Ecotourism In The Maya Biosphere, Guatemala

Environment, Development and Sustainability (2005) 7: 303–318 DOI 10.1007/s10668-004-2944-3


Springer 2005

TOURISTS’ AND LOCALS’ PREFERENCES TOWARD ECOTOURISM DEVELOPMENT IN THE MAYA BIOSPHERE RESERVE, GUATEMALA ROBERT R. HEARNE1,* and C. ALEJANDRO SANTOS2 1

Department of Agribusiness and Applied Economics, North Dakota State University, Bob 5636, Fargo, ND 58105, USA; 2Centro Agrono´mico Tropical de Investigacio´n y Ensen˜anza (CATIE) Managua, Nicaragua (*author for correspondence, e-mail: [email protected]; fax: +1-701-231-7400; tel: +1-701-231-6494) (Received 28 October 2003; accepted 27 May 2004)

Abstract. Ecotourism has been identified as a low impact means to provide income generating opportunities that are complementary to nature conservation as well as the welfare of the local population. The Maya Biosphere Reserve in Peten, Guatemala is one of Central America’s largest and most isolated protected areas. And although the area is contains a well-visited archeological site, it has not been developed for ecotourism. In this study choice experiments are used to analyze the preferences toward alternative scenarios of ecotourism of two important stakeholder groups: foreign tourists and educated local residents. Results demonstrate that these two populations of have unequal but similar preference orderings, especially toward: improved national park management and the presence of guides for wildlife viewing. These stakeholder groups had different opinions toward paved access roads and the presence of illegal colonists within the protected area. Both populations favored an entrance fee toward the absence of an entrance fee. These preferences were generally consistent across other socioeconomic indicators. Key words: ecotourism, choice experiments, choice modelling, Maya Biosphere Reserve, Guatemala, protected area management, Tikal, national parks.

1. Introduction Ecotourism is considered to be a low impact means to provide income generating opportunities that are complementary to nature conservation as well as the welfare of the local population. Ecotourism supports nature conservation by providing an economic demand for natural ecosystems. Entrance fees and tourist expenditures provide financial incentives to national park managers and communities so that they maintain secure, accessible, and pristine visitation opportunities. Ecotourism facilitates rural development by diversifying income generation and complements other low-impact production and extraction activities. Ecotourism provides alternative recreation activities and thus complements beach and cultural tourism, as well as reduces the environmental impact of visitors on congested destinations. Readers should send their comments on this paper to: [email protected] within 3 months of publication of this issue.

304

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

Many national parks and natural areas are conveniently situated near population centers and tourist centers and can easily attract visitors and expenditures. However other natural areas, especially remote areas that are rich in biodiversity, have not been developed for tourism and do not receive tourist expenditures. This constrains the income generation opportunities of the local population and reduces their incentive to support nature conservation. Thus, tourism opportunities need to be developed that: (i) attract tourists and tourist expenditures to otherwise unvisited natural areas, especially those under threat from competing economic activities; (ii) complement nature conservation; and (iii) support the income generation and development needs of the local population. Tourist opportunities that meet the needs of natural area managers, tourists, and locals, need to be based on the preferences of the tourists and locals as well as the physical characteristics of the area. Efforts need to be made to understand tourist demand as well as the capacity of locals to supply services and share their communities with visitors. This is especially important in areas with historic civil conflict and areas where tourists and residents represent distinct social and cultural groups. Research on the analysis of tourists’ and locals’ preferences towards the development of natural areas and tourism services should support the decision making processes of protected area managers, government agencies, and entrepreneurs at the national and local levels. Also, new methodologies to analyze these preferences need to be developed and adopted to the needs of protected area managers and decision-makers. This paper presents an analysis of tourist and local preferences towards nature based tourism development in the Maya Biosphere Reserve in Peten, Guatemala. The Maya Biosphere Reserve, in Guatemala, is one of Central America’s largest and most isolated protected areas. Because of the presence of an extensive tropical rainforest and because of the popularity of the Mayan ruins at Tikal, the Reserve has a substantial tourist potential. However, outside of Tikal, the Reserve has not been developed sufficiently to exploit ecotourism. Entrance fees are not collected, access is not supervised, and basic tourist services, such as lodging and guides, are not provided at the Reserve. Given that the analysis will feature environmental and ecotourism services that have not as yet been offered, a stated preference analysis will be used to assess willingness-to-pay for hypothetical products. Choice experiments (sometimes referred to as choice modelling or contingent choice) is a generalized version of the popular dichotomous choice contingent valuation method (CVM) of valuation of environmental goods and services. As in CVM, survey respondents are asked to declare their preferences in a hypothetical contingent market. However, choice experiments present respondents with a broad menu of options of which price/payment is only a single attribute. For this reason authors have suggested that choice experiments have advantages over CVM, especially in the analysis of substitution and tradeoffs (Boxall et al., 1996; Adamowicz et al., 1998). Also, because price is only one of many attributes that have influence over choices, the impact of yea-saying should be less significant in choice experiments than in CVM (Hanley et al., 1998). Many of the first published choice experiments studies in the environmental economics literature concerned with protected area management. Some of these

