Next Article in Journal
Prediction of Land Use and Land Cover Changes for North Sumatra, Indonesia, Using an Artificial-Neural-Network-Based Cellular Automaton
Previous Article in Journal
What Factors Are Necessary for Sustaining Entrepreneurship?
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Economic Valuation of an Urban Lake Recreational Park: Case of Taman Tasik Cempaka in Bandar Baru Bangi, Malaysia

1
Faculty of Economics and Management, Universiti Kebangsaan Malaysia, UKM Bangi Selangor 43600, Malaysia
2
Institute for Food and Resource Economics, University of Bonn, Nußallee, 2153115 Bonn, Germany
*
Author to whom correspondence should be addressed.
Sustainability 2019, 11(11), 3023; https://doi.org/10.3390/su11113023
Submission received: 11 February 2019 / Revised: 10 April 2019 / Accepted: 11 April 2019 / Published: 28 May 2019
(This article belongs to the Section Sustainable Urban and Rural Development)

Abstract

:
There is growing public support for an outdoor, nature-based urban park, which offers the local population a wide range of recreational services for an improved quality of life. This study estimates the economic value of recreational benefits for the case of a lake-based urban park known as Taman Tasik Cempaka (TTC) in Bandar Baru Bangi in Selangor, Malaysia. The Travel Cost Method (TCM) was used to generate the demand function for park visitation and the recreational benefits were estimated using the Poisson and Negative Binomial (NB) models. The net benefits of recreation per visitor were evaluated at MYR 6.2 per trip while the price elasticity of demand was highly inelastic at −0.48. The result provides support for the imposition of an entrance fee and the subsequent revenue collection may be used for park upkeep and conservation.

1. Introduction

Urban parks are outdoor, nature-based sites that primarily serve a defined urban area to conduct an array of recreational or leisure activities, ranging from indulgence in the tranquil, natural environment, to active and participatory activities such as fishing, bird watching, jogging, children’s playgrounds and other activities specific to the biophysical and landscape features of the parks. Urban parks provide urbanites with the important amenity values that can reduce the visual impact of an environment dominated by asphalt and concrete. In this way, parks decrease anxiety levels and promote public health, thus enhancing the overall quality of life of the population [1,2,3,4]. Urban parks, especially those with a larger number of trees, may also contribute to public health by removing the presence of pollutants from the air [5] and, to a certain extent, mitigate urban warming [6]. Recreation users tend to report fewer medical visits for purposes other than check-ups than do non-park users [7,8], along with reported lower anxiety and stress levels [9].
In Malaysia, like in many countries across the globe, the provision of nature-based recreational services via urban parks have become a necessity for holistic, sustainable urban living. Typically in Malaysia, the state or federal government is responsible for the development and financial provision of such parks, while the local government is entrusted to manage and maintain the parks. Given the open access nature of the resource, services of urban parks are routinely provided free of charge. However, as governments are increasingly faced with budgetary constraints and changing priorities, the quality and long-run maintenance of urban parks may be affected. In addition, if the parks are overused the social net benefits of the parks reduce. Entrance fee is a measure for sustainable finance purposes when the park authorities’ face budgetary constraints. The measure could also be utilized to either reduce the number of visitors or redistribute the time of visits when the park is crowded during certain days or times of day.
The relevant issue is, how can efficient entrance fee levels for public or urban parks be determined? This has important implications for the long-run sustainability of parks. Inefficiency due to no charging or under-charging will hamper the ability of park management to maintain necessary resource upkeep. Furthermore, it induces overly high visitation rates and accelerates resource degradation by users. On the other hand, overestimation of access fees for recreational sites may lower tourists’ recreational benefits, reduce visitation rates and negate the intended purpose of the park’s existence. The imposition of an entrance fee for visiting urban parks is, however, not an easy task. Given that urban parks are public goods, it is the role of municipalities to provide residents with free and equal opportunities to benefit from recreational services and the external effects of such policies, such as their impact on public health, should always be considered. Further, urban parks have an open access character and the cost and technicality of limiting access to the park, as well as benefit-cost analysis of levying a fee, should be considered beforehand. Nonetheless, this study confines itself with the measurement of an entrance fee.
In the context of the demand for recreational services from urban parks, economic theory asserts that the admission fee is efficient when it is set at the level at which tourists’ marginal recreational benefits (MRB) are equal to the associated marginal cost of provision, which includes maintenance, resource degradation and congestion cost. At this level, social welfare (benefits) is at its maximum. However, the marginal cost of provision is normally indeterminate and is likely to increase with visitation rates. Given such a situation, an efficient fee may be determined by equating it with the known or estimated MRB of the representative user.
From the theoretical perspective, recreationists’ willingness to pay (WTP) to obtain a benefit from a given recreational resource reflects the MRB of said resource. Subtracting the actual service and user fees, including the admission fee from the estimated MRB, one can obtain the net recreational benefits or consumer surplus (CS) from a given recreational experience. In the absence of knowledge of the marginal cost of provision, as is always the case, an admission fee level is deemed inefficient if it is substantially different than the average net recreational benefits per trip.
The MRB or marginal WTP for recreational attributes held by visitors can be estimated using environmental valuation techniques. Knowledge of MRB and net benefits, as noted above, is necessary for the estimation of an efficient admission fee level for tourism/recreational sites. This study presents an application of economic valuation of recreational benefits for the case of TTC in Bandar Baru Bangi, in Selangor, Malaysia. The estimated economic benefits of recreational experience will form the basis for identifying an efficient entrance fee for the park.

