sizable minority (15%) traveled two hours or more (Table 13). The vast majority traveled to the trail by car (Table 14). Most trail users spent little money during their trips including travel expenses and anything else they may have bought for or during the outing. Sixty percent spent $10 or less and nearly a quarter spent nothing at all. However, 14% spent more than $50 (Table 15). Since the average of $40 per person per trip includes four respondents who spent large amounts as part of extended trips (three at $500 and one at $3,500), the median value of $10 per person per trip may be more meaningful. Most people had visited that particular trail five or fewer times during the past 12 months. There were respondents, however, who were regular users of that trail. Nearly 8% reported having used the trail more than 10 times during that time period (Table 16).

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Several questions explored potential problems users may have experienced during their most recent trail trip. The levels of problems reported were universally low on average. Conflicts with other trail users not in their parties was not a problem for the vast majority of users. Over half reported that any other users they may have encountered had no effect on their enjoyment. In fact, a quarter said the others they encountered actually increased their enjoyment. Only 7% indicated that their encounters with other trail users had reduced their enjoyment in some way (i.e., involved some form of trail conflict) (Table 17). Of those who reported that others had affected their enjoyment, the most common way was through conversations or simply seeing others enjoy themselves. The most common way others had a negative effect was simply due to there being too many other people present. This problem with crowding was reported by 10% of those whose enjoyment was affected (Table 18).

Table 17.

Table 18.

The levels of other potential problems examined were very low as well. Respondents rated the levels of 16 potential problems on 5-point scales where 1 indicated "not a problem at all" and 5 "a serious problem." In every case the majority rated them as either a 1 or 2 and none were rated higher than 2.5 on average, indicating that they were no more than a moderate problem overall during their most recent trips (Table 19). The greatest concerns were expressed for items related to lack of trail amenities, particularly restrooms, drinking water, and commercial services.

Table 19.

A series of questions assessed the importance North Carolinians attached to trails in the state. Overall, trails were considered quite important. When asked directly how important they felt it was to have trails available in North Carolina, over 43% reported that they were extremely important. The average response was over 4 on a 5-point scale where 1 was "not at all important" and 5 was "extremely important." Only 5% felt trails were unimportant (responses of 1 or 2) (Table 20). The importance of trails in providing 13 potential benefits was measured on a 5-point scale where 1 indicated that "trails do not provide this benefit at all" and 5 indicated that "trails are extremely important in providing this benefit." The most important benefits were aesthetic beauty, health and fitness, public recreation opportunities, preserving undeveloped open space, and public education about nature and the environment. Each of these benefits had an average rating of over 4 on the 5-point scale and was considered extremely important by more than 45% of respondents. The least important benefit of trails and the only one with a mean rating below the scale midpoint was reducing crime (Table 21).

A final indication of the importance of trails to North Carolinians was respondents' expressed willingness to pay for trail use. When faced with the hypothetical question, "Which of the following amounts would you be willing to pay each year to use trails in North Carolina if the money was used to develop and maintain trails in North Carolina?" and offered $5 increments from 0 to $25, over three-quarters said they would be willing to pay under those circumstances. Nearly a third (31%) would pay $20 to $25 annually. The average willingness to pay was approximately $12 (Table 22).

Table 20.

Table 21.

Note. The mean values and weighted mean values are identical.

Table 22.

Note. Mean willingness to pay = $11.65.

The final group of questions attempted to provide direct guidance for planning future trail developments in North Carolina. The aspects of the current trail system which respondents were least satisfied with overall were the variety and number of trails available in the state (Table 23). The majority felt there were currently not enough trails or enough variety of trails in NC (responses of 1 or 2). Most responded that the information on trails in the state and the number of trails within 15 minutes of their homes was about right. The majority, however, would still prefer to have both an unsurfaced and a surfaced trail even closer to their homes (Tables 24 & 25). The majority also preferred trails over other setting for certain activities. When asked, "If you currently participate in a trail activity in a non-trail setting (e.g., walk for pleasure on sidewalks) would you prefer to do it on a trail if one were available near you?" 80% said they would (Table 26). There was little consensus, though, on how close was close enough for a trail. Respondents were somewhat supportive on average of living immediately adjacent to a trail or greenway (mean of 3.5 on a 5-point scale) (Table 27), but not everyone shared this feeling. Nearly 20% were not supportive of living this close to a trail (responses of 1 or 2). And respondents were equally divided about whether North Carolina should make it a goal to provide a trail within 15 minutes of every resident (Table 28).

