Linking Tourist Willingness to Pay and Beach Management: A Travel Cost Analysis for Balandra Marine Park, Mexico

Abstract

Balandra, one of the most popular beaches in La Paz, Baja California Sur, was declared a flora and fauna protection area in 2012, and in 2019, the Mexican government chose it as the best beach in Mexico during the Tianguis Turistico celebrated on that year. Because of this latter distinction, this beach currently faces overcrowding. Formulating effective management policies depends, to a certain extent, on the knowledge of their recreational value and visitor characteristics. Recreational value allows us to know the benefits of the tradeoffs among the ecosystem services and society and exhibit the value of possible damages to marine ecosystems, like the one caused in 2022 by the fire of a tourist boat inside Balandra. Using the individual travel cost method and applying 159 questionnaires to site visitors, the individual willingness to pay to access Balandra Beach was estimated, resulting in USD 11.11/day/visitor. Recreational economic value (REV) for Balandra was estimated using two essential criteria: first, the total visitors registered in 2021, and second, the daily maximum carrying capacity. Welfare recreational loss was also calculated, derived from the site’s two-month closure, using REV as a proxy. Finally, beach management options and possible environmental and economic policy instruments that could be implemented are discussed.

1. Introduction

1.1. Marine Protected Areas and Tourism: A General Overview

Marine protected areas (MPAs) are a recurrent mechanism to safeguard marine ecosystems and biodiversity [1]. MPAs provide benefits derived from their conservation, such as recreational benefits provided to beach swimmers in coastal zones. To ensure that these benefits are considered and endured, MPA management requires continuous surveillance and monitoring. For example, in several developed European countries, beach access within protected areas is entirely free, and site management and maintenance are financed by local, regional, or national authorities depending on natural protected area categories [2]. Conversely, many sites in emerging economies, like Mexico, lack the federal, state, or local resources needed to guarantee effective biodiversity conservation [3]. Mexico’s MPA budget relies on collecting financial resources derived from a single access fee of USD 5.85 set by Mexican regulation.

In several countries, the coastal and beach tourism industry is a source of significant economic benefits. Natural diversity contributes to a massive demand for touristic services, mainly in MPAs, since using mega-diversity for recreational purposes has generated significant income levels in coastal communities. Nevertheless, there has been an accelerated increase in tourism and the demand for ecosystem services in coastal communities, even small ones, which depend on tourism as a primary or complementary means of livelihood [4].

These zones are highly linked to marine resources and attractions, such as crystal waters and environments suitable for practicing touristic sports and activities. Because of this, coastal zones are relatively relevant due to their social, ecological, cultural, and economic value. Emerging economies seek to exploit their tourism potential to catapult economic development without affecting them [5].

Coastal and marine environments attract hundreds of millions of tourists annually; tourism is a pillar of local economies in some regions [6]. Nature tourism contributes almost 7% of world tourism expenditure, and wildlife tourism contributes USD 343 billion globally [7]. Many tourism activities focus on the “3S market” (sea, sand, and sun), which plays an essential role as a determining factor in tourists’ motivation to visit a destination based on marine resources. Tourism depends on the integrity of coastal resources, i.e., beaches and water. Despite the economic importance of tourism, coastal resources have become scarcer and threatened by it. Several beach destinations have taken measures to deal with them, like long seashore structures or dredging. Such measures regularly bring non-accepted consequences, locally or elsewhere [8]. External pressures on coastal zones linked to tourism, such as land transformation and the construction of industrial developments, water pollution, mangrove depletion, invasive species introduction, and exhaustive use of resources (for example, marine species used as meals and souvenirs), plus climate change, affect tourism viability and tourism coastal destinations [6].

1.2. Beach Management Aspects

Using the 3S market’s aesthetic appeal, the coastal tourist industry applies a variety of aesthetic scales, ranking recreational areas to inform the public where the ‘best beaches’ are located [9]. Beach ratings rank beaches according to their performance against different criteria based on physical, biological, and human use attributes. The evaluation results in an ordered ranking, with the ‘best’ beach at the top [10]. However, several Internet sites publish their own list of the Best World Beaches every year using different criteria to rank these beaches, making methodologies as well as the beach ranking of each method heterogeneous and incomparable [11]. So, there is a high probability that one beach does not appear in more than one rank. Recently, the Beach Ranking Framework has been proposed to homogenize beach ranking criteria; it considers four domains with twenty categories, five per domain [12].

Beach management (BM) seeks to maintain or improve a beach as a recreational resource and a means of coast protection while providing facilities that meet the needs and aspirations of those who use the beach [13]. Several BM schemes focus on the beach rather than the hinterland’s bathing area. Currently, there are different beach classifications: (i) dissipative–reflective, (ii) heavy–light usage, (iii) resort, (iv) urban, (v) village, (vi) rural, (vii) remote, and (viii) urban water [14]. Ecosystem functions and ecological processes are so different among these classifications that different beach types need different management schemes. The essential elements for BM are clean, shallow water and a sandy beach, no dangerous animals/odors, a pleasant water temperature, security, and sound infrastructure/services (toilets, access, lifeguards, shade, and a small shopping precinct) [15].