2. Background of Taman Tasik Cempaka

Bandar Baru Bangi, located in the State of Selangor, Peninsular Malaysia is a relatively new township. It was developed in stages, beginning in 1977. The area used to be a rubber plantation and secondary forests that were gradually cleared to make way for urban settlements. In the early years, there was no place in the area for its inhabitants to indulge in recreational activities. The city grew very rapidly with the establishments of industrial zones, academic and research institutions, the country’s new airport (KLIA), as well as the new administrative centre (Putrajaya), all within 10 to 40 kilometres of the city centre.
At the centre of the designated city, there is a stretch of barren lowland constantly covered by water generated from the rain that made up a small river way, named Air Hitam river, that connected to the main Ulu Langat river. The site is strategically located, as it lies adjacent to the downtown area. This impelled many private entities to propose various commercial-based development ventures on the site, including a large scale hotel facility. Since there were no nature parks within the city, while new residential areas were expanding rapidly, federal and state government authorities were under increasing pressure from public representatives to upgrade the small stream into a water retention and recreational lake area and they did just that in early 1997. Initially, the main aim was to provide river and lake-based recreational activities for the city inhabitants, as well as to create a natural ecological zone, thereby attracting wildlife to the area (survived information from Mohd Fadrillah Taib of the National Landscape Department, principal designer of the site).
The National Landscape Department and the state government further agreed to develop the area on both sides of the water retention lake to incorporate various recreational services to meet the needs and preferences of the various residents of Bandar Baru Bangi and its vicinity. Based on surveyed information provided by the principal designer of the site (Mohd Fadrillah Taib), the aim of expanding the site was the provision of social and ecological services. The social aim was to facilitate sport and family-based leisure activities for the local populace, while the ecological aim was to draw more floras and faunas into the area. An implicit but important intention in expanding the recreational site area was to send a clear signal to the public that the area had been well designated for a nature-based recreational facility and shall not be considered for alternative uses.
The TTC park was developed in three phases and financed by the Federal Government under the Ninth Malaysia Plan. Phase I, with an area of 3.86 ha, was implemented in early 2001, Phase II, with 2 ha in early 2003 and the final phase (4.3 ha) was completed in May 2008 [10]. The entire area (43 ha) is now fully developed as an open-access lake gard|en providing the populace with a wide range of leisure activities and recreational facilities encompassing 17.5 hectares of water bodies and 7.1 ha as habitat for various floras and faunas. Recreational facilities include a 4-kilometer jogging track, skating rink, futsal facilities, lake-view rest huts, rest tunnels, an open-air amphitheatre, kite flying area, fishing spots and open space for family or group-based activities. A public library was also built beside the lake garden. The Bandar Baru Bangi town proper has a population of about 250 thousand (as of 2010), nested within the district of Ulu Langat with a population of some 1.16 million.
The overall budget allocated to develop the three phases of TTC was MYR 13.2 million (survied information from Assistant Director in the National Landscape Department (Mohd Saifuddin Ali)). Further, an interview with Mr. Abdul Razak Abdul Rahman, local agriculture officer in the Department of Development Planning, shared that the Kajang Municipal council relayed that the cost for park upkeep is some MYR28 thousand per month. This includes regular maintenance, cleaning, fertilizing and other associated activities.
Correspondence with the Senior Assistant Director in the National Landscape Department (Mohd Saifuddin Ali) revealed that there was a strong collaboration between the various government agencies, particularly between the National Landscape Department, that provided the development funds, Selangor State Development Corporation, that allocated the land and was responsible for river widening and beautification works and the local authority Kajang Municipal Council, that acts as park manager. Such effective collaboration across federal, state and local governments demonstrates clear awareness of the various authorities about the local needs for open spaces and nature-based recreational facilities. This study attempts to estimate the economic value of recreational services provided by TTC from the user perspective. The information gathered from this study will contribute to a better appreciation of the scale of importance of TTC for the local citizens as well as visitors from nearby areas. Prospects for the imposition of an admission fee based on the estimation of recreation benefits are deliberated at the end of the paper.

3. Literature Review and Methods

A host of valuation techniques to value the recreational benefits of a single recreational site exists in the literature. The choice of method depends on the types of recreational value that one wishes to estimate. If one views recreational services as an aggregate good, then the revealed preference approach, particularly the TCM, is most appropriate. If one wishes to estimate the value of the individual attribute that constitutes the recreational good, the Choice Model (CM) or Contingent Ranking Method (CRM) methods are popular approaches. However, if the change in the level of recreational service provision is the main focus, the more appropriate technique is CM. For the valuation of non-use values of recreational service flows, Contingent Valuation (CVM) and CM are widely used. Most studies on the economic valuation of tourism or recreational services in South East Asia and worldwide have applied the TCM, followed by stated preference techniques such as CVM and CM [11]. Some other studies combined the TCM with contingent behavior questions to estimate recreational benefits or use values (For example, [12]). In general, for the valuation of recreational benefits, the TCM has been the most adopted approach worldwide [13] (see References [14,15,16,17], for more recent applications of TCM).
This study evaluates the recreational benefits of TTC from the perspective of recreation as the entire good, hence the TCM will be employed. The TCM assumes that users or recreationists have to travel to recreational sites to derive recreational services. This requires consumers to incur the cost of travel in return for access to the site and hence, its recreational benefits. The travel cost data and the average number of trips or time spent on site are used to estimate the demand function for recreation at the site. Hence, it is intuitive that respondents who live far away will sustain higher travel costs and therefore pay visits less often. It further assumes that respondents who will not pay any travel costs will not make any visits with recreational benefits then equal to zero.
Suppose that the representative individual’s utility function is an argument of the number of visits to an urban park, z and the quantity of all other private goods in aggregate, q. Suppose the round-trip travel cost for a visit to the facility is given as p. With the price of the aggregate good normalized to equal one and assuming separability between the purchase of q and recreation, the individual’s budget constraint is given by pz + q = I, where I is income. Maximizing the utility function, U(z, q), subject to the budget constraint, yields the standard Marshallian demand function for recreational use of the urban park: z = f(p, I). This demand function can be generated with the travel costs of substitute sites and other demographic factors that affect visitation rates as well as demand curve shifts. However, for localized or neighbourhood urban parks such as the TTC, it is thought that its main role is for routine, leisurely indulgence, either on an individual basis or with family members or close friends. Hence, substitute sites may not influence visitation rates significantly.
Travel costs typically include on-site time costs. Since a neighbourhood urban park is typically close to the homes of the users, on-site time costs may constitute a substantial portion of travel costs and also greatly vary across users. The value of lost opportunities due to recreational indulgence at the park will be the measure of the cost of time. The demand function for TTC visitation can be written:
h = β t c T C + β D i D i
where h (dependent variable) is the number of trips per month or hours of recreational time spent per month in the site, TC is round-trip travel cost which includes on-site time, while Di represents a host of qualitative socio-economic and demographic variables such as age, education level and motivation to visit the site.
To evaluate the net benefits (consumer surplus) of recreational experience to users, the idea is to measure how much h will be lost if the average current price of visiting the site denoted by T C i ¯ increases to a choke price T C * (a price high enough so that the visitor chooses not to visit the site at all). Hence, based on the inverse demand function for recreation from (1), we can integrate the monthly visits or time spent onsite with respect to TC (from T C i ¯ to T C * ) to obtain the monthly consumer surplus (CS), which represents the net benefits obtained by visitors to the urban park:
C S = T C T C * h ( T C ¯ , D i ¯ )