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.

Most respondents did expect to use trails in the near future. Seventy six percent reported that they were planning to use a trail in North Carolina during the next 5 years and another 19% did not know whether they would or not. The mix of activities they expected to engage in was much like present use patterns (Table 29). The activity the largest percent (89.5%) expected to engage in was walking for pleasure. A majority (63%) also planned to hike. Participation in snow activities was anticipated by the fewest respondents. Respondents gave their preferences for potential future trail development by answering the question, "Please indicate how much of a priority you feel each of the following types of trails should be for future development in North Carolina?" Responses were rated on a 5-point scale where 1 indicated "very low priority" and 5 "very high priority." Mean responses for 8 of the 9 trail types examined were slightly above the midpoint of 3 indicating a moderate priority level for each (Table 30). Designated water trail development was seen as the highest priority on average, although the means were very close for all types except off-highway vehicle trails. Future development of off-highway vehicle trails was seen as a lower priority than the others overall.

Table 29.

Table 30.

A wide variety of advice was offered when respondents were asked what one thing they would like to see done to improve the system of trails available in North Carolina. Providing more information on trails and their locations was the most common suggestion to improve the trail system and was offered by 14% of respondents. Additionally, providing more trails in general and providing more information in general were each considered the best way to improve the trail system by over 10% of the respondents (Table 31).

Table 31.

The dollar measure of an economic benefit to a trail user in North Carolina is derived from that user being able to take trips to a trail at the same price each time. The price usually consists of two costs for each user. The round-trip travel costs associated with the operation of a vehicle, if any, and the travel time or opportunity cost of time associated with a user having to drive or walk to a site and return. We evaluate the statistical relation between the number of trips to a trail during the past year and the price per trip. This is referred to as a travel cost model. The resulting analysis leads to the computation of an economic benefit per user. Users obviously do not receive dollar payments or direct adjustments in their annual incomes that are equivalent to the benefit amounts. Rather, the benefits attempt to quantify those intangible values that users receive from having visited trails. In other words, what the trail was worth to them over and above what it might have cost them to visit.

Benefits are different from economic impacts. An economic impact is a change in economic activity generated by users spending money while visiting a trail in a particular area. The impact is a measure of the effects of users' expenditures in a particular geographical area and the operating expenditures made by trail officials. The economic effects are secondary because the effects reflect changes in sales revenues, jobs, net incomes, and tax revenues in the local economy. The economic impact reflects the gain in an area's economic base that supplies the primary users with a trail and other goods and services. Consequently, an economic impact does not measure the economic benefits of a trail to primary users.

Whether to estimate an economic impact or the economic benefits of a trail is often based on the national or regional prominence of that trail. Most important is whether local officials believe that the trail really contributes significantly to economic activity of a local economy. If a trail does not attract visitors from a region outside the local economic base, then users' expenditures will have little to no impact in creating new jobs and incomes in the local economy. Similarly, a trail may primarily attract users who travel lengthy distances but make the majority of their expenditures outside the area. Such trails may have a high economic benefit to the primary users, but no economic impact. Essentially, those users traveling from outside the local economic base may not spend money for local services and goods (e.g., overnight accommodations, meals, and the purchase of hiking equipment). Such users' expenditures produce no secondary effects on the local economy.

Readers should realize that many of the restrictions imposed by us in estimating trail use are due to the availability of survey data regarding user tastes, preferences, and travel behaviors. We studied a geographically diverse sample of trail users. Consequently, the trails differ in site qualities across our sample of users. A further data restriction involved the fact that we may have only one independent observation of a user visiting a specific trail during the season. The following difficulties require a simplification of the travel cost method of analysis to estimate trail use and the economic benefits to people in North Carolina: Disentangling the effects of the quality differences of the diverse trails in North Carolina, differences in each user's proximity to all the trails in North Carolina, and an under sampling of all the trails in North Carolina.