ECOTOURISM IN GUATEMALA

305

surveyed resource users, such as Canadian hunters (Boxall et al., 1996). While other studies focused on the preferences of a broader set of citizens, principally urban residents in Canada, Austrailia, and in Great Britain (Adamowicz et al., 1998; Hanley et al., 1998; Morrison et al., 1999; Blamey et al., 2000). And Lindberg et al. (1999) analyzed residents preferences towards tourism impacts in Denmark. In Central America, Hearne and Salinas (2002) surveyed tourists to analyze attributes of national park management in Costa Rica. Both Costa Rican and foreign tourists preferred improved infrastructure, information, observation towers, and picnic areas. And DeShazo and Fermo (2002) assessed choice set complexity with assorted subsets of up to nine attributes. They estimated the preferences of recreationists in Guatemala and Costa Rica toward tourism development. Respondents demonstrated preferences towards access roads, trails, guides, and wildlife viewing, and the authors concluded that difficulties with complex choice sets can be mitigated in the design and estimation stages. In this analysis the hypothetical goods and services offered to survey respondents are ecotourism services and attributes of protected area management. Nearly identical surveys were presented to two distinct populations: (i) foreign tourists and (ii) educated residents. Thus this analysis can be used to compare and contrast the preferences of both these populations. The results demonstrate a near homogeneity in preferences between the two populations. The second section of this paper presents an overview of the problem of ecotourism development in the Maya Biosphere Reserve in Peten, Guatemala. The third section of the paper describes the methodology used in this analysis. The fourth section of the paper presents the results of the multinomial logit estimation of the data. The paper concludes with a discussion of the policy implications of the study as well as observations about the applicability of this type of analysis in other protected areas. 2. The study area The Maya Biosphere Reserve (MBR) in northern Peten Guatemala covers over 1.6 million ha of lowland, humid tropical forest. It is the largest protected area contained in one country in Central America. Together with the adjacent Calakmul and Montes Azules Biosphere Reserves in Mexico and the Rio Bravo Natural Reserve in Belize, over 4 million ha of forest cover are protected (see Figure 1, source Herrera-MacBryde and Villa-Lobos). Because of its relative isolation and its size, the MBR is rich in biodiversity with a variety of flora, fauna, and endemic species unique to the region. Nearly 130 threatened animal species occur in the MBR. The MBR was established in 1990 under the authority of The National Council of Protected Areas (CONAP). Its creation was a policy response to concerns about increasing deforestation in Pete´n (Shriar, 2002). As a Biosphere Reserve follows a model of sustainable regional development which includes integrated rural development projects (Kuhn, 2000). Since 1990 international development agencies

306

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

Figure 1. The Maya Biosphere Reserve, Peten Guatemala.

and environmental non-governmental organizations (NGOs), such as The United States Agency for International Development (AID), CARE, The Nature Conservancy, and Conservation International, have supported efforts to manage the area and preserve its natural resources. Under the current management plan, the MBR is subdivided into two principle designated land uses. The core zone of the MBR contains four National Parks and three protected Biological Reserves. This zone, which contains 50% of the MBR, is dedicated to nature conservation, archeological preservation, scientific research, and tourism. A multiple use zone is dedicated to sustainable land use practices that complement nature conservation. The core zone of the MBR is supposed to be free of human population. A particular controversy has developed due to the presence of colonists who have migrated to the Peten and settled illegally in the National Parks and Biological Reserves of the MBR. These colonists are mostly from the more populous southern areas of Guatemala, and they have no ancestral or cultural links to the Peten (Sundberg, 1988; Shriar, 2002).