4. Econometric Analysis

The usual approach to model the demand for recreation is using count data models [18,19]. Even if the OLS estimator could be used for such an analysis, it is less efficient in the presence of a count outcome variable [20]. The dependent variable in Equation (1) is typically a positive integer where small observations of the variable (number of trips to the site) constitute a large portion of the data. Count models are more suited for such data [18,19].
The basic count data for most travel cost models is a Poisson regression (PR) where the number of trips by an individual to a park is assumed to be generated by a Poisson process. The probability of observing an individual taking h trips or spending h hours in a month is written:
Pr o b ( h ) = exp ( λ ) . λ h h !
where λ ( = exp X i B ) is the parameter representing expected number of the dependent variable and is a function of the arguments in the demand model as specified in Equation (1). Since λ > 0, it normally takes a log-linear form (Equation(4)):
l n   (   λ ) = β t c T C + β D i D i
and its parameters are estimated using the maximum likelihood estimator
The Poisson model assumes that the mean of distribution equals its variance. However, in practice, this assumption is often not met and the data show a greater dispersion. This is true for our study where our sample comprises over-dispersed data. In these situations, an alternative is to include a stochastic variable ( ε ) that allows for heteroscedasticity across people and allows λ = exp X i B +   ε i to vary according to a specific probability law. If exp ε i follows the gamma ( Γ ) distribution, then the compound count data generation process follows a negative binomial (NB) distribution [21]. The NB distribution has conditional mean λ i and conditional variance λ i   1 + a λ i where a   a   0 describes the over-dispersion of the data.
Pr = [ T = h ] = Γ ( α 1 + h i ) Γ ( α 1 ) + Γ ( h i + 1 ) × ( 1 1 + α λ ) α 1 × ( α λ 1 + α λ ) h i
In this model, if α = 0 , the model reduces to Poisson.
Apart from over-dispersion, there are two additional issues related to the data collected from an on-site survey: truncation and endogenous stratification [22]. Truncation refers to the fact that people with zero trips to the site are not represented. Endogenous stratification refers to the situation where more frequent visitors are more likely to be sampled than less frequent ones. If not properly addressed, these features of an on-site survey lead to an estimation bias. Englin and Shonkwiler [22] demonstrated that a Poisson model may be corrected for truncation and endogenous stratification by means of the same probability mass function shown in Equation (3) but replacing h with h 1 as the response variable. Further, the NB model is corrected for truncation and endogenous stratification by modifying Equation (5) as follows [23]:
Pr = [ T = h ] = Γ ( α 1 + h i ) Γ ( α 1 ) + Γ ( h i + 1 ) × α i h i λ i h i × ( 1 + α i λ i ) h i
Finally, when Equation (4) is estimated using count data models, the consumer surplus for every respondent in Equation (2) could be reduced to the formula below:
C S n = λ n β t c
where subscript n denotes individual n. To obtain an average estimate of CS per trip or per unit of time spent on-site, we simply obtain the value of the reciprocal of the coefficient of the respective travel cost (per trip or hourly basis).

Survey Design and Implementation

A structured survey was conducted to solicit information on the socio-economic and demographic profile of park users. The questionnaire was designed to obtain two types of information: (i) information required to estimate the TCM model and (ii) information on the visitors’ perception of the environmental quality of the facility, their attitude in relation to different functions of the recreational area and the prospect of the imposition of an entrance fee. Assessing the visitors’ perception, attitudes and their motives of trip, helps park management ascertain if it is desirable to update the park with additional facilities.
Questions were posed about the place of residence of the group to see if they are from the town in which the lake is located or other areas and what the purpose of their trip was, for example, whether the trip was on an individual basis or family oriented. Further questions were posed on travel distance, time and money spent on the site, types of vehicles (Car and Van, Motorbike, Bicycle, Walk and public transport) used, the number of friends or family members who came with the respondent and their age, gender, occupation (executive and non-executive job), educational level and monthly household income. Questions on age and income were represented in intervals scales so that respondents did not need to provide their exact age or income. To obtain the information on education variables, the respondents needed to select among different academic ranks (SPM, STPM, Diploma, Degree, Master, Ph.D. and other).
To obtain the second category of information, the questions asked for the motive (leisure or indulgence in sports, viewing of the environment, spending time with friends and family, release of anxiety, tensions and stress) of the visit and the general quality of natural environment (presence of water, pollution, land pollution, noise pollution and social problems). Questions were also posed if, from the visitor’s perspective, the park was overly used and if they would be willing to pay an admission fee.
Preliminary questions were initially deliberated among research members and pretested. The pretesting of the questionnaire was conducted at the park to test for the presence of any cognitive burden and incongruences. After improvement of the questionnaire was made based on feedback from the pre-test, the actual data collection survey was conducted in November 2015 by five trained enumerators. The survey was implemented over a week during the weekdays and weekends. The choice of days was made to minimize any problem of endogenous stratification of samples that may result from the on-site sampling of users. The enumerators were placed at major throughways of the site and directed to select the respondents randomly. The respondents were asked if they prefer the numerator to read the questions for them and fill it in accordingly or if they prefer filling it themselves and asking the numerators if interpretations were needed.
Double counting of respondents from the same group was avoided. Some 620 visitors (out of approximately 3100 visitors during the week) were asked to participate in the survey, with 510 responding to our request, with the others not wanting to participate. The 510 respondents yielded a valid sample of 480. We rejected samples that came from overly long distances, those who visited by chance and incomplete questionnaires.

