Recreation Demand Analysis Research Articles

Correcting On-Site Sampling Bias: A New Method with Application to Recreation Demand Analysis

Collecting data via on-site surveys is convenient and can be cost-effective. However, the on-site sampling scheme over-samples frequent site visitors and omits nonvisitors, which can result in biased and inconsistent estimation of population parameters. A common empirical approach to addressing the sampling issues is to make adjustments directly to the assumed population distribution. We propose an alternative empirical strategy that utilizes the sample distribution and treats endogenous stratification and truncation separately. Monte Carlo simulation shows this proposed empirical strategy has merit. A case study of recreation demand for coastal beaches using on-site survey data is presented.

Handling Excess Zeros in Count Models for Recreation Demand Analysis Without Apology

Hurdle and zero-inflated models are the two foremost methods used to deal with excess zeros. However, their reliance on the nonparticipation assumption is a drawback when applied to recreation demand analysis because there is not a theoretical framework convincingly explaining presence of non-participants. This paper discusses how latent class count models represent a theoretically consistent alternative to handle excess zeros. The theoretical model behind a latent class model does not require the non-participation assumption. Instead, excess zeros is explained as the accumulation of Corner solutions from individuals belonging to different classes.

Recreation demand analysis of sensitive natural areas from an on-site survey

Les espaces naturels sont des sites multifonctionnels contribuant au bien-être humain. Parmi les services écosystémiques qu'offrent ces espaces, les services récréatifs occupent une place grandissante. L'objectif de cet article est d'évaluer la demande de récréation des Espaces Naturels Sensibles ( Ens ) du Département de l'Hérault. Ces sites sont acquis par les Départements pour les préserver de l'urbanisation et les rendre accessibles gratuitement. En estimant les bénéfices récréatifs associés, nous fournissons des informations justifiant la politique de conservation des Ens . Nous utilisons l'approche individuelle de la Méthode des Coûts de Déplacements ( Mcd ). Nous estimons le surplus des visiteurs à hauteur de 59 € (78 US$) en moyenne par visite. Cette valeur est plus importante que celles estimées dans les différentes études portant sur la récréation en France. Cette valeur permet en partie de justifier les coûts engendrés par la politique Ens (acquisition et gestion) par rapport à d'autres sites naturels. En s'appuyant sur deux hypothèses d'affluence de visiteurs, la valeur récréative des Ens de l'Hérault serait de l'ordre de 12,5 millions d'euros par an (16,5 millions US$) pour l'hypothèse basse et 25 millions d'euros par an (33 millions US$) pour l'hypothèse haute.

Estimating Endangered Species Interaction Risk With the Kalman Filter

AbstractTo address the tradeoff between biodiversity conservation in marine ecosystems and fishing opportunity, it is important to quantify the risk of endangered species interactions in commercial fisheries. We propose a Kalman filter suitable for rare events to estimate the endangered leatherback turtle take risk in the California drift gillnet fishery in the years 1990–2010, conditional on spatiotemporal factors that affect take rates. Results suggest interaction risk has remained stable, but with substantial variation over the spatiotemporal distribution of effort. Our methods might also apply to recreation demand analysis with rare event risk, or to applications involving irregularly spaced observations, like trade‐level stock market data.

Using Count Data and Ordered Models in National Forest Recreation Demand Analysis

This research addresses the need to improve our knowledge on the demand for national forests for recreation and offers an in-depth data analysis supported by the complementary use of count data and ordered models. From a policy-making perspective, while count data models enable the estimation of monetary welfare measures, ordered models allow for the wider use of the database and provide a more flexible analysis of data. The main purpose of this article is to analyse the individual forest recreation demand and to derive a measure of its current use value. To allow a more complete analysis of the forest recreation demand structure the econometric approach supplements the use of count data models with ordered category models using data obtained by means of an on-site survey in the Bussaco National Forest (Portugal). Overall, both models reveal that travel cost and substitute prices are important explanatory variables, visits are a normal good and demographic variables seem to have no influence on demand. In particular, estimated price and income elasticities of demand are quite low. Accordingly, it is possible to argue that travel cost (price) in isolation may be expected to have a low impact on visitation levels.