BM focuses mainly on five ramifications: (i) developing tools for integrated coastal management and identifying management priorities, (ii) geomorphological analysis of beach dynamics in the context of erosion events, (iii) critical factors affecting the quality of beaches, wrack removal, and impact mitigation from urban pressures on sandy beaches, (iv) ecological engineering (ecosystem engineering traits of the dune-building grass), and (v) certification processes and considered tools for better control and sustainable beach development and management (e.g., Blue Flag) [16].

Quality is a critical attribute to consider when managing a beach. It could be understood as the uses established for this coastal system, the most common being tourism, fishing, and conservation. This is especially true concerning tourism, where quality is frequently measured in two major areas: environmental and recreational. The beach system especially loses its functionality as a place of leisure and recreation when its natural elements represent a risk to the health of beach users, mainly in environmental and sanitary conditions impacting recreational quality. Recently, the criteria for measuring beach environmental quality parameters were updated to inform authorities with a more effective tool [17,18].

Beaches are seen as an amenity by promotors and tourists; the most critical aspect of understanding amenity is to realize that the amenity of a beach depends on the perception of beach users, meaning that how a beach is managed for amenity purposes is driven by beach users; therefore, it is essential to understand (i) who uses the beach and for what purposes, (ii) how often different users use the beach, (iii) users’ perceptions of the beach and how it is managed, and (iv) what services users need/desire. To understand these elements, it is essential to recognize that the perceptions of beach users vary from beach to beach and country to country [19].

It is also essential to understand where beach users come from and their perceptions, particularly for coastal and tourism managers who wish to provide a beach resource that tourists enjoy, as this benefits an area’s economy [14]. To better understand beach users’ perceptions of the amenity aspects of a location, questionnaires and surveys can be undertaken that provide quantitative information on the spatial and temporal variations in the patterns and nature of beach use. Good perceptions of and attitudes toward the recreational site are essential for tourist demand, helping us to understand visitor behavior and beach use [20]. Sustainable beach use must consider users’ socioeconomic characteristics (preferences, priorities, and willingness to pay) and cultural resources to increase recreation and foster conservation [21]. Conserving and enhancing ecosystem services is a matter of managing human activities and actions within marine environments rather than the physical ecosystem [22].

The need to maintain attractive beaches for tourists has frequently led local managers to clean them [23,24]. However, such cleaning actions usually generate environmental externalities or damage to coastal ecosystems (coastal disruption, marine and coastal biodiversity reduction, and waste disposal) [25,26]. Beach cleaning is often needed to maintain tourism affluence and activity [27,28,29]. Nevertheless, environmental issues could be accumulative.

Tourism also generates social externalities, such as access loss to public or beach facilities, pollution, beach crowding, infrastructure construction, artificial covers on sandy beaches, dunes and mangroves, waterfront and coastal line modifications, and loss of cultural and religious values [30]. In these two situations, coastal ecosystem management tools must prevail, considering the associated costs [31]. These tools must seek to minimize or eliminate externalities.

Externalities could be minimized by applying environmental valuation (EV) and its specific techniques [32]. Among them are the stated preference methods (contingent valuation) and the revealed preference methods (travel cost and hedonic prices). EV contributes to the design of economic instruments for environmental policy oriented to modified consumption or production patterns and strengthens environmental planning instruments (compensatory or regulatory) for conservation [33].

Managers and decision-makers are constantly seeking funding for MPAs; evidence-based valuation of their services is one way to demonstrate the importance of investing in effective protection and management. Environmental valuation is a tool that allows ocean and coastal ecosystems to be effectively incorporated into the national budgetary and planning process as a common language. Incorporating marine ecosystem services into fiscal, planning, and regulatory regimes requires better-quality information about how these ecosystems contribute to human well-being. The value of marine and coastal ecosystems plays a key role in policy and management approaches that enhance the sustainable development of marine environments and contribute to MPA’s sustainable financing by using market mechanisms [22,32,33].

Economic research contributes to the debate on MPAs as a management option by assessing their costs and benefits for society (

Table 1. Beach economic valuation research (USD).

Author (s)/Year	Country	Site	WTP	Method
Zambrano-Monserrate et al. (2018) [36]	Ecuador	Villamil Beach	16.95	ITC
Zhang et al. (2014) [37]	Australia, Queensland	Costa Dorada Beaches	19.97	ITC
Morales-Zarate et al. (2019) [38]	México, Baja California Sur	Los Cabos Touristic Corridor	30.96	ITC
Enriquez-Acevedo et al. (2016) [39]	Colombia	Caribbean Region	Mín. 3.80 Max. 6.80	CV
Legget et al. (2018) [40]	California, Orange County	Long Beach to Huntington Beach	Mín. 5.78 Max. 20.36	ITC
Hynes y Greene (2013) [41]	Ireland	Siverstrand	Min. 11.35 Max. 68.41	Panel ITC
Voke et al. (2013) [42]	United Kingdom	St. David’s, Pembrokeshire	Min. 2.24 Max. 37.56	ITC

Source: Self-elaboration based on mentioned references. ITC: individual travel cost, CV contingent valuation.