5. Results and Discussions

This section discusses the profile of respondents from the survey followed by an empirical analysis of the TCM.

5.1. Profile of Respondents

Based on the data from the survey, as expected, the park was more composed of visitors who live closer to the site (Table 1). The park was also crowded with the age group where the mean age is 29 years old. Most of the respondents were male, single, holder of school certificates and college diplomas and used their own vehicles. Note that while we selected the group of visitors randomly, in most cases the head of the family (man) participated in the interview. Comparing the age distribution of the sample with the age distribution of population truncated with the minimum age in the sample group shows almost similar age structure. The income structure shows that the median income is some 600 MYR below the average country median and 60% of the population mean. The education structure in the sample is well representative of the population group.
Lake visitation was the main intent of the site visit during the survey, relative to other purposes such as shopping at nearby shopping centres or visiting relatives and friends. On park’s visitation, about 40 percent of respondents reported leisure or recreation motive as the main impetus, while indulgence in sports (jogging, futsal, skating and kite flying) constituted a prime motivation for 20 percent of respondents. Other reasons included the viewing of the environment, spending time with friends and family, release of anxiety, tensions and stress. A minority (4%) also mentioned fishing at the lake as their main intention. A large (95%) of respondents declared that they came with either friends or family members. This reflects clearly that the TTC is very much a place for social family and group activities.
A large majority (95%) of the visitors announced their general satisfaction with the quality of natural environment at TTC. Nonetheless, 55 percent of visitors acknowledged water pollution was the main distraction, 38 percent reported the existence of rubbish and littering, while 20 percent complained about slope erosion and noise pollution. Further, about 13 percent reported the nuisance that came from traffic congestion (i.e., parking sites), while 10 percent noted the presence of social problems such as indecent behaviour among the youth and the lack of praying room, especially given that the Bandar Baru Bangi area is inhabited by mainly Muslims.
The respondents were also asked plainly about whether or not they would agree to pay an admission fee for maintenance and conservation of the park. A large majority (65%) seemed opposed to such idea, attributing it to the responsibility of the government, only 15 percent agreed, while the rest simply abstained. Of the 65% who opposed to entrance fee, 80% were those who either visited the park on weekends and reported traffic congestion at the parking sites and that the park was crowded or were those who reported other problems such as water pollution, soil erosion along the banks, littering, as well as the lack of religious facilities, such as a prayer room. Furthermore, the majority of them (79%), indicated that they might revise their decision about the entrance fee depending on the amount of the fee and whether or not their concerns would be addressed by the park authorities. We see the disapproval from the others of the entrance fee either as a form of protest to the possible imposition of a fee (78%) or because they were students (12%) or had an income below 2000 RM (10%).
Furthermore, the respondents were asked about on-site and travel expenditures. The majority of park visitors spent their recreation time on very low travel budgets (Table 2). On average, about MYR5 was spent on food and beverages and MYR6 on petrol.
Details on travel data are shown in Table 3. Group size averaged about 3.33 persons per respondent. The average number of trips per month was 3, with time spent ranging from 0.17 to 3.5 h and an average of close to an hour (0.99 h) per trip. This gives an average time spent on-site per month of 3.83 h. Further, the data reveals that those people who often visited the park stayed for shorter times and vice versa.

5.2. Application of the Travel Cost Model

The following specification was used in the TCM analysis:
V i = α 0 + α 1 T C i + α 2 D _ A g e i > 30 + α 3 D _ E d u i + α 4 D _ B a n g i i + α 5 D _ O c c u p a t i o n i + α 6 D _ P e r s o n a l i + α 7 D _ L e i s u r e i + α 8 D _ G e n d e r i
Table 4 provides the explanation for all the variables, their measurements and mean values. The dependent variables ( V i ) utilized in the model is number of trips to site per month.