ECOTOURISM IN GUATEMALA

307

Despite tourism potential and considerable support from international donors and NGOs, the infrastructure of the MBR has not been developed to receive nature-based tourists. Currently 12% of Guatemala’s tourism visits the MBR, principally to visit Tikal, which is one of the most impressive archeological sites in Mesoamerica. Most tourists arrive in quick excursions flying from Guatemala to the airport in Flores, Peten’s principle city, where they are lodged. Visits to other natural areas and national parks in Peten are rarely included in these excursions. There are no visitor centers, hotels, guides, or even entrance fees in the MBR outside of Tikal. Recently, ProPeten, a local NGO, has promoted hiking excursions into the MBR with some lodging in local communities. Lonely Planet and other tourist guidebooks do mention trekking opportunities for the adventurous traveler. But most of the information on tourism in Peten focuses on Tikal. Indeed the lack of the participation of the local population in the development of Peten as a tourist destination and the nature conservation goals of the MBR has been a constraint upon the general goals of the MBR as well as other community goals. The MBR was created without any substantive consultation of the local population. NGOs and MBR managers implement conservation programs without considering the development plans for the state of Peten. It is noteworthy that the promotion of ecotourism is among the priorities that CONAP has established in its master plan for the MBR, but income generation of the local population is not. Efforts by NGOs to develop ecotourism attractions are not coordinated with Guatemala’s Tourism Institute (INGUAT), which is often expected to provide inputs such as training. Since these tourist initiatives may not be prioritized by INGUAT as being viable, the resulting infighting can lead to cynicism among community members (Sundberg, 1988).

3. Methodology Choice experiments were used to assess preferences for management, conservation, and development strategies for the MBR. Choice experiments are based on Lancasterian consumer theory as well as random utility theory. The former proposes that consumers make choices not on the simple marginal rate of substitution between goods, but based on preferences for attributes of these goods. Random utility theory states that although consumer utility cannot be known, it can be decomposed into random and systematic components, such that for all consumers i, Ui ¼ Vi þ ei

ð1Þ

where Ui is the unobservable, but true utility for alternative i; Vi is systematic component of utility, which can be estimated; and ei is a random component. The systematic component Vi can be expressed as a linear function of explanatory variables, such that

308

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS Vi ¼ b 0 Xi ;

ð2Þ

where Xi is a vector of explanatory variables associated with the significant attributes of visitor choice (Louviere, 2001). The random component of utility allows the researcher to make probabilistic statements about tourist and user behavior. Thus, the probability that the consumer will select alternative i among a set of alternatives U can be presented as PðijUÞ ¼ PðUi > Uj Þ ¼ P½ðb0 Xi þ ei Þ > ðb0 Xj þ ej Þ
8 j 2 U;

ð3Þ

where P(.) indicates the probability of an occurrence. Attributes of MBR management were prioritized and selected based upon the needs and interests of local decision makers. Formal and informal meetings with researchers, administrators, tourists, residents, and interested parties in Costa Rica and Guatemala facilitated this process. An experts’ meeting of economists, foresters, and biologists working for governmental and non-governmental institutions in the region was held in Peten. Certain attributes proposed by the research team were validated by the experts’ group. Others were proposed by this group. A series of four focus groups with tourists was held in Flores. These groups validated the proposed attributes and levels and also tested the survey instrument. Local protected area managers and experts stressed the need to assess the degree of national park management as an important attribute. This group was concerned with public opinion over their level of control of the parks and with their own needs for administering the MBR. Because of this concern, levels of MBR administration were selected as alternative specific constants to identify the three major policy options, as described in Blamey et al. (2000). Table I lists the attributes and levels presented in the choice experiment. Each attribute has a level that is associated with the status quo. Alternative levels reflected options that were considered to be under serious consideration and feasible. Thus, the entrance fees were restricted to either the current condition of no payment or the alternative payments of $5.00 or $10.00 for foreign tourists and 5.00 quetzals or 10.00 quetzals for Guatemalans.1 The combination of these levels implies a full factorial design of 33 · 23 or 216 possible choice set combinations. This was reduced orthogonally to produce 9 choice set combinations which were combined in 24 choices set groups. Each choice set group contained three choice sets, with an additional option of ‘No Preference.’ Choice sets were listed as ‘A,’ ‘B,’ and ‘C’ but with policy labels that were explained to respondents to correspond with the conservation levels seen as the goal of MBR managers. The survey instrument was comprised of three sections. The first was designed to introduce the respondent to the research project and to make them comfortable with the issues. This section included questions about the respondent’s knowledge of and attitude towards the conservation efforts of the MBR. The second section

ECOTOURISM IN GUATEMALA

309

TABLE I. Attributes and levels presented in the choice experiment. Attribute

Description

Levels

Description

National Park Management

All actions affiliated with administration and control of protected areas so that conservation goals are met.