). A straightforward approach to understanding coastal value is the economic rent or monetary production; studies in Spain and Italy have demonstrated that a square meter of beach could produce around 700 and 12,000 EUR/m² [34]. Though some costs are often accessible to estimate, some benefits, like recreational, conservation (future use), and ecosystem functions, are rarely incorporated into the economic rent approach [35].

Valuing a sandy tourist beach lies in its landscape and attributes that bathers can appreciate directly. Still, conveying the value of less aesthetically appealing components of the coastal biome is often more complex. So, BM could become complicated because users have a limited understanding of marine and coastal ecosystem services and lack knowledge of their contribution to human well-being. Demonstrating these critical ecosystems’ economic value could reveal new economic opportunities and the need for better management [22].

Beach-user surveys can determine the economic value of beaches and the surrounding hinterland. This information is often used to demonstrate the value of continuing beach-management activities for coastal defense, as undertaking beach management for this purpose also serves to retain or improve an area’s amenity value [19].

2. Study Area

Baja California Sur (BCS) is positioned as Mexico’s fifth ranked tourism destination and the third destination that received the most tourists by air and sea [43]. BCS’s central natural heritage is constituted by high biodiversity (ecosystems and species), landscapes, and beaches, on which it is possible to perform different social and economic recreational activities [44]. La Paz’s beaches stand out for their low slope and soft white sand; beaches like Balandra, El Tesoro, Coromuel, El Caimancito, and Pichilingue are some of the favorite beach destinations of national and international tourists [45].

Balandra became a natural protected area due to its high marine and terrestrial biodiversity and was declared a natural protected area (BNPA) in 2012; it is located east of La Paz Bay (Figure 1a) and incorporates the largest wetland in the bay of La Paz, Balandra, and El Merito wetlands, which are a refuge and nursery for species of mangrove, seagrasses, fishes, birds, and sea mammals (e.g., humpback whales, orcas, California seals, and dolphins) [46]. It has eight bays with white beaches and a mushroom-shaped rock formation [47]. Balandra’s total surface area is 2512.63 ha. The land area comprises five beaches, with a total surface area of 2.070 million m² [35]. The most extensive beach is Balandra A, or the main beach (

Table 2. Balandra beaches’ surfaces.

Beach	Surface (m2)
Balandra A	599,457.00
Balandra B	523,258.00
Frente 1	144,248.00
Frente 2	243,988.00
Frente 3	559,912.00
Total	2,070,863.00

Source: Extracted from [48].

), with a total surface of 559,457 m², followed by Frente 3, Balandra B, Frente 2, and Frente 1 (Figure 1b) [48].

Tourism is the main activity in the BNPA, but there are also small-scale and recreational fisheries [46]. The effective carrying capacity calculates that the minimum area a tourist should use to avoid crowds in Balandra is 5 m² [48]. Tourism has increased since the federal and state governments began an aggressive marketing and media campaign to promote BNPAs [49]. This started because Balandra was chosen as the best beach in Mexico at the 2019 Tianguis Turistico, which is a festival for promoting Mexican tourist destinations [50]. Because of its unique and barely unmodified landscapes and clean beaches (Figure 2), it is one of the most visited beaches in BCS [49,50,51].

A sudden tourism demand increase has generated some issues in the BNPA: (i) severe over-crowding (the BNPA’s effective carrying capacity (CC) is 350 persons/day in the main beach (Balandra A), and the full CC for the area is 984 persons/day [48]) and, (ii) due to overcrowding, current visitor management strategies consist of two daily visiting shifts, with a maximum of 400 people on Balandra’s main beach, trying to accomplish 800 visitors/day in the BNPA main beach, exceeding its CC by 2.28 times [52]. As a result, long waiting lines to access the BNPA occur during the high season.

The Management Director of the BNPA [53] indicates that the site’s high season of visitors is spring, summer, and winter vacations when the two-shift strategy is regularly implemented. In these three seasons, around 68% of registered visits take place.

On the night of 21 August 2022, the BNPA faced a maritime accident inside the marine zone; a tourist boat caught fire when passing by this zone, which is restricted to naval navigation. This led to marine and coastal pollution by the fuel and oil spill in the influential area where it took place, and the boat’s physically burned waste was left on the beaches. Because of this accident, the BNPA remained closed for over two months that year [54,55]. The estimated cost derived from boat waste removal and beach cleaning was estimated at USD 36,070 by BNPA managers after the cleaning process finished; this amount regards only the main beach, Beach A; some boat parts still remain underwater along the site’s polygon [53]. These issues lead to the following question: Could willingness to pay to access the BNPA contribute to beach management decision-making?

3. Materials and Methods

3.1. Sampling

Managers from the BNPA reported 41,259 visitors in the 2020–2021 season [53]. Applying proportional unrestricted random sampling for each month of the season, the sample for this research was estimated using parameter values p = 0.5 and q = 0.5, with an estimation error (i) of 7.7% and 95% confidence. This resulted in 161 questionaries being applied face-to-face to visitors from April 2021 to March 2022. The response rate was approximately 89%.