5.3. Empirical Results

Equation (8) was estimated using the Poisson and NB models and the results are depicted in Table 5.
The two models are consistent with broadly similar parameter estimates. The log-likelihood of the Poisson model (PR) is slightly higher than the NB model. However, the NB model shows substantially lower AIC and BIC statistics indicating that it fits the data better than the Poisson model. Moreover, the significance of the coefficient for over-dispersion parameter, Ln(a), is suggestive of NB model, which takes over dispersion of the data into account. Note that in our dataset, as reported in Table 3, the mean number of trips is 3.02, while the standard deviation is 3.67, placing the variance at 13.5, over four times the value of the mean. Several others also shown that the NB model performs better than the Poisson model [15,25,26]. Nonetheless, in our estimations the two model estimates are quite similar in magnitude and significance level.
The coefficient for travel cost in both models has the expected sign and is highly significant. Very interestingly, the coefficient for the leisure variable signifies that respondents who admitted recreation/leisure as the top reason for TTC visitation are positively associated with the number of monthly trips to the site. Another striking observation is the negative coefficient for friends, which reflects that respondents tend to have lower visitation rates if they came with friends or family members, relative to visitors who came alone. This may be mainly due to the general leisure, group-based activities, relative to those who came alone for relatively prolonged self-indulgence in certain sports or nature. Occupation and education seem to have no significant effects on visitation rate, suggesting that the TTC is a place for all walks of life. The Bangi residence dummy is also not significant in all the models, signifying that quite well balanced visitors were coming from both Bangi and nearby non-Bangi areas. It is also worthy to note the significance of the D_Age>30 variable, which indicates that the more senior respondents tend to visit the TTC more often.
The consumer surplus estimates are calculated and shown in Table 6. Each of the CS per hour and per trip were estimated based on the overall sample, that is, the reciprocal of the respective travel cost coefficient is divided by the average group size (3.33) to obtain the per individual basis. The two models yield almost the same estimates of consumer surplus (5.9 and 6.2 MYR). Note that the per trip CS was about the same as the per hour basis, as average on-site time per trip was about 1 h. The estimates signify that on average, each individual visitor is willing to pay some MYR 5.9 or 6.2 more per hour of recreation time relative to the travel expenses the individual actually incurred. This estimate is comparable to that of the case of Malaysian Agricultural Park in Shah Alam [27].
We also calculated the own-price elasticity of demand to obtain insights on the sensitivity of time spent on-site and the number of monthly trips due to changes in travel costs. The Poisson and NB models generate an estimate of the own-price elasticity of −0.51 and −0.48, respectively. The expected small estimates suggest that on-site time and visitation rates are highly inelastic to changes in travel cost. For each model, a 10 percent increase in travel cost per hour reduces onsite visitation time and number of trips monthly by 5.1 percent and 4.8 percent, respectively. These estimates are comparable to that of Kaffashi et al. [28] for the case of Malaysian National Park in Penang.
The aggregate benefits of recreational services per time unit (month or year) provided by TTC can be calculated by multiplying the estimated consumer surplus per trip per individual with the known total number of trips or individuals over a month or year. This study surveyed 510 respondents over a period of one week. Given the average group size of 3.3, we can reasonably expect at least a visitor stream of some 510 × 3.3 × 4 = 6732 per month..Multiplying this with the average number of trips per month (3) yields an estimate of 20,196 for total number of trips per month. This generates aggregate monthly recreational benefits of 20,196 × 6 = MYR 121,176. Now, should the park be closed for visitation indefinitely regardless of the reason, these aggregate values of net benefits would be a measure of undiscounted economic loss per month. The estimate of economic loss due to the closure of recreational service provision in present value terms can be simply approximated by dividing the presumably constant annual stream of net benefits by an appropriate social discount rate. Based on a 10 percent discount rate and presuming no change in visitors flow over the years, the lower bound present value of economic loss would be some MYR 14 million.

5.4. Appraisals of Admission Fee

Currently, no admission fee to TTC is being administered. Empirical results have shown that the estimated net benefits (consumer surplus) of MYR6 per trip is quite substantial. It signifies the prospect for an imposition of an admission fee by the responsible authority. We investigated the behaviour of the change in net benefits (i.e., marginal consumer surplus) on average as visitation time increases per visit ceteris paribus and was then compared with that of the change in travel cost (i.e., marginal travel cost or also known as marginal supply cost).
Consumer surplus for every visitor per hour per trip was calculated based on Equation (7) and the estimates of the NB model, while marginal CS was estimated by using the regression coefficient of CS against on-site time and its square (semi-log dependent variable specification). The same procedure was used to calculate marginal travel cost. The calculation is shown in Table 7 while Figure 1 plots the Predicted consumer surplus (Predicted CS), Marginal consumer surplus (Marginal CS), travel costs, as well as marginal travel cost against on-site time per visit. WTP and marginal WTP (MWTP) are also shown. Note WTP is the sum of CS and total travel costs. Theoretically, the marginal WTP curve represents the representative visitor demand curve which links time spent on-site ceteris paribus and travel cost. Each point along the demand curve reflects the marginal benefits accrued for the visitor for each additional unit of time spent on site.
On average, CS per hour reaches maximum (MYR19.91) when MWTP (marginal benefits) equals marginal travel costs and the equilibrium visitation time is 2.39 h. The CS initially rises at an increasing rate until it reaches 1.39 h on-site duration; thereafter it diminishes and becomes negative from the point of intersection of the marginal WTP curve and marginal travel cost at the equilibrium on-site duration (2.39 h). Results show clear declining marginal utility as the consumption of recreation increases for each trip, that is, when on-site time increases beyond 1.30 h.
The imposition of any level of admission fee per trip will shift the marginal supply cost upwards and result in reduced net benefits. Theoretically, the admission fee level per entry can be set at any level below the consumer surplus estimates to allow for some positive net benefits to visitors. Visitors may also optimize their visitation time in order to compensate for the additional cost. For instance, if the admission fee is set at MYR1 per entry, based on Table 7, for a visit duration = 0.89 h, the visitor will incur a decline in net benefits by MYR1. However, by increasing onsite time by 16 min, the additional gain in consumer surplus of MYR1.76, more than offsets the increase in cost, resulting in a small net gain in benefits. In the absence of knowledge of the marginal provisional cost of recreation as argued at the offset of the paper, an admission fee may be set equal to the MWTP at the average on-site visitation time (i.e., 0.99 h). In Table 7, the relevant MWTP lies between MYR2.78 – MYR3.03 and the associated change in net benefits (CS) is between MYR1.61–MYR1.76. Therefore, the admission fee ought to be set equal to the change in CS, that is, between MYR1.6 –MYR1.76, which satisfies the utility maximization condition (MWTP = MC), subject to the average visitation time. This leads to an average change in CS (i.e., proposed entrance fee) of MYR 1.70 per trip per individual on average. This represents a reduction of CS of about 18 percent for the average visitor. An alternative approach is to set the fee at the level equal to the average of the positive marginal change in CS, that is, from the minimum on-site time through the equilibrium duration (2.39 h). This gives a value of about MYR1.30 per trip per individual.
Assuming the lower bound estimated average entrance fee of MYR1.30 per trip was administered and given the highly inelastic demand, it can be expected that park management may be able to raise some MYR26,000 of revenue per month. Such a sum is significant as it constitutes some 93 percent of the monthly budget set aside by the local authorities for the necessary upkeep.
However, the decision to levy charges on the people who use public spaces is not an easy task and several issues should be considered beforehand. First, an evaluation of whether or not the authority is technically able to collect an entrance fee for an open urban lake park with many access points is required. Second, the cost of limiting the access area, administrative costs of a fee system, the legal provision of the imposition of the entrance fee and whether the imposition of the entrance fee falls within the jurisdiction of the local government should be considered. Third, an urban park is a public park (i.e. public good) and it is the task of the municipality (the public sector) to provide the local residents with recreational opportunities for free and give equal access opportunities to those wishing to visit the park. Forth, the imposition of an entrance fee would reduce the number of park visits and that may have negative external effects on public health. Fifth, the analysis discussed here focuses on whether the sum of all benefits exceeds the sum all costs (direct and opportunity costs) and if there are positive net benefits the imposition of an entrance fee is justified based on the premise that the winners could compensate the losers. However, an important part of a policy decision is who gains and who loses. The distribution of benefits and costs is certainly important and should be studied. Finally, if the imposition of an entrance fee is the only option remaining to ensure the sustainability of the park, a fee system that minimizes the unwanted impact of such a policy should be investigated.