No management

Status quo, with minimal personnel, negligible infrastructure, inaccessibility, and incapacity to control areas, offer services, or collect fees

Little management Adequate management

Adequate personnel and infrastructure to control area, but not for public use Accessible areas with adequate personnel and infrastructure to meet conservation goals and to attract and serve tourists

Inexistent

Status quo, due to a lack of information concerning the few existing services. Visitors generally stay in Flores or at Tikal Cabins in rural communities with local services An international standard Ecolodge. operated by a concessionaire under conditions appropriate to the region.

Lodging

Hotel and lodging services available for tourists, in the MBR.

Rural cabins Ecolodge

Wildlife observation

Availability of expert guides

Without guides With guides

Human settlements

Access roads

Entrance fees

The presence of communities of illegal colonists inside the core areas of the MBR

Roads that connect different national parks within the reserve. Currently the road from the airport in Flores to Tikal.

Status quo, no assistance of knowledgeable personnel Knowledgeable personnel to assist in locating and identifying flora and fauna

With illegal settlements

Status quo, communities of colonists with no ancestral link to the regions can remain inside the core areas of the MBR

Without human settlements

Colonists relocated by the Government of Guatemala and given title to other properties

Unpaved

Status quo, because of lack of pavement certain areas are inaccessible during part of the year

Paved

The Government of Guatemala will pave roads and thus improve access for the entire year

$0 for Entrance fees for Guatemalan a daily visit to the and foreign National Parks. Guatemalans receive beneficial rates. Exchange rate for 2000 Q. 7.50 = $1.00

Status quo, no payment currently charged

310

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

TABLE I. Continued. Attribute

Conservation Levels

Description

These are the policy labels listed horizontally to distinguish choices A, B, and C

Levels

Description

$5.00 for foreign Q. 5.00 for Guatemalan $10.00 for foreign Q. 10.00 for Guatemalan

An intermediate payment level

A relatively high level of payment

A, intermediate

Without ecological services nor conservation programs

B, satisfactory

Ecological services and conservation programs with limitations Optimal levels of conservation and high visitor satisfaction

C, excellent

of the survey was the choice experiments. Information on each attribute was provided to respondents so that they could make informed decisions. Each respondent was requested to complete three choice sets, with a ‘No Preference’ option. The third section of the survey included questions about the socioeconomic characteristics of the respondent to use for a generalized analysis. Surveys were conducted between April and June of 2000. The survey of foreign tourists was conducted in either English or Spanish at the departure lounge of the international airport in Santa Elena, Peten. This is the principle conduit for tourists who visit Tikal, and was considered to be an ideal location where travelers would have sufficient time to dedicate to the survey with few distractions. Due to prior experience with similar surveys, it was decided that the target population for the survey of locals should be secondary school-educated locals. The survey of local residents was conducted in Spanish at the site of the University Center of Peten as well as in the central plazas of Flores and Santa Elena. The university offers courses in agronomy, forestry, archeology, and administration of tourism. It is considered to be a center where a variety of secondary school-educated locals would converge. Given that the experimental design produced 24 different choice set combinations, these were each presented to eight different individuals in both populations. Thus each survey sample was comprised of 192 individuals. With each individual responding to three choice group sets, 576 observations were possible for each population. Because a few respondents did elect a ‘No Preference’ option when presented with choice sets, the final data set included 499 observations for foreign tourists and 532 observations for locals. The collected data was analyzed in order to estimate the probability of selecting a labeled alternative corresponding to conservation goals (A ¼ Intermediately Satisfactory, Satisfactory, Highly Satisfactory). The systematic component of the utility function was assumed to be a linear function of the selected attributes with the form:

ECOTOURISM IN GUATEMALA

311

Vi ¼ a1 Satisfactoryi þ a2 Highly Satisfactoryi þ b1 Little Managementi þ b2 Adequate Managementi þ b3 Rural Cabinsi þ b4 Ecolodgei þ b5 With Guidesi þ b6 Without Human Settlementsi þ b7 Paved Roadsi þ b8 $5:00 Entrance Feei þ b9 $10:00 Entrance Feei