3.2. Questionnaire

Due to COVID-19 pandemic restrictions, a Google Forms online questionnaire was used to collect visitor information. The questionnaire is divided into three sections. The first is “Travel”, where visitors are asked about the visit motivation, transportation means, travel, and discretional costs (2021 USD). The second, “Site”, includes questions about visits to the BNPA and the site’s environmental conditions. Third, “Visitor characteristics”, include precedence, distance, and socioeconomic and sociodemographic aspects.

3.3. Travel Cost Method

From an economic perspective, recreational services provided by natural resources (lakes, rivers, estuaries, beaches, forests, and others) have essential attributes and characteristics. These are fundamental to determining the economic value of recreational services. The market system does not assign access to natural resources that offer recreational alternatives. This means that natural resources providing recreational services are public goods. A prevalent methodology to assign their economic value is the travel cost method (TC); this was proposed to the National Park Service of the United States of North America to establish access fees [56]. Through its development over the decades, the ITC method has been preferred over the other travel cost approaches for valuing recreational sites [57].

The ITC’s central assumption is that a visitor must incur travel and other costs associated with visiting a single site to enjoy its recreational services. It seeks to estimate how the demand for environmental assets (quantity or quality) varies, considering how the number of visits will change if the travel cost or access fees to the site also change [57,58]. Tourists visiting a single recreational site have a determined budget, which they compare with actual prices before deciding. Considering the different preferences and income levels, there will be individuals whose willingness to pay (WTP) for accessing the site is high and others whose WTP to access the site is lower. This condition originates a demand curve, which relates travel costs and expenses to the site with the number of visits a tourist makes [59]. The value of a particular site’s recreational service is represented by the area under the demand curve, which aggregates all site visitors and is frequently known in environmental economics as WTP [57,59]. Lastly, ITC estimations present high efficiency for the demand function of recreational services [60].

3.4. Poisson Models for Recreation Demand

Individuals frequently pay just a few visits to a recreational site, with one or two trips as the maximum. Most ITC models are estimated using discrete distributions. Since the number of trips is a non-negative discrete variable, it is the dependent variable. Under this approach, discrete density functions such as the Poisson distribution have been used. The most relevant feature of Poisson models is that they assume equality between the distribution’s mean and variance [57,60].

Recreational demand could be estimated using ITC; the method shows the individual’s willingness to accept or behave in favor of site improvements or how the individual values potential site damage. If such changes happen, such willingness is measured through the number of trips and the associated costs to access the recreational site (and other socioenvironmental variables) [60]. The general ITC outline is

X_ij = f (C_ij,Z_ij,ε_ij)

(1)

where X_ij is the individual number of visits to the recreational site in one year, C_ij is the personal travel cost, Z_ij is a vector of socioeconomic and environmental variables related to the individual and the site, and ε_ij is the stochastic term.

The individual demand model allocates time and income constraints, providing a generic demand function for a single site, assuming that an individual, i, chose x_ij, where j is the number of trips/visits to the site ∀ j = 1, 2,…n. Round TC is c_ij. The individual also consumes a set of associated goods, z_i, also known as weak complementarity goods. TC assumes that the visitor is subject to two constraints [61]:

Income:

$${\mathit{y}}_{\mathit{i}}·({\sum }_{\mathit{j}=1}^{\mathit{n}}{\mathit{x}}_{\mathit{i}\mathit{j}}{\mathit{c}}_{\mathit{i}\mathit{j}}+{\mathit{Z}}_{\mathit{i}}\le {\mathit{y}}_{\mathit{i}})$$

(2)

Time:

$${\sum }_{\mathit{j}=1}^{\mathit{n}}{\mathit{x}}_{\mathit{i}\mathit{j}}+{\mathit{t}}_{\mathit{i}\mathit{j}}+{\mathit{h}}_{\mathit{i}}={\mathit{T}}_{\mathit{i}};$$

(3)

where t_ij is the travel time to the site, j; h is associated with the individual working hours, and T is the individual total available time. It also assumes that visiting time for every site is the same.

With explicit arguments, individual demand i for site j is given by $x_{i1}=f(c_{i1}+t_{i1}{w}_i,\dots ,c_{in}+t_{in}{w}_i,q_1\dots q_n,y_i^f$), where $y_i^f$ is the individual full income ($y_i^f=y_i^0+w_i{T}_i$), or the amount of money the person could earn if he worked his whole available time; w_i, is income after paying taxes; each q_i is the exogenous quality of the jth site; and ${\mathit{y}}_{\mathit{i}}^{\mathbf{0}}$ is the individual adjusted income by $y_i^0=y_i+w_i{h}_i$, where ${\mathit{y}}_{\mathit{i}}$ is the individual income.