6. Conclusions and Policy Implications

Urban parks provide an array of different recreational opportunities.. They help reduce stress and the anxieties of urban life and foster community relations, balancing the built and natural environment and therefore contribute to the overall well-being of the urban community. Inevitably, the development of urban parks and their associated recreational service provisions are an important element in contemporary urban development planning policies. This study presents an economic valuation of the recreational services provision of a small urban, lake-based park, TTC in Bandar Baru Bangi (a small township in Selangor, Malaysia). The historical development of the park demonstrated the clear roles of the relevant federal, state and local authorities, as well as public forces in determining the development decision. To assess the economic benefits associated with the recreational service flow, the TCM was used to estimate the net benefits accrued to park users. An important aim of the study was to identify an appropriate level of an entrance fee. The imposition of a fee may be necessary in light of park sustainability issues that potentially emanate from increasing budgetary pressures faced by park managers as well as park overuse.
The profile analyses of some 480 respondents indicate that the park is well liked and suited for neighbourhood, family-based recreation for Bangi residents and its vicinity, with an average trip of 3 times per month and an average visit duration of around 1 h per visit. However, there have been notable environmental problems cited by respondents. These include water pollution, soil erosion along the banks, littering, traffic congestion at the parking sites and a lack of religious facilities, such as a prayer room. A well-planned development program with adequate financial resources would be needed in order to see substantial park improvements and consequently ensure the long-run sustainability of quality recreational services.
Empirical results show the undiscounted aggregate value of recreational benefits to the tune of MYR121,176 monthly, implying that the park provides considerable benefits for the community. An average entrance fee of some MYR1.30 per individual per entry could be imposed to utilize the considerable net benefits visitors are enjoying from park visitation. Given the large net benefits as well as highly inelastic demand, it is expected that the imposition of an admission fee may not reduce net benefits and visitation rates substantially. Some MYR28,000 of revenue per month can be expected to be raised from such a policy. Revenues from the entrance fee could be directed towards park upkeep and conservation. However, before levying an entry fee, various issues such as distributional issues of levying an entrance fee, right of equal access to public space and the positive external effects of park visits on public health must be considered, with the possible fee system minimizing the unwanted impacts. Furthermore, before implementing an entrance fee system, further appraisals with respect to legal provisions and the additional administrative costs are important. These may encompass the issue of tickets or pass accessibility, as well as problems of multi-entry points given the vast space of the park.
Policymakers can use the result of this study for better management of the park. Given that there is a high demand for the park and that it generates significant net benefits, policymakers must react by providing the park with more amenities. There is a need for the enhancement of the quality of current recreational services, especially in respect to the problems with pollution and limited parking areas as highlighted in this study.

Author Contributions

Conceptualization, J.O. and Y.J.; Formal Analysis, J.O. and Y.J.; both authors wrote and reviewed the paper.

Funding

This research received no external funding.