ð4Þ

These variables are all considered to be discrete changes from the status quo levels presented. Assuming independently and identically distributed (IID) error terms with an extreme value type 1 Gumbel distribution, a multinomial logit model can be used to estimate the attribute coefficients. Violations of IID would be reflected in violations of the independence of irrelevant alternatives (IIA) (Louviere, 1996). In order to test for IID/IIA violations, a series of Hausman–McFadden tests were conducted. This test analyzes the impact of the removal of a choice from the data set on the probability of the selection of the remaining choices. Because the elimination of each choice from the complete data set resulted in a singular covariance matrix, these tests were not conclusive. Thus, the multinomial logit model was used. 4. Results and discussion The research was designed under the assumption that there are two distinct populations, middle-class Peteneros and foreign tourists. The survey instrument was identical for both groups, with the exception of the language used and the denomination of the entry fees in U.S. dollars for tourists and quetzals for locals. This is in accordance with established practice throughout Central America, where nationals pay a reduced entrance fee. Table II presents the socioeconomic characteristics of the survey respondents. Both samples are relatively well educated. The sample of foreign visitors demonstrates a high percentage of low-income and lowexpenditure backpack tourists. An initial test of the equality of the preference orderings between the two populations was conducted under procedures outlined by Swait and Louviere (1993) and Blamey and Bennett (2001). The equality of the combined coefficients and scale parameters was rejected with a likelihood ratio test of the form: 2½log Lðpooled dataÞ  log LðforeignÞ  log LðlocalÞ
¼ 68:08  v211 ; ð5Þ where, log L is the log-likelihood estimated from the corresponding model with 11 degrees of freedom for the ten restricted parameters and the varying scale parameter l. Given that the critical value for the v2 distribution at the 95% confidence level is 19.68, the equality of the combined coefficients and scale parameters between the two populations was rejected. The relative scale parameter

312

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

TABLE II. Socioeconomic characteristics of survey respondents. Socioeconomic characteristics Foreign n = 192 Origin

Percentage USA and Canada Europe Latin America Other

48 42 8 2

Sex

Male Female

44 56

Education

Primary and secondary University Postgraduate

14 70 16

Annual income

$24,000 or less $24,000 to $48,000 $48,000 and greater

55 25 20

Daily expenditure

$30 or less $31 – $50 $50 or more

53 20 27

Peten Other

92 8

Sex

Male Female

59 41

Education

Primary Secondary University

3 53 43

Annual income

Q24,000 or less Q24,000 or more

76 24

Guatemalans n = 192 Origin

lforeign =lnational was estimated to be 0.80, and the variables for the data from the survey of national tourists were adjusted for the pooled data. This enabled the comparison of the b coefficients of the two groups. The resultant likelihood ratio of 67.73 was calculated and compared to the critical value of 14.07, and the hypothesis of identical preferences across populations was rejected. This test demonstrates that the two populations, Peteneros and foreign tourists, represent different preference orderings. Thus, two models of preference orderings are presented. Results for the multinomial logit analyses for both populations are presented in Table III. Both populations had significant preferences for the ‘Adequate Management’ over the status quo. Foreign tourists also preferred ‘Little Management’ to the status quo, although they had stronger preferences towards the higher management level. Both populations preferred ‘Observation with Guides’ relative to the status quo of ‘Without Guides.’ Foreign tourists declared a significant preference for ‘Ecolodge’ in relation to the current situation of ‘No Lodging.’ This was the lodging option which promised the most comfort. Foreign tourists did not have a significant preference for ‘Rural Cabins,’ and the local population did not have a significant preference for any lodging alternative.

ECOTOURISM IN GUATEMALA

313

TABLE III. Results of multinomial logit model of preferences for attributes. Attributes

National Park Management

Lodging Wildlife Observation Human Settlements Access Roads Entrance fees

Alternative Specific Constant, conservation levels

Foreign Tourists (n = 499) (significance of the model v29 ¼ 170)

Local Residents (n = 532) (significance of the model v29 ¼ 145)

Coefficient

Standard error

Coefficient

0.3845*

0.1570

0.0760

0.1498

0.9913**

0.1565

1.3292**

0.1499

0.0731 0.4070** 0.5453**

0.1495 0.1430 0.1273

0.0061 0.0650 0.4139**

0.1361 0.1404 0.1229

Without human settlements Paved $5.00 or 5.00 quetzal $10.00 or 10 quetzal B satisfactory

)0.2302*

0.1207

0.2314*

0.1137

)0.3118** 1.2632**

0.1202 0.1575

0.4247** 0.2316*

0.1176 0.1404

0.6026**

0.1663

)0.1928

0.1472

0.6229**

0.1261

)01693

0.1236

C excellent

0.2584*

0.1360

Levels

Little management Adequate management Rural cabins Ecolodge With guides

0.2281*

Standard error

0.1159

* significant at the 90% confidence level; ** significant at the 99% confidence level (P[|Z| > z]).