Once the general demand model has been defined, estimating the associated parameters for each trip determinant is possible. The most common model in TC is the Poisson count model. The demand curve is determined for each individual site i in a given population, given as $x_i^{*}=f\left(z_i\right)+{\epsilon }_i$, where $z_i=(p_{ij}, j=1,\dots ,n;q_j, j=1,\dots ,n, y_i^f)$ and $p_{ij}=c_{ij}+w_i{t}_{ij}$. Poisson models specify the demanded quantity, trips, as an integer non-negative random number, with the mean independent from exogenous regressors. The expected functional form for Poisson models is typically exponential. For single site models, such as this, the general count model is written as $\mathrm{Pr}\left(x_i=n\right)=f\left(n,z_i,\beta \right), n=\mathrm{0,1},2,...,$ and the probability density function is $\mathrm{Pr}\left(x_i=n\right)=\left(e^{-\lambda }{\lambda }_i^n/n!\right), n=\mathrm{0,1},2\dots$, where ${\lambda }_i>0$ and is specified as an exponential function ${\lambda }_i=\exp (z_i,\beta )$.

However, if the assumption of equality between the Poisson model’s mean and variance is empirically invalid or if data are generated by a mechanism that structurally excludes zero counts, the regressor will probably be biased. In this case, truncated count models are recommended [62].

3.4.1. Truncated Models

Truncated count models must be used if at least one of the following three situations exists: (i) if data came from a mechanism structurally excluding zero counts, as in this case (Figure 3a), under this assumption, Poisson distribution must adjust only when data values begin at one; (ii) if on-site sampling is conducted, it ensures that questionnaires are applied to visitors [62,63]. The on-site sampling allows only visitors with trips x_i > 0 to be interviewed. In the on-site interviewing process, there is a high probability of interviewing visitors with a high frequency of site visitation since these individuals are more likely to be selected, known as truncated error or truncated demands [60]. (iii) If a population that visits a recreational site is considered and divided into strata based on the number of trips, such that the stratum, i, contains individuals who make i trips, it causes endogenous stratification [63].

This phenomenon occurs when the systematic variation in the selected proportion depends on the characteristics of the individuals in the sample or when the proportion of individuals chosen systematically varies from the population proportion. For distributions presenting these issues, the truncated Poisson model and the model with endogenous stratification can be applied [63,64].

3.4.2. Negative Binomial Models

The estimator for the truncated model could be biased and inconsistent under the presence of over-dispersion (α), defined as the excess of conditional variance over the corresponding conditional mean of the dependent variable (when the ratio of variance/mean is higher than one). In such conditions, the negative binomial distribution must be used as an extension of the Poisson distribution [60].

For the negative binomial distribution, the functional form of λ is ${\lambda }_i=\exp (x_i\beta +\xi )$ with a gamma distribution with mean of 1 and variance α. In addition, the random independent variable is λ and its variance $\lambda +\alpha {\lambda }^2$. The ratio of mean/variance is $1+\alpha \lambda$, such that the over-dispersion degree is a function related to λ and α. If $\alpha \to 0$, it implies no data over-dispersion, and the negative binomial distribution is reduced to a Poisson distribution on its limit. However, when using negative binomial models, if the on-site sampling problem exists, if a mechanism that structurally excludes zero counts was used to collect data, or if there is a presence of stratified bias, it is highly recommended to use negative binomial truncated models [62,65]. Once the model is calibrated and validated, WTP could be estimated.

4. Results

4.1. Descriptive

The sample shows an average number of visits to the BNPA of 6.81 times in the last five years, with 12 days of staying and five persons traveling with the interviewee. For monetary variables, averages are in USD: (i) 652 for monthly income, (ii) 44 for travel cost to BNPA, (iii) 65 for daily feeding expenses, (iv) 125 for daily lodging rate, and (v) 230 regarding total cost. Visitors’ gender is distributed 62% male and 38% female. Visitors by origin are disaggregated as follows: (i) 24% domestic, (ii) 27% American (USA), (iii) 21% European, and (iv) 28% another. The 42% of domestic visitors are locals (from La Paz City). In total, 72% manifest the current visit as the first to the BNPA. Only 45% of visitors considered that the BNPA is of a good conservation status. Schooling disaggregates as follows: (i) 2% elementary, (ii) 28% high school, (iii) 68% college, and (iv) 2% postgraduate (master or Ph.D.). Lastly, 63% manifest that visiting the BNPA and the mushroom-shaped stone were the main reasons for visiting (Figure 3b–e).

4.2. Recreational Demand Model

For estimating the demand model, trips to the BNPA (V) were used as the dependent variable, and eight independent variables were used to explain the dependent variable (

Table 3. Variables included in BNPA recreational demand models.

Variable	Description
V	Dependent variable. Number of trips to BNPA in the last five years, including current visits
tc	Travel cost natural logarithm (transportation cost plus cost of the trip or expense made to get to BNPA)
first	1: if is the visitor’s first visit to BNPA, 0: no
usa	1: If visitor’s origin is from USA, 0: Other
bnpa	1: Visiting BNPA or mushroom-shaped stones was their main trip motivation, 0: Other
stay	Natural logarithm of staying in La Paz City (days)
college	1: If the visitor’s schooling is college, 0: Other
income	Natural logarithm of visitor’s declared monthly income
pers	Reciprocal of the number of persons traveling with the visitor

). The Truncated Negative Binomial Model (TNB) was chosen because it presented the best fit to the data and was more significant in the respective statistical tests than the other estimated models (

Table 4. TNB model for BNPA recreational demand. Dependent: V.

