Tourism Development in Small Islands Across the World*










Jerome L. McElroy

Professor of Economics

Department of Business and Economics

Saint Mary's College

Notre Dame, IN  46556   USA

TEL: 574-284-4488

HOME: 574-234-2827

FAX: 574-284-4566









*An earlier version of this paper was presented at the Islands VII Conference, Institute of Island Studies, University of Prince Edward Island, Charlottetown, PEI  (June 26-30, 2002).


            This study develops the Tourism Penetration Index and applies it to 51 islands.  The results confirm the most penetrated comprise Caribbean, Mediterranean and Northern Pacific islands typified by large resorts, crowding, short stays and the replacement of man-made attractions for lost amenities.  The least penetrated islands comprise mainly Pacific and Indian Ocean destinations characterized by small facilities, long stays, and limited infrastructure.  The intermediate islands primarily fall into two groups: Caribbean islands advancing to the high-density stage, and other destinations experiencing rapid growth and resource conflicts.  The study concludes with planning implications and suggestions for further research.  Key Words: tourism development, penetration index, islands, world.



            The postwar history of small islands has been marked by two favorable development factors: the march of decolonization and the global spread of international tourism.  In the first case, since 1960 roughly 30 tropical/temperate islands across the five major oceanic basins have become politically independent (McElroy and Mahoney, 2000).  Other island territories have achieved significantly greater internal self-government and have used this new-found autonomy—the so called “resource of jurisdiction”—to create tax havens and diversify into other nontraditional activities like off-shore finance and ship registry (Baldacchino and Milne, 2000).  In the second case, the remarkable transformation of tourism into the world’s largest industry—accounting for roughly a tenth of global GDP, employment and capital formation (WTTC, 2001)—has coincided with the restructuring of small island economies away from traditional exports like sugar and copra toward mass tourism and related construction.  The results have transformed insular landscapes across the Caribbean, Mediterranean and North Pacific and created the so-called “Pleasure Periphery” of North America, Europe and Japan respectively (Turner and Ash, 1976).

(a) The Problem

            However, much of this growth has been overly rapid, unplanned and intrusive and has damaged insular ecosystems (Briguglio and others, 1996).  In the Caribbean, tourism expansion has directly or indirectly caused deforestation and erosion of upland forests for condominium developments and road works, as well as beach loss, lagoon pollution and reef damage from sand mining, dredging and boat anchoring (McElroy and de Albuquerque, 1998).  Nearly 30 percent of the reefs are at high risk because of runoff and discharges of untreated municipal and hotel waste and pollution from pleasure yachts and cruise ships (Bryant and others, 1998).  Partly as a result, since 1985 fish catches are off nearly 50 percent in gross tonnage (UNEP, 1999b).

            In the Mediterranean, large-scale coastal hotel/marina and infrastructure construction has filled in salt ponds, disfigured shorelines, and polluted nearshore waters with sewage (Pearce, 1989).  In highly developed islands like Balearics and Malta, tourism has been associated with the rapid decline of traditional pursuits and renewable resource uses, the rise of realty inflation (Beller and others, 1990), and paralyzing summer crowding and other sociocultural intrusions that threaten insular lifestyles and identity (Lanfant and others, 1995).  In developing Indian Ocean islands, the situation is similar.  Tourism along with unplanned urbanization is associated with sand mining, mangrove destruction and coastal pollution.  Mauritius and Seychelles are ranked second and third in the world in terms of endangered native plant species (UNEP, 1999a), and some beach-based resorts are under threat from sea level rise.

            The Pacific, in transition from subsistence to a cash economy, is undergoing substantial threats from commercial agriculture and fishing, logging and coastal tourist development.  In popular resort areas delicate mangroves have been harvested for construction material and reefs scarred by trampling and collecting by tourists (Lobban and Schefter, 1997).  Development on Guam has been compared to suburban Los Angeles, and even the Galapagos has allegedly been overrun by excessive visitation (Lindberg and Hawkins, 1996).  As a result of these forces of modernization, the region boasts the largest number of bird extinctions in the world, and seven times more endangered species than the Caribbean (UNEP, 1999a).

            Since tourism’s biocultural base is in decline across the island world, and because of the continued pressures from increased globalization expected for the future (UNEP, 2002), researchers have called for greener, lower-density tourism styles and begun to explore the causes of this policy failure.  A variety of structural and institutional factors have been suggested (McElroy, 2002): Four stand out.  First, there is strong pressure on policy-makers to create labor-intensive opportunities in insular economics buffeted by (1) the loss of traditional agricultural (sugar, copra) preferences due to the formation of regional trading blocks (EU, NAFTA), and (2) by the fall-off in aid stemming from islands’ declining geopolitical significance in the post-cold-war era (Commonwealth Secretariat, 1997).  Second, the scale discrepancy between large, heavily capitalized, high-volume international travel interests (air and cruise lines, hotel chains, tour operators) and small, fragile insular ecosystems produce an inherent propensity for environmental overrun, and tourism’s on-site consumption character is conducive to socio-cultural disruption.  Third, over time the cumulative impacts of mass tourism’s uneven dynamics in small islands often catch decision-makers unprepared.  While visible economic benefits accrue linearly (jobs, foreign exchange, etc.), costs are often hidden until dangerous thresholds are crossed (beach erosion, fish kills, traffic congestion).  Finally, despite its pervasiveness, there is no standard measure of tourism’s overall impact to provide policy-makers with an early warning signal for broadly assessing potentially nonsustainable levels of visitation (U.S. Congress, 1992).

