Urbanities Willingness to Pay for Trees and Forests in Recreation Areas

Methods

To estimate how much people were willing to pay for particular attributes of recreation areas such as trees, forests, and lakes, as well as facilities such as picnic areas and bike trails, we contacted park and forest preserve users from northwest Chicago and its northwestern suburbs randomly from telephone directories. Those who agreed to participate were mailed a questionnaire that included written descriptions of 16 pairs of hypothetical park settings that differed in terms of 22 attributes found to be important in earlier work on park choice (4, 5,11). These attributes included vegetation, terrain, water features, recreation facilities, types of users, maintenance, travel time from home, and entry fees. The choices were designed to permit estimation of the importance of each attribute in choosing a site. Respondents were asked which one park in each pair they would prefer for an outdoor day-trip in the Chicago area. They were then asked whether they would realistically prefer to go to the park chosen or engage in some other outdoor activity instead.

Out of 478 park users who were contacted by phone, 285 agreed to complete the mail survey, and 210 returned usable surveys. Thus the total return rate was 44%, which is quite good for this type of survey.

From the choices made by 210 respondents among the descriptions of diverse hypothetical parks, we constructed models (discrete logit choice models) to predict the likelihood of an individual choosing any option described in terms of the 22 attributes—as well as the option of choosing not to visit a recreation site (8,9). Since one of the attributes that varied over the choices was entry fees, we could estimate willingness to pay to have other attributes at a site. This was determined by what increase in user fees would be needed to just offset the effect of a particular attribute on the probability of visiting a site. For example, to estimate willingness to pay to have a bike trail at a site we first calculated how the probability of choosing a site would increase if a bike trail were added. Then we used the model to estimate how large a fee would be required to reduce the probability of selecting this site back to its original level. This fee represents maximum willingness to pay for the trail, because if users had to pay more than this, the probability of selecting the site would be lower than if there were no trail and no special fee. The entry fees included in the choice options ranged up to a maximum of $3.00.

Results and Discussion

For the sample of respondents as a whole, all the park attributes had a statistically significant influence on choices, with the exception of presence or absence of police patrols. Of perhaps greatest interest to arborists, foresters, landscape architects, and others who plan for and manage vegetation on urban recreation sites is the willingness of users to pay for trees and forests. On the average, and with all other attributes held the same, users were willing to pay up to $1.60 more for a visit to a site that was “mostly wooded, some open grassy areas under trees” than for a site with “mowed grass, very few trees anywhere.” Two other choices represented intermediate amounts of vegetation. Individuals were willing to pay $0.21 more per visit for “mowed grass, scattered trees, no woods” and $0.99 more for “mowed grass, scattered trees, some dense woods” than for the option with fewest trees (Table 1).

The willingness of users to pay for the trees and forests on a site becomes substantial given the relatively high value per visit and the large number of visits to some urban parks and forest preserves. For example, the Ned Brown Forest Preserve located northwest of Chicago and managed by the Forest Preserve District of Cook County receives more than 2,500,000 visits per year. An estimate of $1.00 per visit attributed to this Forest Preserve’s vegetation—a mixture of mowed grass, scattered trees, and forests—would total up to $2,500,000 per year, suggesting that the trees and forests on the area are indeed valuable assets that merit considerable attention.

While the willingness of users to pay for trees and forests is not surprising, it is somewhat surprising that users were willing to pay the most for the most heavily forested areas. Many studies of visual preferences for urban forest environments suggest that a somewhat open stand of trees or a savannah-type environment would be the most preferred (3, 7, 10). However, such studies of visual preferences were based on photographs of forest settings that probably conveyed more information about the visual environment than the short verbal descriptions of vegetation used in this study.

While users are willing to pay a substantial amount for trees and forests, they are willing to pay even more for water resources; lack of crowding, vandalism, or litter; and a certain age distribution of users (Table 1). In terms of what users were willing to pay, trees and forests ranked sixth overall among 20 site attributes (excluding entry fees and travel times). These results suggest that trees and forests certainly can claim their share of attention by managers of urban parks and recreation areas.

The above findings offer strong support for the planting and preservation of trees in urban recreation areas, but public support for tree maintenance efforts beyond those essential for tree survival is difficult to identify from this study. Several of the attributes included in the choices do, however, point to the importance of maintenance—and might be extended to suggest support for tree maintenance. For example, average willingness to pay, per visit, was $0.33 for mowed grass, $0.49 for “structures and facilities well maintained,” and in excess of $3.00 for an absence of lit- ter/trash, and an absence of vandalism. Taken together, these responses reflect support for overall maintenance of the area that perhaps may extend to trees.

Variations in preferences and values

The foregoing results reflect the average preferences of the total set of respondents. However, we have reason to suspect that there may be substantial variations in their preferences for an willingness to pay for trees and forests. Individuals landscape their homes in different ways, they are drawn to various kinds of settings in urban parks and forest preserves, and they engage in a wide range of outdoor recreation activities—many of which require quite different outdoor environments. In previous research, variations in preference have been related to two major attributes of forest sites; the density of vegetation and the intensity of development. Inner city natives seem to prefer neatly maintained sites with not too many trees, while suburban residents prefer densely forested, natural sites (6, 12).