Variable	TNB
	N = 143
tc	−0.2223
	(2.65) ***
first	−1.0314
	(4.61) ***
usa	1.0252
	(5.80) ***
bnpa	−0.4341
	(2.47) **
stay	−0.0048
	(2.02) **
college	1.4564
	(7.26) ***
income	−0.6785
	(7.10) ***
pers	−1.0212
	(2.37) **
cons	8.7471
	(8.95) ***
ln(α)	−0.2859
	(2.02) **
Pearson R2	0.8867
Pseudo LL	−356.7948
Chi2 (8)	148.24
Pr > Chi2	0.0000

Source: Self-estimation based on survey data. ** p < 0.05; *** p < 0.01.

).

Generally, the parameter associated with tc presents the expected sign and is statistically significant at 1%. All variables are statistically significant at the traditional confidence levels (10, 5, and 1%). For global models’ significance, the pseudo-log-likelihood (Pseudo-LL) demonstrates that the best model will be considered if its value is the closest to zero in absolute values. This criterion indicates that the TBN model is reliable in explaining trip demand to BNPA. The recommended values for statistical significance for these measures are −15.13 and −10.83 for α = 0.0001 and α = 0.001, respectively. The Chi² evaluates the null hypothesis that all coefficients are zero; its value rejects this hypothesis and is statistically significant at 1% [65,66].

Several statistical R² tests measure the count models’ goodness of fit [65,67], highlighting that for this kind of model, it is more beneficial to use Pearson’s R² or Deviation R² [68,69]. In this study, the goodness of fit of the models is defined by Pearson R², a measure that yields values above the recommended values established for cross-section data (ranging from 0.20 to 0.40) [70]. Given these measures, it is proven that the TNB model has the best goodness of fit and statistical significance, given the independent variables. The statistical significance of α indicates that there is no over-dispersion.

The estimated model suggests that if travel costs increase, the BNPA visits will decrease; since its elasticity is inelastic, this negative effect on visits will be very small, under 3%. The visitors’ staying time will negatively affect the visits. The income elasticity is negative (as the parameters are also negative), indicating that visits to BNPA are an inferior positive, meaning that the demand will reduce if income increases. A first-time visitor to the BNPA whose primary reason for traveling was to visit the recreation site may not repeat the visit. A positive effect on visits may occur if the visitor’s origin is from the United States of North America or has attended college. If the number of persons traveling with the interviewee increases indiscriminately, the likelihood of increased visitation will be negatively affected. Then, the model should be capable of predicting the number of visits declared by the interviewee; in this case, predicting a high number of visits is easier than predicting a low number of visits since visits’ central tendency parameters are biased to the right (Figure 4).

4.3. Willingness to Pay Calculation

The demand function’s semi-logarithmic form precludes traditional estimates of WTP. Estimating WTP when the model has a semi-logarithmic form requires two steps [71]. First, the travel cost elasticity must be calculated by the equation ${\epsilon }_{tc}^V=-\beta ·(1/\overline{x})$, where β is the travel-cost-associated coefficient and $\overline{x}$ is the average visits to the recreational site. Secondly, WTP must be estimated using the following formula: ${WTP=\overline{x}/\epsilon }_{tc}^V$ (

Table 5. WTP, recreational economic value, and welfare loss for BNPA beaches (USD).

Concept	TNB Model
WTP	11.11
REV-AA	1,345,074
REV-2021	458,269
REV-5 m2	9.13
WL-2MA	223,920
WL-5B2M	629,535
REV-5BCC	3,781,579

Source: Self-elaboration. Exchange rate: 20.10 MXN/USD.

). The elasticity calculation gives a value of 0.0326, which indicates an inelastic demand curve.

The recreational economic value (REV) for Balandra Beach A—or the main beach—was estimated assuming three scenarios: first, assuming beach A’s annual maximum effective carrying capacity (MCC) is reached (REV-AA); second, using the total visitors in 2021 (REV-2021); and third, the REV per square meter considers the maximum surface area a tourist occupies (REV-5 m²) as the value of the five beaches reaching their MCC (REV-5BCC) divided by the total surface of the BNPA and multiplied by five. Lastly, the recreational welfare loss caused by the two monthly closures of the BNPA was also estimated under two scenarios. First, the monthly MCC in the main beach (WL-2MA) is assumed. Second, under the same assumption, welfare loss is estimated for the five beaches part of the BNPA complex (WL-5B2M); the results are shown in Table 5. Appendix A shows the individual REV for each beach of the BNPA assuming a yearly maximum effective carrying capacity.

5. Discussion

5.1. Recreational Economic Value

WTP estimations could be helpful for site management, such as establishing or modifying access fees. Results indicated that visitors are willing to pay USD 11.11/person/day above the current access fee. The value of the estimated WTP matches with several examples of fees from Latin American countries, like Costa Rica, Peru, Colombia, and Ecuador [72]. An approach to using these monetary schemes is to use them as a demand control mechanism for the number of visitors to reduce overcrowding and diminish anthropogenic pressure on coastal ecosystems [73].