(b) Literature

            Carrying capacity has been a policy concern in small islands pursuing a mass tourism strategy particularly since the appearance of Butler’s (1980) destination lifecycle model.  This theory emphasizes the dynamic, market-driven thrust of tourism development and argues that successful destinations pass through a regular sequence of growth stages that parallel the S-shaped logistic curve.  Progress along the development continuum involves increasing industry institutionalization, facility scale, visitor saturation and cumulative ecological impact (Dann and Cohen, 1991; Butler, 1991).  Butler’s six stages include emergence, involvement, growth, consolidation, maturity and/or stagnation followed by decline or rejuvenation.  The concept is analogous to the product lifecycle in marketing literature whereby a new product is launched, achieves acceptance and growth until competitors gain market share, and innovation or repositioning is necessary to stave off sales and profit declines (Heywood, 1986).

            Although the model has been applied to over a dozen resort areas, these case studies lack standardized approaches, uniform measures and rigorous quantification (Getz, 1992).  As a result, conceptual and empirical difficulties remain including problems of empirically defining the stages, questions about the inevitability of the progression, and complications involving multiple products with multiple cycles in a given destination (Agarwal, 1994).  Despite these limitations, a number of authors continue to view the lifecycle model as a useful, descriptive, and analytic framework for understanding the general processes and patterns of tourist development (Hovinen, 2002) not only for individual destinations but also at the more macro level as a regional model of tourist evolution.  For example, Holder (1988) loosely invoked the lifecycle – his so-called “self-destruct theory of tourism”—to broadly characterize the postwar growth of mass tourism in popular Caribbean destinations.  Similarly, Minerbi (1988) and Choy (1992) sketched tourism’s development stages across a number of Pacific islands.

            In the same vein, more recent work has attempted to (1) reduce the number of lifecycle stages to more empirically definable levels, and (2) to develop more comprehensive measures of overall tourism impact.  In the first case, de Albuquerque and McElroy (1992) developed an abbreviated three-stage version of Butler’s model for the Caribbean.  By descriptively interpreting over a dozen (mainly economic) indicators, they roughly clustered 23 small islands into three levels of increasing tourism penetration: low-density emerging destinations, growing intermediate islands and mature high-density resort areas.  Follow-up studies (McElroy and de Albuquerque, 1994; McElroy and others, 1993) including a number of Pacific islands suffered the same limitations.  To overcome these shortcomings, in the second case, McElroy and de Albuquerque (1998) constructed a Tourism Penetration Index (TPI) and successfully applied it to the small-island Caribbean.  This instrument more comprehensively measures the cumulative scale and impact Butler envisioned for the lifecycle process, more systematically clusters destinations into discrete development stages, and provides a more balanced imprint of tourism on island economy, society and environment.  Most recent applications have extended the model to other small islands across the world (McElroy and Olazarri, 1997: McElroy and de Albuquerque, 1999; McElroy, 2002).




( c )  Scope

            The present study does four things.  First, it develops the TPI and discusses its strengths and weaknesses.  Second, it applies the model to 51 small (roughly one million population or less) islands across the world and interprets the results according to the scaled-down three-stage version of Butler’s lifecycle.  Third, it emphasizes the obvious planning challenges appropriate for each stage of tourist development.  Finally, it assesses the general usefulness of the TPI as a comprehensive measure of overall tourism impact as well as its function as an early warning signal particularly for those destinations experiencing and/or approaching high-density development and potentially damaging levels of visitation.


            Similar to Briguglio’s (1995) small-island vulnerability index, three major criteria governed the construction of the Tourism Penetration Index.  The first was that it be simple to formulate and interpret with rising TPI scores indicating increasing tourism penetration.  The second was that it be comprehensive enough to include all major dimensions of tourism impact.  The third was that it allow wide applicability through the use of standard, readily accessible data.  To achieve these purposes, the TPI was developed from three separate indicators which measure economic, socio-cultural and environmental penetration.

(a) Variable Selection

            The TPI was constructed in three steps.  First the variables were selected to measure tourism development across the three impact dimensions.  Second, standardized indices based on these variables were calculated.  Third, TPI scores were calculated as the unweighted average of the three standardized impact indices.  In the case of variable selection, given the widespread unavailability of tourism’s contribution to GDP in many destinations, visitor spending per capita (per resident population) was selected as a measure of overall economic impact because of its standard usage as well as its correlation with other measures of tourism development (Liu and Jenkins, 1996).

            The most common measure of tourism’s socio-cultural impact is the ratio of visitors to the local population.  A more rigorous indicator is the Tourism Penetration Ratio, i.e. the number of stayover visitors times the average length of stay divided by the population times 365 (CTO, 1993).  This study used a similar measure that includes one-day excursionists and cruise visitors and can be called the average daily visitor census or density per 1,000 population.  It is calculated as: the number of stayover tourists times the average length of stay plus excursionists divided by the host population times 365.  The resulting ratio is multiplied by 1,000 to yield the average daily visitor census or density per 1,000 population.  A figure of 100, for example, suggests that there are 100 daily visitors per 1,000 population indicating that tourist activity results in a 10 percent increase in the daily island population throughout the year.  This indicator was chosen as an aggregative, indirect proxy measure of hose-guest irritation and/or crowing and/or socio-cultural pressure.