Leisure and the Net Opportunity Cost of Travel Time in Recreation Demand Analysis: An Application to Gros Morne National Park

Using count data models that account for zero-truncation, overdispersion, and endogenous stratification, we estimate the value of access to recreational parks. The focus is on the empirical estimation of the proportion of the wage rate that best approximates park visitors' opportunity cost of travel time within the cost of their trip and its effects on estimated consumer surplus. The fraction of hourly earnings that corresponds to the opportunity cost of travel time is endogenously estimated as a function of visitor characteristics, rather than fixed exogenously. In this case, which deals with a relatively remote recreational site, the relevant opportunity cost of time for most visitors appears to represent a smaller fraction of their wage rate than commonly assumed in previous similar studies.

Microeconometric Strategies for Dealing with Unobservables and Endogenous Variables in Recreation Demand Models

The past decade has witnessed significant advances in the microeconometric analysis of recreation data. In this review, we focus on two areas in which these innovations have been especially prolific: accounting for unobserved preference heterogeneity and controlling for unobserved and possibly endogenous site characteristics, such as congestion. Failure to appropriately address these issues with the nonlinear models typically used in recreation demand analysis can severely bias parameter and welfare estimates. We consider these issues of widespread importance within and beyond recreation demand applications. We also expect these estimation challenges to become more ubiquitous as the field gradually moves toward region-wide, multisite applications in reaction to large-scale environmental changes.

Recreation demand analysis under truncation, overdispersion, and endogenous stratification: An application to Gros Morne National Park

Using on-site survey data from Gros Morne National Park in Newfoundland, this paper estimates and compares several truncated count data models of recreation demand. The selected model accounts not only for the truncated and overdispersed nature of the data but also for endogenous stratification due to the oversampling of avid users. Moreover, the model that allows for a flexible specification of the overdispersion parameter dominates on the basis of goodness of fit. The results are used to estimate the users’ value of access to the park.

The Fast Decay Process in Outdoor Recreational Activities and the Use of Alternative Count Data Models

Despite a number of significant advances in count data modeling during the last two decades and the growing popularity of these models in recreation demand analysis, standard count data models are inadequate to address the fast decay process of the dependent variable and the associated long tail. This article demonstrates how one and two‐parameter alternative count data models can be used to properly model the fast decay process and the associated long tail commonly observed in recreation demand analysis. Econometric results from an illustrative application suggest satisfactory performance of four of the eight alternative count data models proposed in this article.

Recreation Demand Choices and Revealed Values of Leisure Time

An approach to jointly estimating an endogenous marginal value of time function and a recreation budget share equation is developed. The specification requirements for the marginal value of time function, to ensure consistency with the maintained hypothesis of two binding constraints on choice, are articulated. The estimating model consistent with these requirements that nests several conventional treatments of the marginal value of time is highly significant. The general endogenous marginal value of time model outperforms other approaches and shows a more complex relationship between the marginal value of time and the individual's wage than has been used in previous work. (JEL J22, Q26)

Size and Bag Limits in Recreational Fisheries: Theoretical and Empirical Analysis

Size and bag limits are among the most common forms of regulations for recreational fisheries. In this paper, we theoretically study and compare the short- and long-term impacts of these policies on individual anglers and fish stocks. Particular attention is paid to the issue of release mortality, which can have important consequences for policy effectiveness. Theoretically, we show the conditions under which these policies will be successful in achieving biological objectives. Implications for recreation demand analysis are discussed. We also study these policies using a simulation model of various policy combinations for the Gulf of Mexico red snapper fishery.