Our strategy involved the specification of a "pooled sites" model. All annual trips to the last site visited by our sample of users are pooled into a single trail equation, which is called a hypothetical travel cost model. The travel cost model incorporates a measure of each user's perceived quality of trail conditions at the last trail visited to account for the varying conditions among the trails. In addition, users were asked the annual number of recreation trips that they would have taken had there been a hypothetical improvement in site conditions. We combine the revealed (actual) trip counts with the stated (hypothetical) trips. We then grouped the two different observations by the same users into a panel data set. This allowed us to calculate individual differences in welfare benefits between revealed and stated behaviors of trail users.

In the hypothetical travel cost framework, we augment the travel cost method by constructing a hypothetical scenario of a proposed improvement to any existing adverse site conditions. We then analyzed users' demand for trail trips by combining data on actual trip taken during the past 12 months and the stated trips that would have been taken, if the site conditions that they perceived had been ideal. Since the revealed or actual trip counts change with the hypothetical proposal, and not the trip prices, the prices are unbiased measures in the hypothetical proposal (Layman, Boyce, & Criddle, 1996).

Since the proposed action in many management cases is to improve quality by eliminating adverse conditions, users are expected to respond by intending to take more trips during a season (Loomis, 1983). Also, some nonusers who are interested in outdoor recreation but took no trips during the past season are expected to begin using trails. However, not captured by the hypothetical proposal is the increased satisfactions of users with the improved conditions at their existing trip counts.

As shown in Figure 12, a change in the shape of the demand curve and the increased area under the curve can be attributed to an increase in stated trips due to the hypothetical improvements in trail conditions. This increase is in part due to the entry of the zero trip-takers during the past season who would have visited trails under more appealing conditions. The safer and better maintained trails may attract the non-users from further distances by enticing them to pass more intermediate substitute trails. The mean revealed (current) trips of 6.4 under existing conditions is less than the stated mean of 10.2 annual trips under ideal quality conditions. Of the 5,630,142 individuals 18 years and older living in North Carolina (Office of State Planning, personal communication, November, 1998), approximately 68% had not visited a trail in the past 12 months, while 32% visited a trail during the past 12 months, according to our telephone survey. Using the mean estimate of 6.4 annual trips per user, individuals took approximately 11.5 million trips in 1998 to the estimated 3,660 miles of trails in North Carolina (D. McBane, personal communication, November, 1998).

Figure 12. Demand Curves for Existing and Hypothetical Ideal Trail Systems.

Incorporating an indicator of trail quality into the estimation of recreation demand enabled us to compute the change in trail benefits between the actual and hypothetical demand for trails. The trail benefit per user is the welfare loss associated with losing trail access. Consequently, benefit values are compensated because they reflect the dollar benefits that a user should receive if, hypothetically, the user were denied access to the trails during a year.

Based on the results of this analysis, it is clear that a statewide allocation of more resources to improve trail conditions would result in far greater public use and benefits. The mean annual trail benefit was $184 per user or $28.75 per individual trip ($184 / 6.4 annual trips) under current trail conditions in North Carolina. This result is consistent with the values of $26.10 per trip for day hiking and $36.95 per trip for walking found in previous research (Bergstrom & Cordell, 1991). The estimated 3,660 miles of trails in North Carolina therefore provides $331.5 million in economic benefits to trail users annually, or $90,575 in economic benefits to users per mile of trail. Under stated trip conditions (i.e., if trails were to be improved to meet users' ideals), the economic benefits of trails increase to $294 per user. If we added the 14% of the respondents who reported zero annual trips, but indicated that they would have taken one or more trips if trail conditions were improved, to the 32% who currently use trails, 46% would have visited trails under stated quality conditions. The total estimated welfare benefits with trail quality at an ideal state would be $761.4 million or $208,038 per mile of trail. (Total benefits were obtained by multiplying 5,630,142 individuals 18 years and older by 46% by $294 per user.)