            The census measure was selected in preference to the so-called Tourism Density Ratio (visitors X average stay divided by land area X 365) –to which it correlates closely—because the former focuses exclusively on social as opposed to environmental impacts.  To measure environmental penetration, the use of more direct measures like per visitor electric or water consumption or annual rate of arable land loss or deforestation was precluded by the unavailability of the data.  Instead, the number of hotel rooms per square kilometer was chosen to measure tourism’s impact on the physical environment.  Normally progression though Butler’s stages of tourism development involves enlargement of the hotel plant, increasing facility scale, and expansion of the associated air, sea (ports and marinas), and land (road networks) infrastructure.  Hotel rooms per km2 was selected as a commonly accessible indirect proxy for tourism’s imprint on the insular landscape and fragile ecology.

            The strength of these indicators, as constituted in the TPI, is that they represent an integrated comprehensive measure of overall tourism impact in small-island countries.  However, because of their highly aggregative nature, they remain rudimentary first approximations and require some interpretive caveats.  First, they do not measure the spatial concentration of tourism, a particularly important dimension for islands where sunlust activity is concentrated along the seashore.  Likewise they take no account of dualistic development in single-island countries and mask individual island differences in archipelagic states.  Second, the indices also mask the seasonability so characteristic of tourism like high winter visitor densities in the Caribbean and intense summer visitation in the Mediterranean.  Third, as single-year, cross-sectional indicators, none of these variables capture an island’s long-term experience with and adaptation to tourism development.  Finally, as land-based indicators they may not accurately measure tourist activity in certain destinations that cater principally to sea-based yachting tourism.

(b) Index Construction

            In order to develop the TPI scores, the three impact indicators were standardized according to the following formula:


TPIij =      (Xij – Min Xi)

                (Max Xi – Min Xi)

Where TPIij stands for the degree of tourism penetration for the jth island with respect to the ith variable, i taking on the value of 1 to 3 since there are three variables, and j taking on the values of 1-51 for the 51 islands in the sample (see below).  Xij stands for the value of the ith variable for the jth island.  Max Xi and Min Xi stand for maximum and minimum values of the ith variable for all islands in the sample.  If an island has a value of Xij equal to the maximum (minimum), its standardized score would be one (zero).

            The overall TPI scores for each island were then estimated by taking an unweighted average of the three standardized indices according to the formula:


TPIj = ∑TPIij, i=1,2,3; j=1,2…51


Since different weights could not be established a priori either from the theory or the literature, this equivalent impact formulation assumed that each type of impact (economic, social, environmental) was as important as the other two in contributing to overall tourism penetration.

To operationalize the index, a sample was selected of 51 small islands of roughly one million population or less for which complete data were available.  This data constraint excluded a number of small non-sovereign islands like Azores, Balearics, Faroes, Madeira, Mayotte, Wallis and Futuna and others.  To ensure uniformity, all data were taken from standard sources: the tourism data from the Compendium of Tourism Statistics (WTO, 2001), and the population and area figures from The World Factbook (CIA, 2001).  The resulting sample includes 23 islands in the Caribbean, 16 in the Pacific, five in the Indian Ocean, four in the Atlantic, two in the Mediterranean (Malta and Cyprus) and Bahrain in the Persian Gulf.  Three have slightly more than one million inhabitants (Hawaii, Mauritius, and Trinidad), and all are less than 20,000 Km2 in area except Iceland and Solomons.


            Table 1 presents the basic data, and Table 2 records the three impact variables, their standardized indices and their combined TPI scores and the overall destination rankings.  Although quite aggregative and rudimentary, the TPI scores yield results that broadly confirm what is expected.  The 51 islands are loosely ranked from most (St. Maarten) to least developed (Comoros).  The more traditional, developed, and accessible Caribbean, Mediterranean, and Northern Pacific destinations populate the top half of the rankings while the more remote, and recently emerging South Pacific and Indian Ocean destinations dominate the bottom half.  The sample divides roughly into three distinct groupings based on discrete levels of development as revealed by the TPI informed by historical observation and a sense of the literature.  They are most developed, intermediate, and least developed and correspond generally to three abbreviated stages of the destination lifecycle.

(Tables 1 and 2 here)

            The most developed islands form a subgroup of 13 internationally visible highly developed destinations characterized by average per capita visitor spending approaching $10,000 and an average daily visitor density over 170 tourists per 1,000 residents.  Tourists thus represent the rough equivalent of a 17 percent increase in the daily year-round population.  Their insular landscapes are crowded on average with nearly 25 hotel rooms per km2 of area.  This group includes three clusters: (1) five traditional resort islands comprising Bermuda, the British (BVI) and U.S. (USVI) Virgins, Hawaii and Malta; (2) six more recent entrants into the mass tourism market including Aruba, Caymans, St. Maarten, Turks/Caicos, Guam and Northern Marianas; and (3) two small Dutch Antilles, Saba and St. Eustatius, known for their dive tourism.  This last result points up one limitation of the TPI.  It overstates the impact in these two destinations and the BVI largely because the high visitor densities (ranging from 240 to 337 per 1,000 population) fail to capture the marine based nature of their tourist activity.  On the other hand, the TPI does capture the heavily built character (over 60 rooms per km2) of Bermuda, St. Maarten and Malta.  In addition, the relatively low per capita visitor spending in Malta ($1,709) may partly reflect the deep discounting by European tour operators common in Mediterranean resort islands.