Influence of On-Site Choices on Recreation Demand

The role of on-site trips in recreation demand modeling is examined by means of the travel cost method. An understanding of on-site choices of site attractions and costs incurred by visitors is important if analysts are to estimate the popularity of the various attractions and, ultimately, to derive site benefits. We investigate two primary questions. First, can on-site trips to attractions be adequately modeled? Next, how important is onsite travel in valuing site benefits? Visitor survey data for the North Carolina Zoological Park were used to determine that economic site benefits are underestimated by 24% if expected on-site travel costs from recreation demand analysis are excluded.

An Estimable Dynamic Model of Recreation Behavior with an Application to Great Lakes Angling

This paper develops an estimable model of recreation behavior in which the recreation decision is cast as a dynamic programming problem. The model is illustrated with an application to salmon and trout fishing on Lake Michigan. The application considers various factors affecting the trip decision, including the expected catch, the weather, the opportunity to participate in a fishing derby, and the time elapsed since the last fishing trip. Catch is modeled as a Poisson-distributed random variable. Estimation results are promising, but a number of practical obstacles must be surmounted for the model to be regularly applied in recreation demand analysis.

Count Data Models and the Problem of Zeros in Recreation Demand Analysis

AbstractIn this paper we develop a count data model for consumer demand which explicitly allows for a large number of zero observations for the dependent variable, and separation of the participation versus quantity decisions. The advantages of the model over traditional censored and count demand models are brought out, and the appropriate consumer surplus measures are derived. By introducing a random error term into the traditional count model demand function, the appropriate measure of expected consumer surplus for count models is derived. The model is illustrated using a recreational survey of the general population.

Estimating and Comparing Demand Functions for Personal Use Christmas Tree Cutting at Seven Utah Sites

Abstract The travel cost model of recreation demand analysis was applied to seven USDA Forest Service Ranger Districts in Utah. The objectives were to (1) estimate consumers' surplus and total willingness to pay values for the recreational component of personal use Christmas tree gathering, and (2) compare these Utah results with each other and with the Markstrom and Donnelly (1988) results from Colorado, the only other travel cost analysis of Christmas tree gathering from public lands. The results were that per-trip consumer' s surplus estimates ranged from $6.71 to $31.17, compared to Markstrom and Donnelly's (1988) estimate of $10.05. There is sufficient intersite variation that the demand functions from any site cannot be readily applied to any other. West. J. Appl. For. 6(2):42-46.

The Effect of Common Sources of Regression Error on Benefit Estimates

Discussions of the sources of error in recreation demand analysis are common in the existing literature. The most traditional line of thought (e.g., Gum and Martin) considers the error component in predicting the individual's recreation behavior to arise from unmeasured socioeconomic factors. Others (e.g., Hanemann) attribute at least some of this error to fundamental randomness in human behavior. Applied statisticians (e.g., Hiett and Worrall) on the other hand, suggest that recall of annual number of recreational trips (i.e., the quantity demanded) is subject to substantial error. Still others (e.g., Brown, et al.) have argued that recall of explanatory variables, such as travel expenses, contains error. Despite the discussion of potential error sources, there is no work which relates the sources of error in the demand relationship with benefit measurement.1 Of practical importance in recreational demand estimation is whether error assumptions have an influence on welfare measures, since it is for public policy decision purposes that recreational demand estimation is usually undertaken. Will the estimated consumer surplus derived from demand analysis be independent of the assumed source of the error in the regression? The purpose of this paper is to explore how the assumptions about sources of error influence estimates of recreation benefits. A discussion of the various sources of error in de-