These populations differed in their preferences towards paved access roads. Foreign tourists had a significant preference for the status quo of unpaved access roads, while locals declared a significant preference for pavement. It is possible that this reflects a greater concern by foreigners for protecting the tropical forest ecosystem by maintaining remoteness. Locals would be expected to appreciate the facilitated transport and increased commerce that is possible with pavement. The divergent opinions might also reflect that local residents might have a better understanding of the benefits of pavement. Since the interviews were conducted during Guatemala’s dry season, tourists might not appreciate the difficulties encountered on unpaved roads during the rainy season. Residents and tourists also disagreed in their preferences towards colonists inside the MBR. Foreign tourists had significant preferences towards maintaining the status quo and allowing illegal migrants to remain in the MBR. Local residents of Peten’s urban centers of Flores and Santa Elena declared a significant preference for the removal of these colonists. Perhaps the most intriguing result of the multinomial logit model is the respondents’ stated preference for an entrance fee. Both populations demonstrated a significant preference for paying an entrance fee of either $5.00 or 5.00 quetzals as opposed to the status quo of no entrance fee. This significant result contradicts economic theory that postulates that ceteris paribus an individual would prefer to pay

314

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

more to less and prefer to pay zero than to have a positive payment. Furthermore, the positive preference for a payment negates the possibility of estimating willingness to pay from the results of the multinomial logit model. However, this result should not be discounted, nor rejected automatically as the result of yea-saying behavior. It merely demonstrates a recognition of the benefits of user fees. Although both populations demonstrated a preference for an entrance fee of $5.00/5.00 quetzals over no payments, they also showed a preference for an entrance fee of $5.00/5.00 quetzals over an entrance fee of $10.00/10.00 quetzals. This result is presented in Table III, which comes from the same model as presented in Table II, only with the entrance fees recoded such that the $5.00/5.00 fee is the baseline to which the other alternatives are compared. Table IV shows that for foreign tourists the $5.00 entrance fee is preferred to both the $10.00 fee and the $0 fee at the 99% confidence level and that for local residents the 5.00 quetzal fee is preferred to both the 10.00 quetzal fee and the 0 quetzal fee at the 90% confidence level. In order to test for the impact of the respondents’ socioeconomic characteristics on their preferences, mixed multinomial logit models were employed. These are mixed models because they presuppose that the probability of selecting an alternative is a function of the attributes of the alternative and the attributes of the respondent (So and Kuhfeld, 1995). Two characteristics were considered to be particularly important; these were daily expenditure for the foreign tourists and income for the local residents. Daily tourist expenditure is an indicator of the type of tourist that arrives and the impact of that tourist on the local economy. This is particularly important because Peten is a destination for both low expenditure backpackers and high-expenditure archeology enthusiasts. In general, the impact of these socioeconomic variables on the declared preferences was negligible. Table V presents all of the significant results. Tourists with higher daily expenditures were more likely to prefer the presence of guides. Locals with higher levels of income were significantly less likely to prefer the higher levels of national park management. 5. Conclusions and observations This analysis demonstrates that there is a significant demand for improved national park management and ecotourism services in the MBR. This demand, demonstrated by the positive willingness-to-pay for entrance fees, comes from TABLE IV. Test for preferences for entrance fee of $5.00/5.00 quetzals to alternatives. Entrance fees

Fee levels

0 $5.00 or 5.00 quetzal $10.00 or 10.00 quetzal

Foreign tourists n = 499

Local residents n = 532

Coefficient

Coefficient

)1.2632** ) )0.5506**

Standard Error 0.1575 ) 0.1306

).2316* ) ).2508*

All other estimates and results are identical to those presented in this Table IV. * significant at the 90% confidence level; ** significant at the 99% confidence level (P[|Z| > z])

Standard Error 0.1404 ) 0.1404

ECOTOURISM IN GUATEMALA

315

TABLE V. Results of mixed multinomial logit models with significant interactions with socioeconomic indicators. Attributes

Levels

Foreign tourists (n = 499) Coefficient

National park management

Lodging Wildlife observation Human settlements Access roads Entrance fees Alternative specific constant, conservation levels Interactions with daily expenditures