            Many of these most developed destinations share a relatively unique profile.  According to the literature (McElroy and de Albuquerque, 1992), these mature, affluent areas advancing to the top of the resort cycle are characterized by high visitor and hotel room densities but relatively slow visitor and room growth rates.  Their market is by and large dominated by shorter-staying visitors (6.2 nights average) with a strong preference for hotels, large-scale (comfortable) facilities, and man-made attractions.  They also exhibit the highest levels of hotel occupancy, promotional spending, and (for the Caribbean) cruise passenger traffic.  As a partial indicator of their integration into the global tourist economy, they tend to display the lowest degree of seasonality through special year-round packages (honeymoon weekends, conventions, carnivals, regattas, etc…).  They also tend to exhibit a relatively high degree of man-made attractions (casinos, golf courses, conventioneering etc.).  Many of the older established destinations are also among the most frequently cited in the literature for tourism-induced ecosystem damage, marine pollution, over-crowding, host tensions, and declining vacation quality (Beckhuis, 1981; Towle, 1985; Beller and others, 1990; Jenner and Smith, 1993; Briguglio and others, 1996).  As one specific example, the USVI’s National Park was recently placed on the United States 10 most endangered list because of “over fishing, silt runoff from poorly designed developments on land, careless snorkelers and scuba divers, boat anchor dropping on coral reefs and seagrass beds and boats running aground on shallow reefs” (Larson, 2003:1).

            In conjunction with abundant natural amenities, the success of these mature destinations has been fostered by significant locational advantages and long-standing metropolitan commercial ties.  The geographical proximity of the Caribbean to North America, of the Mediterranean to Europe, and the Northern Pacific to Japan has facilitated tourism growth because of their relatively cheap access to the most lucrative origin markets in the world.  Inter-island proximity also helps explain the rapid diffusion of tourism across the Caribbean and the popularity of Guam and Saipan as Japanese weekend golf and honeymoon destinations.  Historically many of these islands also functioned as nodes in center-periphery trade, and these links nourished the inflow of foreign private hotel investment encouraged by generous, pro-growth tax incentives.  In addition, traditional political ties fostered the aid-financed transport infrastructure that forms the capital base of the visitor industry.

            The least developed islands contain 16 primarily Pacific and Indian Ocean islands located at the low end or beginning stage of the resort cycle.  As a group, they average $231 in per capita visitor spending, i.e. less than three percent of the most developed islands’ level.  Likewise, they average only eight visitors per 1,000 population and roughly one room per km2 of land area.  With some exceptions (Marshalls, Tuvalu), they tend on average to be larger in area and population than the mature destinations and to exhibit more diversified economies.  The largest islands like New Caledonis, Fiji and Solomons possess valuable resources—nickel, gold and silver respectively—while over half the labor force in Vannatu and Iceland is engaged in subsistence/small-scale agriculture and fishing respectively.

At the bottom of this grouping are five low-income remote outposts with minimum tourism development that resemble MIRAB states heavily dependent on emigration, remittances, foreign aid and public employment (Bertram and Watters, 1986).  They include Solomons and Kiribati in the Pacific, Cape Verde and Sao Tome/Principe along the west African coast, and Comoros in the Indian Ocean.  At the top end are several destinations with a couple of decades of tourist experience and some highly developed tourism zones masked by the aggregative TPI.  They include Maurituis, Reunion, New Caledonia, Fiji, and Iceland.  For example, Mauritius, the world’s second largest exporter of woolen knits, boasts the largest visitor industry in the Indian Ocean; and tourism earnings now rival the contribution of traditional sugar exports (Benhamou, 1993).  Finally, the middle rung is occupied by islands advancing up the cycle.  During the 1990’s Trinidad/Tobago experienced some of the most sustained increases in stayover arrivals, visitor expenditure and hotel room growth in the Caribbean (CTO, 2001) while Samoa in the Pacific achieved international recognition as an ecotourism destination for its natural beauty and cultural uniqueness.

Although it is difficult to simply classify these least penetrated islands because of their heterogeneity in size and overall level of development, many share characteristics common to Butler’s early tourism stages.  These include small-scale facilities and infrastructure (some lack jetports), limited visitor growth, and less disturbed cultures and ecologies than their mature, high-density counterparts.  They tend to spend least on promotion, have the lowest proportion of rooms in large (100+) hotels, the highest ratio of regional (inter-island) visitors, and their average length of visitor stay is the longest (10 nights), roughly two weeks in five cases (New Caledonia, Kiribati, Reunion, Solomons, and Tonga).  They also exhibit greater appeal for more adventurous travelers and niche segments than for the mass market.

In contrast to the ‘pleasure periphery,’ the later arrival and slower progression along the resort cycle of these Pacific, Indian and African islands have derived principally from their remoteness.  This isolation, in combination with small-scale facilities, has hindered the growth of direct air connections with metropolitan origin markets while the resulting limited passenger traffic has further curbed expansion of hotel room capacity (Kissling, 1989).  Many also share a less extensive colonial and commercial history with the west and low levels of socio-economic modernization.  Because of the alternative employment afforded by their diversified economies, policy-makers by and large have been less aggressive in promoting mass tourism, given air access and infrastructure constraints, and more supportive of selective nature and culture tourism styles, the so-called “South Pacific Way” (Yacoumis, 1989).  Finally, in some cases (Comoros, Fiji), recent political instability has retarded development.