Alternative Ways to Measure Recreation Values by the Travel Cost Method

The travel cost method is widely used for estimating the economic value of outdoor recreation services. Much early research was empirical applications of the Hotelling-Clawson model (Brown, Singh, Castle; Clawson; Clawson and Knetsch; Trice and Wood). Recently, model specification and functional form problems have attracted greater attention (Allen, Stevens, Barrett; Burt and Brewer; Cesario and Knetsch; Gum and Martin; Smith 1975b; Ziemer, Musser, Hill). A major focus of attention is estimating the recreation participation equation (often called step one of the travel cost method). It measures participation in an activity based on travel costs and other explanatory variables. However, misspecification of this equation is only one possible source of inaccuracy. The procedure for estimating consumer surplus once the participation equation has been derived also can affect the values derived from recreation demand analysis. This paper considers how alternative ways to estimate economic value from participation equations can affect travel cost study results. Empirical study of the economic value of the St. Lawrence River-eastern Lake Ontario bass fishery in New York illustrates the point. The underlying theory of the travel cost method comes from neoclassical demand theory. Demand equations are derived for an outdoor recreation activity by observing how individuals respond to various costs of travel to use the resource. The net economic value (NEV) of the resource in its current use is the amount that users would be willing to pay in excess of their actual expenditures. This is the sum of the users' consumer surplus (Knetsch, Smith 1975a). Assuming that any net user benefits are associated with the recreation site itself rather than the trip, this value represents the net economic value of the recreation resource (Brown, Singh, Castle; Dwyer, Kelly, Bowes). Empirical applications of the travel cost method use visitation data from points of origin at various distances from the site. The traditional approach uses distance zones as units of observation. The quantity variable (visitation rate) is the expected participation rate for a representative individual from a particular origin zone.' Observations of individuals rather than zone averages generally will lead to efficiency gains in estimating the travel cost coefficient and other parameters (Brown and Nawas). The second step is to derive the implied demand for and economic value of the resource from the participation equation. Several methods have been used. One calculates the site's economic value as the area under an aggregate demand curve for the site (Cesario and Knetsch; Grubb and Goodwin; Knetsch, Brown, Hansen). Another method uses the participation equation to derive a demand curve and economic value for the site from each origin. Then aggregation across all origins measures the site's total economic value (McConnell and Norton). This study compares these methods for estimating recreation demand curves and economic value.

Measuring the Cost of Time in Recreation Demand Analysis: Comment

In the February 1981 issue of the Journal, McConnell and Strand present a model of recreation demand in which the opportunity cost of an hour of travel to a recreation site is assumed not to equal the average hourly income of the recreating family. They test the model with data from a 1978 survey of sportfishermen in the Chesapeake Bay region. Simply stated, the model hypothesizes that the sportfishermen will behave as if the price of a recreational trip is the sum of cash costs and the opportunity cost of travel time. This opportunity cost is hypothesized to be some ratio, k, of average family income per hour times the travel time. McConnell and Strand estimate the value of k by regressing recreation days against cash costs and opportunity cost of time as separate variables. The ratio of regression coefficients then is an estimate of k. They go on to estimate the ratio, k, of the opportunity cost of travel to the average income for sportfishermen in the Chesapeake Bay region. The authors hypothesize that the ratio (k) falls between 0 and 1, implying that sportfishermen value the time in transit lower than their average income when working. Sufficient conditions for k to equal one include (a) the tax rate is zero, (b) marginal earnings are constant, and (c) nonwork income equals zero. This comment is written to point out certain aspects of the study which may have led to an underestimation of the value of k. It is argued that McConnell and Strand have misspecified their demand equation in such a way that their estimate for k (.612) is biased downward. McConnell and Strand hypothesize the demand relation

Measuring the Cost of Time in Recreation Demand Analysis: Comment

Measuring the Cost of Time in Recreation Demand Analysis: Comment

Measuring the Cost of Time in Recreation Demand Analysis: Reply

We appreciate the comments of Johnson and Ward on our note measuring the value of time. We would like to emphasize that the purpose of our note was to offer an empirical approach to measuring value of time which permitted the value to vary across individuals but which inferred this variation from individuals in the sample. We realize that empirical approaches to measuring the value of time are currently evolving. The comments by Johnson and Ward help elucidate some basic issues in valuing time. We respond first to Johnson and then to Ward.