If this fee scheme could be implemented, the BNPA could collect finance resources under three outlines: (i) low collection, which REV-2021 represents; (ii) medium collection, as REV-AA shows; and (iii) high collection, as shown by REV-5BCC estimations. If this scheme operates, recreational demand for the BNPA will be reduced by 3.26% at maximum. The estimated direct expenditure for nature-based tourism in La Paz is USD 13 million [74], and the estimated REV under the previous collection classification represents 3.52, 10.34, and 29.09%, respectively, of this amount. Another way to set the new fee is by using the REV-5 m2; in such cases, the difference between the fee set using the WTP will not be so different. However, the collected resources will be higher under the WTP scenario.

One way to manage these financial resources might be through the Advisory Council (AC) of the BNPA, which is an instance of the local society participating to contribute to the management and administration of the MPA. The AC seeks to promote an agreement-building process to establish commitments and responsibilities among MPA stakeholders [75].

5.2. Welfare Loss

On the other hand, estimated welfare loss represents the value of the damages to the natural capital stock of the BNPA, affecting local, domestic (national), and foreign visitors [76]. Welfare loss is frequently related to measuring the impact of an undesirable environmental change. This is measuring the change in consumer surplus with a decline in the environmental asset, and it must be highlighted that welfare losses due to environmental degradation or damage in emerging economies arise from different sources like urban pollution, soil erosion, reduction both in quality and quantity, or an environmental good or service. If enough agents are affected by the environmental change, then prices will change. This change is uncertain but is not independent of the associated causes.

Welfare loss will increase with increasing instability regarding complex environmental ecosystems, such as coastal ecosystems. The reduction in coastal ecosystem services implies a reduction in ecological resilience, which increases the risk that local human communities will lose some of these essential ecosystem services. These points are relevant for both rich and poor countries but are more relevant for poor countries; the magnitude of welfare losses due to environmental degradation is even greater. The reduction in ecosystem services in quality and quantity has been more pronounced in emerging economies [77]. The estimated welfare loss for the two-month closure represents 5.92% of REV-5BBC, 16.64% of REV-AA, and 48.86% of REV-2021. Welfare loss allows for dimensioning the cost of damage to the BNPA ecosystem. It also allows us to compare the ecosystem damage cost versus the cleaning cost, which is 17.45 times bigger than WL-5BBC and 6.37 times bigger than WL-2MA.

5.3. Beach Management

Beaches are an essential source of ecosystem service flow for tourists, coastal tourism activity, and localities promoting them. However, coastal ecosystems, including beaches, are threatened by anthropogenic actions. There is a need for information that could be used to encourage more efficient management for MPAs. An essential piece of this information is linked to its recreational economic value. Most MPAs’ management plans in Mexico lack the economic value component, or any other type of value, which is essential when negotiations on the recreational site’s economic importance arise, or in case of damages made to, or inside, the site ecosystem.

There are three main aspects to consider when managing an MPA. First, beach management and carrying capacity, which refer to effectively managing beaches and nearshore areas, require a multifaceted approach. Coastal-Defense Beach Management integrates both physical and anthropogenic factors. Physical parameters encompass beach morphology, including type, width, chromatic properties, and gradient. Anthropogenic factors include access modalities, built environment characteristics, anthropogenic disturbance levels, and algae presence [9]. A critical metric in this management framework is carrying capacity, defined as the maximum user density that maintains acceptable environmental quality. Research indicates an optimal carrying capacity range of 3–6 m² per user [78]. Exceeding this threshold may result in a quantifiable reduction in the perceived recreational value of the coastal environment [9]. Therefore, if occupancy in the BNPA continues to exert pressure on the site due to the intensification of the two shifts that currently operate, in the long term, beach width and steepness will be reduced due to this anthropogenic use. Also, beach width is affected by climate change; combined factors can exacerbate the deterioration of the coastline and its components. As a result, the beach area and quality perception will be reduced. These put more pressure on rural beaches, which often are less defended; on the other hand, high-occupancy beaches need beach recharge.

Second, maintenance protocols, safety measures, and spatial management should be considered. Beach maintenance strategies are tripartite, focusing on detritus removal, the preservation of geomorphological parameters, and continuous monitoring of coastal defense and amenity factors [79]. Anthropogenic marine detritus, primarily attributed to visitor activities, poses significant threats to coastal ecosystems and aesthetic values. Safety protocols incorporate proactive lifeguard services and systematic water quality assessments adhering to established standards [19,80]. Beach safety is an essential matter in high-occupancy beaches; lifeguards, tourists’ security surveillance, safety, first aid equipment, and first responders in emergencies for remote beaches are some aspects that local, state, or federal authorities should provide to beach users. Some key factors for safety are handrails, warning signs, maritime signaling, shallow slopes or steps for access, and public access over/around structures [81]. Beaches demonstrating superior safety and hygiene metrics may qualify for international quality certifications, such as the Blue Flag Award [82]. Spatial management through beach zoning strategies is implemented to mitigate conflicts arising from diverse recreational activities [19,83]. This approach employs spatial and temporal segregation, considering activity popularity, participant volume, and temporal distribution to optimize resource allocation and user experience [84].