The intermediate destinations represent the largest and most dynamic group.  Their average TPI scores fall cleanly between the most and least developed.  For example, average per capita visitor spending is approximately $2,000, average daily density is 55 visitors per 1,000 population, and average rooms per km2 is six.  In most cases, these islands are characterized by very rapid visitor growth and hotel and infrastructure construction.  In contrast to the most developed destinations, they tend to have higher rates of seasonality and lower levels of promotional spending and cruise ship traffic.  In terms of lifecycle progression, they comprise a mix of 22 islands of varying size and economic structure that display increasing tourism scale, ecosystem impact and international visibility.  All have fairly considerable tourism experience but they are marked by a diversity of tourism styles.  At the high end are traditional mass market destinations like Antigua, Bahamas, Bahrain, Barbados and Cyprus.  In fact, Bahamas is more accurately a mature resort because of the high-density concentration of activity in the Freeport-Nassau complex, but its intermediate TPI score results from the archipelago’s large land area and the low level of tourist development across the outer “Family Islands.”

The intermediates also include a large number of mid-size small islands with smaller tourism sectors inside more diversified economies: in the Caribbean, Dominica, Grenada, St. Lucia and St. Vincent in transition from preferential dependence on the shrinking EU banana market plus St. Kitts/Nevis (sugar) and Curacao (petroleum); in the Indian Ocean, Seychelles (tuna) and Maldives (fishing); and the French overseas dependencies of Polynesia (farming) and Guadeloupe and Martinique, replacing sugar with banana exports to France.  In recent decades, Dominica (St. Vincent to a lesser extent) has become recognized as an ecotourism destination “marketing its mountains, forests and lack of development as tourism assets rather than liabilities” (Weaver, 2001:168).  Finally, they include a handful of the least populated islands in the sample: Bonaire and Palau, famed dive destinations; tiny Niue and Cook Islands, MIRAB-like societies heavily dependent on remittances and subsidies from New Zealand; and Montserrat, a popular North American retirement haven since the 1960s whose TPI ranking has declined because of a devastating volcanic eruption in 1995 that rendered over half the island uninhabitable.

Anguilla and Antigua at the top of the group may graduate to most developed high-density status in the next two decades.  Recent performance suggests a slow-down that may be a harbinger of saturation and/or declining popularity.  Between 1993 and 2000, growth was fluctuating and essentially flat in stayovers and visitor expenditure in Anguilla, and in stayovers and hotel rooms in Antigua (WTO, 1999, 2002).  However, over the same period, many other intermediate destinations experienced robust growth, the hallmark of this stage in the lifecycle.  For example, stayovers and expenditure roughly doubled in Maldives while cruise traffic nearly tripled in St. Lucia and doubled in Dominica and St. Vincent.  Stayovers increased 40 percent in both Guadaloupe and Martinique and expenditure doubled in Polynesia.  In addition, rapid hotel room growth marked the environment in a number of islands: over 50 percent in St. Lucia, Maldives and Palau, and between 30 and 50 percent in Dominica, Grenada, Martinique, St. Lucia and Seychelles.  Many destinations undergoing such noticeable changes are experiencing resource-use conflicts and planning challenges as land, labor and capital migrate from traditional pursuits to tourism.


In broadbrush, the TPI presents a global picture of tourism development in small islands across the world.  Because of its aggregative nature and other limitations, its individual rankings may be less important than the clustering of destinations at the low, intermediate and high end of the lifecycle.  The results are useful as a first approximation for policy-makers in at least two related ways: (1) as an indirect early warning signal particularly for those destinations experiencing or approaching high-density and potentially nonsustainable development, (2) and, with Butler’s framework as a backdrop, as revealing a menu of the principal planning challenges that surface along the three major stages of the cycle.

For the most developed mature destinations, for example, the key task is to sustain vacation quality.  This will require at least three major policy directions: (1) restoring environmental damage and curbing further incursions into fragile areas; (2) managing visitor densities less intrusively by dispersal through time and space; (3) and expanding length of stay and visitor quality and by developing smaller-scale specialty alternatives to mass tourism: heritage, scientific, nature, retirement, village and so on.  Bermuda represents such a success story.   The stagnation of Bermudian tourism in the mid-1980’s plus citizen complaints of visitor saturation prompted widespread public discussion and long-term reassessment.  Residents favored retaining the Island’s upscale image and protecting its unique biodiversity and heritage.  Policy-makers responded with ceilings on bed capacity, vehicles and cruise ships as well as specific controls on construction design and landscaping.  Natural amenities and historical architecture remained intact while tourism stabilized in the 1990’s somewhat below 1980’s levels (McElroy, 2001).  Five elements played a role in this case: a tradition of environmental conservation, widespread community awareness and participation, strong destination identity, policy commitment to the long term, and a robust offshore financial sector (banking, insurance, ship registry) to absorb any tourism declines.