Interaction with resident income

Little management Adequate management Rural cabins Ecolodge With guides Without human settlements Paved $5.00 $10.00 B – satisfactory C – excellent

Satisfactory conservation * daily expenditure Excellent conservation * daily expenditure With guides * daily expenditure Little management * income

Standard error

Local residents (n = 380)

Coefficient

Standard error

Coefficient

Standard error

0.337*

0.159

0.332*

0.158

1.235*

0.638

0.982**

0.158

0.995**

0.156

3.285**

0.633

0.071 0.414** 0.502**

0.151 0.144 0.129

0.085 0.398** 0.248

0.151 0.143 0.179

)0.026 )0.093 0.291*

0.168 0.164 0.155

)0.175

0.158

)0.249*

0.122

)0.232*

0.122

)0.317** 1.284** 0.621** 0.952**

0.121 0.159 0.168 0.185

)0.337** 1.310** 0.645** 0.668**

0.121 0.160 0.168 0.128

0.266

0.202

0.266

0.202

)0.154*

0.061

).005

0.061

0.144*

0.062

Adequate management * income

0.556** 0.144 0.035 0.370**

0.158 0.183 0.184 0.142

)0.089

0.152

)0.215

0.147

)0.423**

0.141

* significant at 90% confidence, ** significant at 99% confidence (P[|Z| > z])

tourists who have made the effort to visit Peten but would like to diversify their activities in the region in order to appreciate the flora and fauna of the MBR. The acceptance of entrance fees to pay for improved park management demonstrates that tourism can financially support nature conservation within protected areas. And expanded nature-based tourism services should provide financial incentive to local residents to encourage tourism and support the conservation goals of the MBR. Tourists’ preferences for guides and lodging should stimulate local entrepreneurs to provide these services. Foreign tourists and local residents had similar, although unequal, preferences towards ecotourism development in the MBR. Both groups preferred adequate park management, wildlife viewing with guides, and an entrance fee. These preferences did not significantly vary across the socioeconomic indicators of the survey respondents. The demand for tourism services does not vary significantly between different expenditure groups. There is a complementary interest in local residents

316

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

in supplying these services. The surveyed population of educated, middle-class Peteneros would be those most likely to be contributing the services required to support expanded tourism. The preferences of tourists and locals diverged in respect to the presence of colonists within the MBR’s core zone and paved access roads. This divergence in opinion might reflect general sensibilities of the international community towards access to rainforest and colonization. The target population of secondary school-educated Peteneros is an important stakeholder group. Many of the local survey respondents would be expected to benefit directly from the economic activity that tourism would attract. But this target population is not representative of the entire local population because it does not include the peasant farmers and forestry workers who live near to and within the MBR. Further studies need to be conducted in order to understand how these rural stakeholders would respond to and perhaps benefit from ecotourism development. And further efforts need to be made so that these individuals consider themselves beneficiaries, along with the better educated Peteneros, of any increased tourism. The feasibility of nature-based tourism in the Peten is dependent not only upon the natural beauty and unique species that can be encountered in the MBR, but also upon the accessibility of the area, the available services, the sense of security that is provided to tourists, the availability of complementary tourist and recreational activities, and the affordability of travel. Given that the area including southern Guatemala, Belize, and southern Mexico has substantial tourism there appears to be affordable regional travel with alternative complementary activities. Thus the current constraint upon tourism in the MBR is the lack of facilities within the National Parks, the services offered, and the sense of security offered to tourists. This research has demonstrated positive preferences towards improved National Park and increased tourist services by both local residents and foreign tourists. This consensus among two critical stakeholder groups should encourage greater interest in ecotourism development among investors and decision makers. A remaining concern that can continue to inhibit tourism in Guatemala is the perception that the country is unsafe with continued civil discord. The development of the MBR’s ecotourism potential should become a priority for CONAP and the MBR managers. This development should become part of a strategy of providing incentives to local populations to cooperate with the nature conservation goals of the MBR. Other complementary efforts to provide improved security in tourist zones and to promote the MBR as an area of integrated naturebased and archeological tourism should prove beneficial. The development of nature-based tourism within the National Parks should assist in the overall nature conservation goals of the MBR, especially since much of the area remains isolated and in designated Biological Reserves. Choice experiments is an appropriate tool to analyze preferences towards nonmarket environmental goods and services and ecotourism development. It is appropriate to analyze changes from the status quo to a broad menu of possible alternatives. However, results of this study demonstrate that researchers should be

ECOTOURISM IN GUATEMALA

317

careful with the inclusion of a ‘zero’ fee option. Respondents preferred a payment of $5.00 or 5.00 quetzals over the absence of an entrance fee. This is an interesting result that does not imply irrationality. However, given this preference the results could not be used to present the respondents’ marginal willingness-to-pay for the other attributes. If the estimation of respondents’ marginal willingness-to-pay is an important research objective, researchers should carefully pretest when they use very low payment levels.