Lastly, infrastructure provision and public education initiatives should be considered. The provision of adequate infrastructure is a key determinant of coastal recreational quality. Essential amenities include hygienic facilities, gastronomic services, medical aid stations, transportation infrastructure, and accessibility enhancements such as boardwalks [85]. Complementing these physical elements is a comprehensive public information and education strategy. This involves disseminating data on safety protocols, zoning regulations, available facilities, access points, and ongoing maintenance activities. Information distribution utilizes multiple channels, including strategically placed signage, informational documentation, public awareness campaigns, and digital platforms [86]. The objective is to convey clear, readily comprehensible information to visitors, enhancing user experience and compliance with coastal management protocols [85,86].

5.4. Limitations

The ITC method does not consider multi-purpose or multi-site travel or lodging and feeding expenses. MPAs’ main characteristic is to protect pristine, high-biodiversity, or unique sites. Because of this, it is not easy to compare REV between sites. In addition, visitors’ preferences could influence the distance traveled or cost. Another limitation refers to when specialized equipment is required to perform a recreational activity; in that case, maintenance costs could vary depending on how specialized the equipment is [87]. In addition, ITC estimations are static and valid only for the period analyzed; therefore, the results must be taken within the surrounding context.

Establishing sustainable finance mechanisms is often identified as the critical barrier to effectively managing an MPA [1]. Without active management, tourism activities have been shown to cause environmental degradation and create inequities and conflict within coastal communities [86,87,88,89,90,91].

In emerging economies like Mexico, increasing fees does not guarantee that the collected amount will return to the MPA that generates it. When visitors and tour operators are not informed about where the fees are going or cannot see quality improvements in the site in categories like management and infrastructure, they are unwilling to pay fees or any increase because they think higher fees would reduce visitation to the BNPA [92]. Nevertheless, this effect could be divided between domestic/local tourists and foreign visitors. A lower visitation rate could be expected for foreign visitors; domestic/local visitors could approach zero, with fees above the social optimum [93]. Recent research for Mexico, which relates to WTP and crowding, indicates that international tourists have a very low acceptance of crowded recreational sites, and local tourists have the highest acceptance level of crowded recreational sites. On the other hand, international tourists have the highest WTP for avoiding crowded recreational sites, and local tourists have the lowest WTP for accepting crowded recreational sites [94].

6. Conclusions

Effective MPA management brings ecosystem benefits that could promote local economies directly or indirectly. Tourism is the sector most benefitted by conservation effects and actions. Therefore, MPA managers must be aware of challenges and opportunities around them; this way, they could obtain and use better-quality economic information to guide environmental and conservation policies in their favor.

The economic resources collected through a new fee scheme based on the WTP estimation in this research could strengthen the whole gradient of management strategies and conservation efforts, generating a possible and viable sustainable finance scheme. If implemented, it would help to settle the budget constraints that Mexican MPAs face when allocating economic resources to surveillance, monitoring, and cleaning programs. The research shows that positive and negative externalities could be internalized to be considered in the decision-making process.

The tourism industry (3S market) benefits from coastal and marine ecosystems; therefore, they should be morally committed to assisting MPA managers in coastal area conservation. For example, this assistance could be through a “blue” fee, levy, or tax, using the estimated WTP in this research as a baseline. Revenues raised from increased fees (tax or levy) represent less than one-third of the direct expenditure of nature-based tourism for the La Paz municipality, so it is possible to think of market-based environmental economic instruments to support BNPA management.

The results provide evidence for the economic viability of modifying entry fees to finance necessary improvements in beach management without drastically affecting visitation to the BNPA. It also could help reduce overcrowding since the price elasticity is highly inelastic, as increasing the access fee (tax or levy) would have little impact on the total number of visits. Nevertheless, increasing fees could have a two-way negative impact, diminishing local/domestic visits more severely than international tourists.

Adjusting or increasing access fees to the BNPA might be an objective to generate revenues from entrance fees. Still, it could not be the best strategy to search for sustainable funding if there are gaps in knowledge about visitation levels, tourism revenues, tourist beach quality perception, and associated management costs. Results also exhibit that BNPA managers can obtain higher income through site access fee modification without increasing pressure on the coastal ecosystem while covering maintenance, surveillance, and operational costs.

Ecological damages could be measured by revealed preference methods in environmental economics. The estimated welfare economic loss caused by the two-month closure of BNPA is, perhaps, higher than the cost of beach and coastal cleaning. An important percentage of the cost of effective beach management could be recovered through entry fee modification. This value could be considered a baseline to establish fines or charges for ecosystem damages associated with anthropogenic activities or to evaluate the amount of damage caused by incidents or unforeseen events generated by tourism activity inside the BNPA’s influence polygon.

On a final note, poorly planned and managed tourism activity generates long-term cumulative environmental impacts that are invisible to managers in the short term. Because of this, conservation policies must consider the economic value component as a management tool and to have continuity and sustainable financing.