To identify variation in preferences in this study, respondents were clustered into “market segments”—based on similarities in their park visitation patterns and socio-economic characteristics. Each of the five “market segments” that contained a sufficient number of respondents to support statistical analysis was then analyzed separately.

Across the segments, trees and forests varied in importance from second to tenth among the 20 site attributes (excluding user fees and travel times). The only site attribute that was more important than vegetation for all five segments was crowding. While the ranking of the attributes varied significantly over the five segments, trees and forests always ranked higher than terrain, childrens playgrounds, or athletic fields.

To illustrate some of the variation in choices made by different users, consider an example involving four parks that differ only in vegetation. Table 2 shows how users from each market segment, as well as the total, would distribute themselves over the four parks (assuming all other attributes remain the same). All five segments would make some use of each park. Segment 3 would make the least use of a park with very few trees, while segments 2 and 5 would make the heaviest use of densely forested sites. Four of the five segments would be most likely to choose the most heavily wooded site, although one of those segments would be almost as likely to choose the second most forested site. One segment would be most likely to choose the second most forested site.

Three market segments (containing 29% of the respondents) were attracted to sites with higher fees. This may have been out of a belief that sites with higher fees would somehow be more desirable due to lower levels of use, absence of anti-social behavior, etc. Consequently, we were able to estimate the willingness of users to pay for particular attributes for only two segments. This situation also lowered the “average-willingness- to-pay” values for attributes calculated for the entire set of respondents, including the values presented above for trees, forests, and maintenance.

Willingness of users to pay for the most wooded site over the least wooded site were $1.23 per visit and $2.75 per visit for the two market segments for which comparisons could be made. This is a fairly substantial range of values and indicates how much preferences for trees and forests can vary among individuals.

Mowed grass was viewed positively by two of the five market segments (i.e., it increased their probability of choosing a site), and well-maintained structures and facilities were viewed positively by four of five market segments. For the two market segments for which calculations could be made, one was willing to pay an average of $0.36 for mowed grass and the other was willing to pay an average of $1.03 per visit to have grass that was not mowed. This variation in preferences for “naturalness” may have signficant implications for managing sites with trees and forests.

The characteristics arid preferences of the five market segments are complex and difficult to interpret straightforwardly. The following examples offer some insight into two of the segments—with reference to trees and forests. The segment with the highest willingness to pay for a heavily forested site has the strongest aversion to mowed grass, likes hiking, but not picnicking. This conjures up the image of users who prefer a relatively “natural” forest for their outings. The segment that appears least attracted to heavily forested sites is heavily oriented to state and county parks and likes fishing, swimming, picnicking, and athletic fields—activities associated with open and fairly treeless areas.

Continued research on the preferences and choices of forest users will better define market segments, perhaps focusing on particular activities or groups of activities. We are currently focusing on urban and suburban bicycle trails—the recreation facility that had the highest average utility in this study. At the same time, different levels of tree and forest maintenance should be specified in the choices—perhaps through photographs. Users differ in their preference for urban forest areas that are “well manicured” versus those that present a “natural environment.” It is not yet clear what standards of tree and forest management the recreationists want, but given the high cost of maintenance, it is reasonable to conduct additional research to identify their preferences and willingness to pay for the results of these activities.

The fact that some segments preferred sites with higher fees made it difficult to estimate the willingness of users to pay for a particular attribute. Individuals probably associated higher fees with other attributes such as reduced crowding and more “respectable” users. The same may be true with litter/trash, vandalism, and conflict among users. In future research we must design and present the choices in a manner that minimizes such confounding. Given users’ willingness to pay for site attributes, the range of fees should be extended beyond $3.00.

Two potential sources of bias must be considered in interpreting the results of this study. First is non-response bias, due to the fact that only 44% of the park users randomly contacted by telephone actually completed the survey. If this 44% differs in some way from the users who did not complete the survey, then their choices may not be representative of the general population of park users. This potential source of bias is present in any research in which respondents have a choice of whether to participate or not.

The second potential source of bias is due to the hypothetical nature of the choices in the scenario. We do not know how well choices among hypothetical, verbally described parks correspond to real choices among real parks. Data from an earlier study of park choices in Iowa (4) suggest that people’s real choices are reasonably similar to their choices among hypothetical parks, but a definitive test of the correspondence has not yet been made.

Summary

Trees and forests are important to the users of urban parks and forest preserves. This importance is reflected in users’ willingness to pay to have these features in a park and their likelihood of choosing a park with these attributes. The dollar values of trees are substantial and compare favorably with dollar values of other park resources. Although the willingness of users to pay for tree maintenance in parks was not directly addressed, their willingness to pay for other maintenance suggests that tree maintenance may also be important, and the results argue strongly for those maintenance activities essential to maintaining a forest environment in parks. The data also revealed substantial variations in individual preferences for vegetation in urban parks—with implications for trees, forests, and their management.

Annotation

Reports on a logit choice model for urban recreation areas that was developed from an experiment in which Chicago residents who use parks and forest preserves were asked to choose between paired descriptions of hypothetical recreation areas.

This manuscript was written and prepared by U.S. Government Employees on official time and therefore in the public domain.