The key challenge facing many intermediate destinations is controlling the quantity and quality of growth.  During this phase of increasing integration with the global tourist economy, these islands must attempt to constrain the natural propensities of international air, hotel and cruise interests for large-scale facilities and high-volume visitation so that these growth imperatives do not exceed the insular economic (labor, utilities, etc.) and socio-environmental absorptive capacities.  This will require at least three major initiatives: (1) preventing further encroachment on renewable resource uses (agriculture, fisheries); (2) sequencing large developments in stages over long horizons; and (3) engaging residents not only in participatory decision-making, but also providing them with a stronger financial stake in the industry.  This last can be accomplished by tax and other incentives for small-scale, local labor-intensive enterprises and for local purchases by hoteliers/restaurateurs/developers.  Targeting inter-island regional tourists is also warranted since such visitors support smaller-scale local service suppliers and “tend to travel to more geographically dispersed areas of the destination country, thereby facilitating wider distribution of the income from tourism” (Sinclair and Voker, 1993: 213).  Growth can be further limited at critical habitats and stressed sites by raising access fees to also secure funding for biodiversity conservation and management (Lindberg, 1991).

The Seychelles represents a good example of a mid-stressed destination managing tourism through localization and ecotourism in a macroeconomic context of falling copra prices and fish catches (Shah, 2002).  For three decades the government has managed Cousin Island, a defunct coconut plantation converted to a wildlife refuge for Hawksbill turtles, Seychelles Warblers and other rare terrestrial and marine species.  Ten percent of all Seychelle visitors take day-trip adventures to the site in local vessels.  Fees finance conservation of the Island and also support scientific tourism and environmental educational programs.

            Maldives provides another illustration of an intermediate destination accommodating tourism growth in an unfavorable macroeconomic context: the collapse of its fresh fish export market to frozen fish and the closure of a major UK air base.  Over the past two decades, tourism has been principally responsible for directly expanding employment and stimulating indirect activity in building materials, handicrafts and shell souvenirs, and other cottage industries.  To manage this growth, the government has crafted an environmentally integrated and socially segregated tourism style whereby small uninhabited islands across the sprawling archipelago are leased to foreign investors to construct self-contained, high-quality, low-density resorts.  Strict building and operational codes serve to preserve the natural patrimony and to attract affluent visitors from Europe and Japan (Domroes, 1999).  Developers are also required to provide their own utilities, waste disposal and staff accommodations.  In order to minimize western influence on the local Muslim population, the government also regulates visitor activities, i.e. prohibiting nude bathing as well as overnight stays on inhabited islands.  The outcome of all such planning is that Maldives has become one of the best water sport and underwater diving destinations in the world (Cockerell, 1995).

In contrast to the Seychelles and Maldives, noted for their long-range planning to control the pace of development (Innskeep, 1994), Antigua provides an intermediate case of nonsustainable growth.  Centuries of deforestation for sugar culture presaged the post-war growth of mass tourism.  Between 1975-1980 total visitation doubled, and doubled again between 1980-1990.  During these years more mangrove swamps and offshore reefs were damaged or killed than in all previous island history (Coram, 1993), through intense coastal commercial colonization.  Subsequent to these asset losses, overnight visitors have still not recovered from their early 1990 levels (WTO, 2002).  This policy failure derives from the absence of the five ingredients illustrated in the Bermudian case: (1) a legacy of environmental neglect, (2) the absence of a clear destination identity anchored to the “native genius of the place,” (3) the poor performance of non-tourist diversification into manufacturing and domestic agriculture, (4)marginal citizen and NGO decision-making participation, and (5) the political directorate’s persistent preference for short-term economic gain over long-term sustainable natural resource planning and management.

For the least developed destinations, the key challenge is to establish international visibility.  Ideally, this requires a major community-wide planning effort to achieve three things: (1) to identify the islands’ unique assets/attractions, (2) to construct the transport and facility infrastructure for sustainably accessing these assets, and (3) to determine a destination identity compatible with the native natural and cultural “genius of the place.”  This is a formidable task, but fortunately these islands have ample policy room to maneuver because of their early position in the resort cycle.  They have sufficient time to develop a tourism style that is socially acceptable, environmentally compatible and economically viable, and there are a number of successful models of integrated planning available (Manning and Dougherty, 1999).

These destinations also possess some of the best opportunities for designing more sustainable alternatives to mass tourism.  Although recent growth has elevated it to intermediate status, Dominica is one of the best examples of comprehensively planned ecotourism.  Based on its unique natural assets (riverine ecology, spectacular falls, black-sand beaches), the government is fostering small-scale local tourism emphasizing “the forested interior, volcanic phenomena, the remnant of Carib Indian culture of the east coast and colonial historical sites” (Weaver, 1998: 194).  Other ecotourism possibilities Weaver identifies for low-density Pacific islands include hiking trails in Samoa, village tourism in the Solomons, and controlled diving activity in the Enipein Marine Park on Pohnpei in the Marshall Islands.


This analysis applied the Tourism Penetration Index to 51 small islands across the world to broadly assess tourism’s overall pressure on fragile insular societies and ecosystems.  In the context of an abbreviated three-stage version of Butler’s lifecycle model, the TPI classified islands into least, intermediate and most penetrated destinations.  In so doing it provided an early warning signal for resort areas crossing the threshold to the high-density stage and possibly nonsustainable tourism development.  As a highly aggregative rudimentary tool, however, the index suffers from a number of limitations, in particular the failure to capture seasonal variations as well as the geographic concentration of visitor flows in dualistic and archipelagic countries.  A first order of follow-up research should address these and other deficiencies to improve the TPI’s performance and reliability.