Acknowledgements Research was conducted while Hearne was Assistant Research Professor and Santos was Graduate Student at CATIE, the Tropical Agriculture Research and Higher Education Center in Turrialba Costa Rica. Research was supported by a scholarships provided by the German Service for Academic Interchange (DAAD) and the Swedish International Development Agency (SIDA). The authors thank Gilberto Paez , CONAP, and The Nature Conservancy for their support during the early stages of this research.

Notes 1

During data collection in 2000, the rate of exchange was 7.81 quetzals per dollar.

References Adamowicz, W., Boxall, P., Williams, M. and Louviere, J.: 1998, ‘Stated preference approaches for measuring passive use values: choice experiments and contingent valuation’, American Journal of Agricultural Economics 80, 64–75. Blamey, J. and Bennet. R.: 2001, ‘Yea-saying and Validation of a choice model of green product choice’, in J. Bennett and R. Blamey (eds.), The Choice Modelling Approach to Environmental Valuation Cheltaham, UK, Edward Elgar, pp. 178–201. Blamey, R., Bennett, J., Louviere, J., Morrison M. and Rolfe J.: 2000, ‘A test of policy labels in environmental Choice Modeling Studies’, Ecological Economics 32, 269–286. Boxall, P., Adamowicz, W., Swait, J., Williams, M. and Louviere J.: 1996, ‘A comparison of stated preference methods for environmental valuation’, Ecological Economics 18, 243–253. DeShazo, J. and Fermo, G.: 2002, ‘Designing choices sets for stated preference methods: the effects of complexity on choice consistency’, Journal of Environmental Economics and Management 44, 123–143. Hanley, N., Wright, R. and Adamowicz, V.: 1998, ‘Using choice experiments to value the environment’, Environment and Resource Economics 11(3–4), 413–428. Hearne, R. and Salinas, Z.: 2002, ‘The use of choice experiments in the analysis of tourist preferences for ecotourism development in costa rica’, Journal of Environmental Management 65(2): 153–163. Herrera-MacBryde, O. and Villa-Lobos J.: undated, Fact Sheet on Peten Region and Maya Biosphere Reserve. Smithsonian Institute. Available at http: //www.nmnh.si.edu/botany/projects/cpd/ma/ ma13.htm. checked 10/2003. Kuhn, M.: 2000, ‘Biosphere reserves as planning models for sustainable regional development: Schorfheide-Chorin, Germany’, Journal of Environmental Planning and Management 43, 897–904.

318

ROBERT R. HEARNE AND C. ALEJANDRO SANTOS

Lindberg. K, Dellaert. B. and Rassing, CR.: 1999, Resident tradeoffs. A choice modeling approach Annals of Tourism Research 26(3), 554–569. Louviere, J.: 1996, ‘Relating stated preference measures and models to choices in real markets: calibration of CV responses’ in D. Bjornstad and J. Kahn (eds.), The Contingent Valuation of Environmental Resources: Methodological issues and research needs. Cheltenham, UK, Edward Elgar. Louviere, J.: 2001, ‘choice experiments an overview of concepts and issues’, in J. Bennett and R.Blamey (eds.), The Choice Modelling Approach to Environmental Valuation. Cheltaham U.K. Edward Elgar, pp 13–36. Morrison, M., Bennett J. and Blamey, R.: 1999, ‘’, Water Resources Research 35(9), 2805–2814. Shriar, A.: 2002, ‘Food security and land use deforestation in rural Guatemala’, Food Policy 27, 395–414. Swait, J. and Louviere J.: 1993, ‘The role of the scale parameter in the estimation and the comparison of multinomial logit models’, Journal of Marketing Research 30:, 305–314. So, Y. and Kuhfeld, W.: 1995, Multinomial Logit Models. Technical paper available at http: //ftp.sas.com/techsup/download/technote/ts273.pdf. Sundberg, J.: 1988, NGO landscapes in the Maya biosphere reserve, Guatemala. The Geographical Review 88(3), 388–390.