Acknowledgments

The author would like to thank the three anonymous reviewers for providing constructive and insightful comments. The usual caveat holds.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Chiesura, A. The role of urban parks for the sustainable city. Landsc. Urban Plan. 2004, 68, 129–138. [Google Scholar] [CrossRef]
  2. Tzoulas, K.; Korpela, K.; Venn, S.; Yli-Pelkonen, V.; Kazmierczak, A.; Niemela, J.; James, P. Promoting ecosystem and human health in urban areas using Green Infrastructure: A literature review. Landsc. Urban Plan. 2007, 81, 167–178. [Google Scholar] [CrossRef] [Green Version]
  3. Bullock, C.H. Valuing Urban Green Space: Hypothetical Alternatives and the sSatus Quo. J. Environ. Plan. Manag. 2008, 51, 15–35. [Google Scholar] [CrossRef]
  4. Latinopoulos, D.; Mallios, Z.; Latinopoulos, P. Valuing the benefits of an urban park project: A contingent valuation study in Thessaloniki, Greece. Land Use Policy 2016, 55, 130–141. [Google Scholar] [CrossRef]
  5. Salazar, S.S.; Menendez, L.G. Estimating the non-market benefits of an urban park: Does proximity matter? Land Use Policy 2007, 24, 296–305. [Google Scholar] [CrossRef]
  6. Feyisa, G.L.; Dons, K.; Meilby, H. Efficiency of parks in mitigating urban heat island effect: An example from Addis Ababa. Landsc. Urban Plan. 2014, 123, 87–95. [Google Scholar] [CrossRef]
  7. Ho, C.H.; Payne, L.; Orsega-Smith, E.; Godbey, G. Parks, recreation and public health. Parks Recreat. 2003, 38, 18–27. Available online: https://www.questia.com/magazine/1P3-333980511/parks-recreation-and-public-health (accessed on 28 May 2019).
  8. Wolch, J.; Byme, J.; Newell, J.P. Urban green space, public health, and environmental justice: The challenge of making cities ‘just green enough’. Landsc. Urban Plan. 2014, 125, 234–244. [Google Scholar] [CrossRef]
  9. Bratman, G.N.; Daily, G.C.; Levy, B.J.; Gross, J.J. The benefits of nature experience: Improved affect and cognition. Landsc. Urban Plan. 2015, 138, 41–50. [Google Scholar] [CrossRef]
  10. National Landscape Malaysia. Project Outcome Evaluation Report of New Public Parks: Taman Tasik Cempaka, Bangi Selangor. Unpublished report. 2008. [Google Scholar]
  11. Jamal, O.; Rahajeng, A. Economic Valuation of Jogjakarta’s Tourism Attributes: A Contingent Ranking Analysis. Tour. Econ. 2013, 19, 187–201. [Google Scholar] [CrossRef]
  12. Alberini, A.; Longo, A. Combining the travel cost and contingent behavior methods to value cultural heritage sites: evidence from Armenia. J. Cult. Econ. 2006, 30, 287–304. [Google Scholar] [CrossRef]
  13. Gürlük, S.; Rehber, E. A travel cost study to estimate recreational value for a bird refuge at Lake Manyas, Turkey. J. Environ. Manag. 2008, 88, 1350–1360. [Google Scholar] [CrossRef] [PubMed]
  14. Pérez-Álvarez, R.; Torres-Ortega, S.; Díaz-Simal, P.; Husillos-Rodríguez, R.; de Luis-Ruiz, J.M. Economic Valuation of Mining Heritage from a Recreational Approach: Application to the Case of El Soplao Cave in Spain (Geosite UR004). Sustainability 2016, 8, 185. [Google Scholar] [CrossRef]
  15. Grilli, G.; Landgraf, G.; Curtis, J.; Hynes, S. A travel evaluation of the benefits of two destination salmon rivers in Ireland. JORT 2018, 23, 1–7. [Google Scholar] [CrossRef]
  16. Torres-Ortega, S.; Pérez-Álvarez, R.; Pedro Díaz-Simal, P.; De Luis-Ruiz, J.M.; Felipe Piña-García, F. Economic Valuation of Cultural Heritage: Application of Travel Cost Method to the National Museum and Research Center of Altamira. Sustainability 2018, 10, 2550. [Google Scholar] [CrossRef]
  17. Bertram, C.; Larondelle, N. Going to the Woods Is Going Home: Recreational Benefits of a Larger Urban Forest Site—A Travel Cost Analysis for Berlin, Germany. Ecol. Econ. 2017, 132, 255–263. [Google Scholar] [CrossRef]
  18. Creel, M.D.; Loomis, J.B. Theoretical and Empirical Advantages of Truncated Count Data Estimators for Analysis of Deer Hunting in California. Am. J. Agric. Econ. 1990, 72, 434–441. [Google Scholar] [CrossRef]
  19. Herriges, J.A.; Kling, C.L. Valuing Recreation and the Environment: Revealed Preference Methods in Theory and Practice; Edward Elgar: Cheltenham, UK, 1999. [Google Scholar]
  20. Hellerstein, D.; Mendelsohn, R. A Theoretical Foundation for Count Data Models. Am. J. Agric. Econ. 1993, 75, 604–611. [Google Scholar] [CrossRef] [Green Version]
  21. Cameron, A.C.; Trivedi, P.K. Regression Analysis of Count Data; Cambridge University Press: Cambridge, UK, 2013; Volume 53. [Google Scholar]
  22. Englin, J.; Shonkwiler, J.S. Estimating Social Welfare Using Count Data Models: An Application to Long-Run Recreation Demand under Conditions of Endogenous Stratification and Truncation. Rev. Econ. Stat. 1995, 77, 104–112. [Google Scholar] [CrossRef]
  23. Martínez-Espiñeira, R.; Amoako-Tuffour, J. Recreation demand analysis under truncation, overdispersion, andendogenous stratification: An application to Gros Morne National Park. J. Environ. Manage. 2008, 88, 1320–1332. [Google Scholar] [CrossRef] [PubMed]
  24. Highway Planning Unit. Ministry of Works Malaysia. Available online: http://www.kkr.gov.my/en/node/10746 (accessed on 28 May 2019).
  25. Curtis, J.; Stanley, B. Water quality and recreational angling demand in Ireland. JORT 2016, 14, 27–34. [Google Scholar] [CrossRef] [Green Version]
  26. Hynes, S.; Greene, W. A Panel Travel Cost Model Accounting for Endogenous Stratification and Truncation: A Latentclass Approach. Land Econ. 2013, 89, 177–192. [Google Scholar] [CrossRef]
  27. Jamal, O.; Basri, A.T.; Redzuan, O. Recreational values of Malaysia agricultural park, Shah Alam (in Malay). Int. J. Manag. Stud. 2004, 11, 165–179. [Google Scholar]
  28. Kaffashi, K.; Radam, A.; Shamsudin, M.N.; Rusli Yacob, M.R.; Nordin, N.H. Ecological Conservation, Ecotourism, and Sustainable Management: The Case of Penang National Park. Forests 2015, 6, 2345–2370. [Google Scholar] [CrossRef] [Green Version]
Figure 1. Willingness to pay, travel costs and recreational benefits against on-site time per trip, ceteris paribus.