A second potentially promising avenue for further study concerns the influence of political status, i.e. the character of jurisdiction, on island tourism development.  Although non-sovereign jurisdictions make up 45 percent (23/51) of the total sample in this study, they are markedly over-represented among the most tourism developed and clearly under-represented among the least penetrated.  To illustrate, 12 of the 13 most developed destinations with the highest TPI scores are non-sovereign dependencies or states (Hawaii), while only two of the 16 least tourist-penetrated are dependencies (New Caledonia and Reunion).  Such evidence though impressionistic suggests that non-sovereign political status may confer particular advantages for tourism growth.  These could include geographic proximity to and ease of travel (no passports, same currency) from major mother country origin markets, ready access to investment capital and aid-financed transport and communications infrastructure, special tax and duty-free concessions for gift/liquor purchases as well as other favorable advantages that deserve further examination.



Agarwal, S.  (1994).  The resort cycle revisited: Implications for resorts.  In C.P. Cooper

and A. Lockwood (Eds.), Progress in Tourism, Recreation and Hospitality

Management (vol. 5) (pp. 194-208).  Chichester, UK: Wiley.


Baldacchino, G. & Milne, D.   (2000).  Lessons from the Political Economy of Small

Islands: The Resourcefulness of Jurisdiction.  New York, USA: St. Martin’s Press.


Beekhuis, J.  (1981).  Tourism in the Caribbean: impact on the economic, social and

natural environment.  Ambio  10(6): 325-331.


Beller, W., d’Ayala, P.,  & Hein, P. (Eds.).   (1990).  Sustainable Development and

Environmental Management of Small Islands.  Paris, FR: UNESCO-Parthenon.


Benhamou, F.  (1993).  L’ile Maurice: Un tourisme Ltarmonieuz mais fragile.  espace 

121: 42-46.


Bertram, G. and Watters, R.F.  (1986).  The MIRAB process: Earlier analysis in context. 

Pacific Viewpoint  27(1): 47-59.


Briguglio, L., Archer, B., Jafari, J. & Wall, G. (Eds.).   (1996).  Sustainable Tourism in

Islands and Small States: Issues and Policies.  London, UK: Pinter.


Briguglio, L.  (1995).  Small island developing states and their economic vulnerabilities. 

In G. Paoletto and R. Kuhr (Eds.), International Symposium on Small Islands and

Sustainable Development (pp. 7-51).  Tokyo: United Nations University.


Bryant, D., Nielson, D. & Tangley, L.   (1998).  The Last Frontier Forests: Ecosystems

and Economies on the Edge.  Washington, DC: World Resource Center.


Butler, R.W.  (1991).  Tourism, environment, and sustainable development. 

Environmental Conservation.  18(3): 201-209.


Butler, R. W.  (1980).  The concept of a tourist area cycle of evolution: implications for

management of resources.  Canadian Geographer 24: 5-12.


Caribbean Tourism Organization.  (2001, 1993).  Caribbean Tourism Statistical Report. 

Christ Church, Barbados: CTO.


Choy, D.J.L.  (1992).  Lifecycle models for Pacific island destinations.  Journal of Travel

Research  30(3): 26-31.


Cockerell, N.  (1995).  Maldives.  International Tourism Reports No. 2, pp. 25-43. 

London, UK: Economic Intelligence Unit.


Commonwealth Secretariat.  (1997).  A Future for Small States: Overcoming

Vulnerability.  London: CS.


Coram, R.  (1993).  Caribbean Time Bomb.  New York, USA: William Morrow.


Dann, G., & Cohen, E.  (1991).  Sociology and Tourism.  Annals of Tourism Research. 

18(2): 155-169.


de Albuquerque, K. & McElroy, J.L.  (1992).  Caribbean small-island tourism styles and

sustainable strategies.  Environmental Management  16: 619-632.


Domroes, M.  (1999).  Tourism in the Maldives: The resort concept and tourist-related

services.  Insula  8(3): 7-14.


Getz, D.  (1992).  Tourism planning and the destination lifecycle.  Annals of Tourism

Research.  19(4): 752-770.


Heywood, K.M.  (1986).  Can the tourist-area lifecycle be made operational?  Tourism

Management.  7(3): 154-167.


Holder, J.  (1988).  Pattern and impact of tourism on the environment of the Caribbean. 

Tourism Management 9(2): 119-127.


Hovinen, G.  (2002).  Revisiting the destination lifecycle model.  Annals of Tourism

Research  29(1): 209-230.


Innskeep, E.  (1994).  National and Regional Tourism Planning: Methodologies and Case

            Studies.  London, UK: Routlege.


Jenner, P., & Smith, C.  (1992).  The Tourism Industry and the Environment.  London,

UK: Economist Intelligence Unit.    


Kissling, C.  (1989).  International tourism and civil aviation in the South Pacific: Issues

and innovations.  Geojournal  19(3): 309-315.


Lanfant, M.F., Allcock, J.B., & Bruner, E.M.   (1995).  International Tourism: Identity

and Change.  London, UK: Sage Publications.


Larson, C.  (2003).  Group ranks VI national park among U.S.’ 10 most endangered.  The

Virgin Islands Daily News  January 15: 1. 





Lindberg, K.  (1991).  Policies for Maximizing Nature Tourism’s Ecological and

Economic Benefits.  Washington, DC: World Resources Institute.


Lindberg, K., & Hawkins, D.E. (Eds.).   (1993).  Ecotourism: A Guide for Planners and

Managers.  North Bennington, VT: Ecotourism Society.


Liu, Z.H., & Jenkins, C.J.  (1996).  Country size and tourism development: A cross-

nation analysis.  In L. Briguglio, B. Archer, J. Jafari and G. Wall (Eds.),

Sustainable Tourism in Islands and Small States (pp. 90-117).  London, UK:



Lobban, C.S., & Schefter, M.   (1997).  Tropical Pacific Island Environments.  Mangilao,

GM: University of Guam Press.


Manning, E.W., & Dougherty, T.D.   (1999).  In T.V. Singh, & S. Singh (Eds.),

Tourism Development in Critical Environments (pp. 1-20).  New York, USA: Cognizant

Communication Corporation.


McElroy, J.L.  (2002).  The Impact of tourism small islands: a global comparison.  In F.

DiCastri, & V. Balaji (Eds.), Tourism, Biodiversity and Information (pp. 151-167).  Leiden, Neth: Backhuys Publishers.


McElroy, J.L.   (2001).  Island tourism: a development strategy for biodiversity.  Insula

            10 (2): 21-22.


McElroy, J.L.,  & Mahoney, M.   (2000).  The propensity for political dependence in

island microstates.  Insula 9 (1): 32-35.


McElroy, J.L. & de Albuquerque, K.  (1999).  Measuring tourism penetration in small

islands.  Pacific Tourism Review  3(2): 161-169.


McElroy, J.L.,  & de Albuquerque, K.   (1998).  Tourism penetration index in small

Caribbean Islands.  Annals of Tourism Research  25 (1): 145-168.


McElroy, J.L., & de Albuquerque, K.   (1992).  An Integrated Sustainable Ecotourism for

            Small Caribbean Islands.  Bloomington IN:  Indiana University, Center for Global

            Change and World Peace, Occasional Paper No. 8, Series on Environment and



McElroy, J.L. & de Albuquerque, K.  (1994).  Island tourist profiles across the

destination life cycle.  Insula  3(1): 12-18.




McElroy, J.L., de Albuquerque, K. & Dioguardi, A.  (1993).  Applying the tourist area

destination life cycle model to small Caribbean and Pacific islands.  World Travel

and Tourism Review 3: 236-244.



McElroy, J.L., & Olazarri, P.  (1997).  A tourism penetration index for small island

            destinations.  Bank of Valletta Review 16 (Autumn): 1-10.


Minerbi, L.  (1988).  Alternative Forms of Tourism in the Coastal Zone: Searching for

Responsible Tourism in Hawaii Honolulu, HI: University of Hawaii, Dept. of \

Urban and Regional Planning. 


Pearce, D.  (1989).  Tourist Development.  Harlow, UK: Longman Scientific and

Technical Publishers.


Shah, N.   (2002).  Bikinis and biodiversity: tourism and conservation on Cousin Island.

In F. diCastri, & V. Balaji (Eds.), Tourism, Biodiversity and Information (pp. 185-196).  Leiden, Neth: Backhuys Publishers.


Sinclair, M.T., & Voker, R.W.A.   (1993).  The economics of tourism in Asia and the

Pacific.  In M. Hitchcock, V. King,  & M. Parnwell (Eds.), Tourism

in South-East Asia (pp. 200-213).  London, UK: Routledge.


Towle, F.  (1985).  The island microcosm.  In J.R. Clark (Ed.), Coastal Resources

Management: Development Case Studies (pp. 589-749).  Washington, DC:

National Park Service.


Turner, L., & Ash, J.   (1976).  The Golden Hordes: International Tourism and the

Pleasure Periphery.  New York, USA: St. Martin’s Press.


United Nations Environment Programme.  (1999a).  Environmental Outlooks for the

IslandsCountries of the Caribbean, Indian Ocean and South Pacific.  Nairobi,

Kenya: UNEP.  <http://www.unepch/earth/sidsgeo.htm>


United Nations Environment Programme.  (1999b).  Global Environmental Outlook

2000.  London, UK:  Earthscan Publications.  <>


United Nations Environment Programme.  (2002).  GEO-3: Outlook for 2002-2003. 

Nairobi, Kenya.  <>


United States Congress.  (1992).  Science and Technology Issues in Coastal Ecotourism. 

Washington, DC: G.P.O., Office of Technology Assessment.


Weaver, D.B.  (2001).  Mass tourism and alternative tourism in the Caribbean.  In D.

Harrison (Ed.), Tourism and the Less Developed World: Issues and Case Studies

(pp. 161-174).  New York, USA: CABI Publishing.


Weaver, D.B.  (1998).  Ecotourism in the Less Developed World.  New York, USA:

CABI Publishing.


World Travel and Tourism Council.  (2001).  Year 2001:  Tourism Satellite Accounting

Research.  London, UK:  WTTC. <>


World Tourism Organization.  (2001, 1999).  Compendium of Tourism Statistics. 

Madrid, Spain: WTO.


Yacoumis, J.  (1989).  South Pacific tourism promotion: A regional approach.  Tourism

Management  (March): 15-28.