Figure 1. Willingness to pay, travel costs and recreational benefits against on-site time per trip, ceteris paribus.
Sustainability 11 03023 g001
Table 1. Socio-economic profile of respondents (n = 480).
Table 1. Socio-economic profile of respondents (n = 480).
Socio-Economic CharacteristicsVariableShare(%)
Area of residenceBangi area
Others
64
36
SexMale62
Female38
Age 29 (number of years)
StatusSingle64
Married33
Widow3
EducationSchool certificates and college diplomas68
Undergraduate and above32
OccupationExecutive
Non-executive
24
76
VehicleOwn vehicle (car and van)57
Motorcycle12
Bicycle3
Walk9
Public transport19
Is lake visitation the main aim?Yes
No
85
16
Prime motivation of visitsRecreation/Leisure
Other reasons
41
59
Sports as main motivation?Yes
No
21
79
Visitation with family and friends?Yes
No
95
5
Table 2. Expenditures on food and petrol to the park.
Table 2. Expenditures on food and petrol to the park.
Expenditure ItemsMean (MYR)Standard Deviation
Food and beverages5.287.76
Petrol66.72
Toll0.522.81
Table 3. Descriptive statistics of variables per respondent (n = 482).
Table 3. Descriptive statistics of variables per respondent (n = 482).
VariableMeanMinimumMaximumStandard Deviation
Distance (one way distance, km)8.30.5275.9
Number of trips (per month)3.021203.67
Time spent on-site per visit (hours)
Time spent on-site per month (hours)
Travel cost per hour (MYR)
0.99
3.83
25
0.17
1
4
3.5
5
20
0.6
4.14
22
Travel cost per trip (MYR)222.912821
Income (MYR/Family/Month)33001000160003837
Group Size3.33171.46
Table 4. Definition of variables, measurements and mean values.
Table 4. Definition of variables, measurements and mean values.
VariableVariables DefinitionMeasureMean
T C i Travel cost per trip (MYR)The per trip cost includes distance and opportunity cost of time for each respondent. Time cost is round trip travel time (in hours) multiplied by a third of hourly wage rate where the wage rate was equal to daily income over daily working hours (8 h). The distance cost was calculated by multiplying the round-trip distance by variable cost component per kilometer. The cost per kilometer rates (0.7 for cars/vans) was adopted based on the rates by the Highway Planning Unit, Ministry of Works, Malaysia [24]. Discounts of 60% and 90% were accorded to the use of motorcycles and bicycles, respectively.22
V i Number of trips per respondent per monthThe average number of trips each respondent frequented every month3.02
D_Agei > 30Dummy for ageNominal data, 1 for age 30 and above, 0 for others0.06
D _ E d u i Dummy for educationNominal data, 1 for an undergraduate degree and above, 0 for others0.57
D _ B a n g i i Dummy for Bangi residenceNominal data, 1 for the residence of Bangi, 0 otherwise0.64
D _ O c c u p a t i o n i Dummy for occupation categoryNominal data, 1 for a professional occupation, 0 otherwise0.24
D _ F r i e n d s i Dummy for visitation with friends/familyNominal data, 1 for visitation with family or friends, 0 for all others 0.95
D_LeisureiDummy for leisure motiveNominal data, 1 for visitation derived by leisure/recreation, 0 for all others0.56
D_GenDummy for genderNominal data, 1 for male respondents, 0 for females0.61
Table 5. Results of count data models corrected for truncation endogenous stratification.
Table 5. Results of count data models corrected for truncation endogenous stratification.
Dependent Variable: Number Of Visits To Site Per MonthPoisson (PR)Negative Binomial (NB)
Constant1.965
(0.2388 ***)
1.942
(1.4866 ***)
D_Age > 300.532
(0.142 ***)
0.985
(0.193 ***)
D_Bangi−0.171
(0.151)
0.135
(0.207)
D_Occupation0.091
(0.135)
−0.215
(0.699)
D_Edu0.014
(0.123)
0.112
(0.792)
D_Friends−0.768
(0.213 ***)
−0.971
(0.380 ***)
D_Leisure0.605
(0.249 ***)
0.513
(0.319 ***)
D_Gender0.029
(0.157)
0.176
(0.663)
Travel Cost Per Trip−0.051
(0.011 ***)
−0.048
(0.015 ***)
Ln(a)/−0.450 *
(0.261)
AIC301.8279.5
BIC338.9318.3
Log-Likelihood−187.5−193.6
Note: Standard errors in parenthesis ( *** p < 0.01, * p < 0.10). Source: Authors estimation.
Table 6. Consumer surplus and own price elasticities.
Table 6. Consumer surplus and own price elasticities.
VariablePoissonNegative Binomial
Consumer Surplus Per Hour Per Individual (MYR)5.96.2
Consumer Surplus Per Trip Per Individual (MYR)5.86.0
Own-price elasticity of demand−0.51−0.48
Source: Authors estimation.
Table 7. Estimated consumer surplus, travel cost and willingness to pay per individual visitor.
Table 7. Estimated consumer surplus, travel cost and willingness to pay per individual visitor.
Time on Site (hours)Predicted CS (MYR)Marginal CS (MYR)Travel Cost (MYR)Marginal Travel Cost (MYR)WTP (MYR)MWTP (MYR)
0.052.25-4.41-6.66-
0.223.070.825.150.758.231.57
0.384.091.016.000.8510.091.86
0.555.311.226.950.9512.262.17
0.726.741.438.011.0614.742.48
0.898.351.619.171.1617.522.78
1.0510.101.7610.451.2820.553.03
1.2211.951.8411.841.3923.783.23
1.3913.791.8513.341.5027.133.35
1.5515.551.7614.941.6130.503.37
1.7217.131.5816.651.7133.783.29
1.8918.431.3018.451.8036.883.10
2.0519.360.9320.341.8839.702.82
2.2219.870.5122.291.9542.162.46
2.3919.910.0424.292.0144.212.05
2.5619.49−0.4226.332.0445.831.62
2.7218.64−0.8628.392.0547.031.20
2.8917.40−1.2330.442.0547.840.81
3.0615.87−1.5332.452.0148.320.48
3.2214.14−1.7334.411.9648.540.22
3.3912.30−1.8436.281.8848.580.04

Share and Cite

MDPI and ACS Style

Othman, J.; Jafari, Y. Economic Valuation of an Urban Lake Recreational Park: Case of Taman Tasik Cempaka in Bandar Baru Bangi, Malaysia. Sustainability 2019, 11, 3023. https://doi.org/10.3390/su11113023

AMA Style

Othman J, Jafari Y. Economic Valuation of an Urban Lake Recreational Park: Case of Taman Tasik Cempaka in Bandar Baru Bangi, Malaysia. Sustainability. 2019; 11(11):3023. https://doi.org/10.3390/su11113023

Chicago/Turabian Style

Othman, Jamal, and Yaghoob Jafari. 2019. "Economic Valuation of an Urban Lake Recreational Park: Case of Taman Tasik Cempaka in Bandar Baru Bangi, Malaysia" Sustainability 11, no. 11: 3023. https://doi.org/10.3390/su11113023

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop