Jerome L. McElroy

Professor of Economics

Fellow, Center for Women’s Intercultural Leadership

Saint Mary’s College

Notre Dame, Indiana USA






Kate Sandborn

2004 Graduate

Department of Business Administration and Economics

Saint Mary’s College

Notre Dame, Indiana 46556











Abstract.  Why has the postwar march to independence stalled among small tropical islands?  Why do dependent islands continue to vote for the status quo?  The primary explanation in the literature is the substantial economic benefits conferred by political affiliation: preferential metropolitan trade, investment and migration opportunities and subsidized infrastructure funding.  This study compares 16 dependent with 19 independent islands in the Caribbean and Pacific across 25 socio-economic and demographic indicators.  The former significantly out-perform their larger sovereign rivals across most indices.  Results suggest the dependencies have more successfully restructured their colonial economies, have progressed further along the demographic transition, and comprise a new insular development case: the small service-driven dependent island economy.



























Since the founding of the United Nations in 1945, over 80 former colonies have achieved independence.  Many have been warm-water islands situated in all major oceanic basins.  A sampling includes Jamaica and Barbados (Caribbean), Malta and Cyprus (Mediterranean), Seychelles and Maldives (Indian Ocean), and Fiji, Solomon Islands and Vanuatu (Pacific).  However, the pace of decolonization has slowed considerably in recent decades.  Since 1983 only four countries have been admitted to the UN as independent states: three former U.S. Trust Territories in the Pacific—Micronesia, Marshall Islands and Palau—and most recently (2002) East Timor (Timor Leste).  Dozens of small, mostly island dependencies, with varying degrees of sub-sovereign jurisdiction, remain scattered across the globe content with the political status quo.  This stalling of the independence juggernaut has prompted the UN to declare the 1990s as “The Decade for the Eradication of Colonialism,” and to call for a repeat performance for 2001-2010 (UN, 2000).  What lies behind this insular propensity for dependence?



Despite the heterogeneity of these non-sovereign islands in size, history, geography and constitutional arrangements, there are some common political, cultural and economic reasons identified in the literature that apply in varying degrees across the spectrum.  For example, in the metropoles after the first wave of independence during the postwar generation, the pressure to decolonize subsided.  Several factors were responsible.  First, in some cases metropolitan policy either neglected the territories because of more pressing matters or was inconsistent and/or lacked the flexibility for devising innovative solutions (Aldrich and Connell, 1997).  Second, with the demise of the cold war, the dependent islands lost much of their strategic geopolitical value (Ramos and Rivera, 2001).  Third, particularly since the escalation of global terrorist attacks, metropolitan policy shifted away from status concerns toward enlisting the territories in the fight against drug traffic, money laundering and illegal migration (Lamp, 2001).  Fourth, the metropolitan powers became increasingly willing “to respect the wishes of the electorate(s) of the dependent territories on constitutional matters” (Hintjens, 1997: 536), and islanders by and large have repeatedly opted to retain colonial ties.


There are many reasons why the status quo is appealing to islanders.  First has been the UN recognition that both free association with or integration into another sovereign state “or the emergence into any other political status freely determined by a people” (UN, 1970: 123). constitute, along with independence, legitimate avenues to self-determination (read decolonization).  Second, many non-sovereign islands are sufficiently satisfied with the jurisdictional autonomy already achieved over local finance, taxation, natural resources and other matters (Baldacchino and Milne, 2000).  Third, in some islands a generation of labor-intensive tourism growth and accelerating globalization have produced immigration pressures which have threatened “the definition and distinction of island identities” (Connell, 2001: 48).  As a result, preserving local culture has become more prominent in political discourse that status concerns (Daniel, 2001).  Other dependencies have had difficulty finding internal consensus on a way forward and/or squandered energy on domestic squabbles (Giacalone, 2001).


However, the most important reason for continuing dependence is the widely held insular perception that going independent in a globalized world would damage the territories’ substantial economic privileges.  According to McElroy and Mahoney (2000: 32), these include: “free trade and export preferences for island produce and manufactures, acess to lucrative metropolitan capital and labour markets, grants and welfare assistance, the subsidized provision of quality infrastructure, external defense and disaster relief.”  A host of additional advantages that apply variously include: (1) access to metropolitan citizenship, (2) minimum wage and health care center-periphery parity in the Francophone islands (Miles, 2001), (2) “flexible finance, environmental and commercial registration regulations” (Armstrong and Read, 2005: 11), and (4) the political stability and metropolitan oversight/security that generate widespread investor confidence.  Many of these benefits originally established to compensate for small insular size and geographic remoteness would be swept away by sovereignty.


According to Rivera (2001: 161), “The perception that the economies of the non-independent countries are more ‘modern’ and prosperous than those of most of the independent countries . . .” is based on fact.  A large stream of recent literature supports the claim.  For example, Poirine (1998) found per capita income in non-sovereign islands double their sovereign counterparts primarily, he argues, because of their significantly higher levels of aid.  Using a large 105 country sample of small countries (less than 3 million population), Armstrong and Read (2000) found similar superior performance recorded for dependencies, 90 percent of which were islands.  Other explanatory variables included differences in economic structure—services (tourism) and finance (offshore) in preference to agriculture—natural resource availability, and geographic location.  McElroy and Mahoney (2000) described major socio-economic and demographic differences favoring dependencies in separate samples of Caribbean and Pacific islands.  Finally, using a worldwide sample, Bertram (2004) advanced the argument that superior non-sovereign per capita income and growth were a direct function of the intensity of political dependence (and the assumed closeness of center-periphery trade, aid and investment linkages) and the economic performance of the metropolitan patron.  Given such strong evidence, he concluded (2004: 353): “. . . there is no clear incentive for presently dependent island territories to see independence, and good grounds for them to hold on to the status quo.”


Scope and Method

The present study focuses on small, warm-water islands in the Caribbean and Pacific.  Its purpose is twofold: (1) to empirically explore and define social, demographic as well as economic differences between non-sovereign islands and their sovereign counterparts, and (2) thus to provide a more comprehensive justification for the propensity for dependence.  It extends an earlier analysis (McElroy and Mahoney, 2000) in three ways: (1) the exclusive focus on small islands less than 3 million population, (2) the use of roughly double the number of independent variables, and (3) the use of statistical inference to test island differences.


Thirty-five islands were selected based on the small size criterion and the availability of nearly complete published data.  They comprised 16 dependent and 19 independent countries.  From the Caribbean they included the following non-sovereign states: Anguilla, Aruba, Bermuda, British Virgin Islands (BVI), Cayman Islands, Guadeloupe, Martinique, Montserrat, Turks/Caicos, United States Virgin Islands (USVI) and Netherlands Antilles.  This last is an aggregation of Bonaire, Curacao, St. Maarten, Saba and St. Eustatius.  The sovereign Caribbean states were: Antigua/Barbuda, Bahamas, Barbados, Dominica, Grenada, Jamaica, St. Kitts/Nevis, St. Lucia, St. Vincent and the Grenadines, and Trinidad/Tobago.  In the Pacific, the dependents were American Samoa, French Polynesia, Guam, New Caledonia and Northern Mariana Islands, while the independents included: Fiji, Kiribati, Marshall Islands, Micronesia, Palau, Samoa, Solomon Islands, Tonga and Vanuatu.  These binary status classifications were based on UN definitions.  While some studies (Bertram, 2004; McElroy and Mahoney, 2000) have treated the U.S. Associated States of Marshall Islands, Micronesia and Palau as non-sovereign, according to Corbin (2001: 139) they have achieved “sufficient sovereignty for acceptance as full members in the UN General Assembly . . .”


Twenty-five variables were selected to test for distinct socio-economic and demographic differences.  To measure economic behavior and structure, ten variables were chosen.  Eight were macroeconomic indicators: per capita income and electricity consumption, the unemployment and labor force participation rates, the distribution of GDP into agriculture, industry and services, and land area, a proxy for resource availability.  Because of the significance of tourism in tropical islands, two tourism measures were used: the ratio of stayover tourists [excluding one-day (cruise passenger etc.) visitors] to the resident population, and per resident visitor expenditure.  Four standard social/health indicators were employed: life expectancy, adult literacy, infant mortality and the number of phones per 1,000 population.  Eleven demographic measures were used: population size, growth and density; distribution into young (0-14 yrs.), working age (15-64 yrs.) and old (65+ yrs.) cohorts; median age, and birth, death, net migration and fertility rates.


All except the tourism data were taken from the World Factbook (CIA, 2004).  The tourism data came from the Compendium of Tourism Statistics (WTO, 2004).  To develop distinct statistical profiles of the dependent versus the independent islands, average values were calculated for each group, and statistical differences were determined across the 25 variables using a two-sample means test.  Given past island literature (Caldwell et al., 1980; Dommen and Hein, 1985; Beller et al., 1990) and the recent research reviewed above, it was hypothesized that, as a group, non-sovereign islands in comparison with their sovereign island neighbors would exhibit: (1) higher levels of economic performance, (2) a stronger service orientation and tourism intensity, (3) and greater social progress in terms of higher life expectancy and literacy and lower infant mortality.  Additionally it was assumed the more advanced dependents would demonstrate greater demographic maturity as evidenced by lower fertility, natality and population growth rates.



Table 1 presents the basic data for the 35 islands.  Table 2 displays average values for the two island groups classified by political status across the 25 indicators.  It also presents results of the two-sample means test.  These outcomes broadly suggest statistically distinct socio-economic and demographic profiles that basically conform to the differences hypothesized above.  The non-sovereign countries differ markedly in size, structure and behavior from their sovereign counterparts.  For example, in terms of basic resource availability, they average only a fourth and a third respectively of sovereign island land area and population.  However, the lack of statistical difference tends to support the view of Armstrong and Read (2005) that size has no major influence on island behavior.


On the other hand, the dependents’ strong economic performance compensates for their relative resource scarcity.  They average between 2-3 times higher levels of per capita income, $16,381 versus $6,145, analogous to the findings of Poirine (1998).  Similarly dependents average 2-3 times higher levels of per capita electricity consumption, 4,246 to 1,537 kwh, a common proxy variable for income and a standard of living indicator.  The labor force participation rate (LFPR), those employed and unemployed divided by the population, is a macroeconomic indicator of overall labor utilization.  As expected, it is higher in the dependencies, 45 to 43 percent, but the different is insignificant.  Likewise, the average non-sovereign unemployment rate is lower, roughly 11 versus 14 percent, but again the difference is not statistically significant.  Thus, the labor variables do not discriminate between the profiles and may suggest the strength of other similar but undefined/measured economic forces operating across the insular spectrum.

(Tables 1 and 2 about here)


It is a different story with economic structure.  The dependent islands average a much lower ratio of primary production and a much higher ratio of tertiary output.  For example, in non-sovereign countries agriculture accounts for less than 5 percent of GDP while it generates nearly 16 percent in sovereign countries.  On the other hand, services absorb approximately 80 percent of GDP in dependents in contrast to 65 percent in independents.  These statistically significant differences suggest the former  have progressed further than the latter in restructuring their colonial economies away from traditional low value-added staples like sugar and copra toward more income-elastic tourism and offshore finance.  These results parallel the findings of Armstrong and Read (2000) but go a step further and specifically point toward the important economic impact of tourism in the dependent islands.  To illustrate, tourists spend an average of nearly 6 times more per resident in the non-sovereign as in the sovereign islands, i.e. $6,740 versus $1,174.  In addition, the ratio of tourists (stayover visitors) to resident population is four times higher in the dependencies than in their independent counterparts.  In short, the island territories are considerably more tourism penetrated.


Such evidence indirectly indicates that—in addition to tropical amenities and favorable location—non-sovereign islands have more effectively implemented an endogenous policy of tourism development.  They have thereby hitched their fortunes to the fastest growing industry in the postwar world and the largest in the global economy.  McElroy and Morris (2002) identified similar differences in tourism intensity to explain part of the superior performance of African islands against their mainland counterparts.  The results also underline Bertram’s (2004) thesis on the significance of metropolitan linkages since it is demand in the main tourist origin markets in North America, Europe and Japan that fuels tourism growth in the Caribbean and Pacific peripheries.  Finally, these results support McElroy’s (2005) contention that small, tourism-driven, dependent islands represent a special development case distinct from the widely discussed MIRAB model (migration, remittances, aid, bureaucracy)(Bertram and Watters, 1985).  This conclusion is based in part on the subsidized transport and communications infrastructure and metropolitan ease of travel (common language, currency, customs) facilitated by their affiliated political status.


Social variables also discriminate the two island profiles.  Given their higher affluence and the income elasticity of medical services, the dependents average significantly higher life expectancy (77 yrs.) than their independent counterparts (70 yrs.).  Likewise and partly stemming from their higher population densities and assumed greater health care access, the non-sovereigns experience significantly lower infant mortality, i.e. 10 deaths per 1,000 live births versus 23 for the poorer sovereign countries.  The wealthier dependents also exhibit higher average adult literacy rates than the independents, 97 to 91 percent, although the difference lies only at the 10 percent level of statistical significance.  Taken together, all such evidence indirectly suggests the closer metropolitan ties of the dependencies may foster access to superior health care and education in the dependencies.


Not surprisingly, Table 2 also displays different demographic profiles.  For example, the dependencies clearly have an older age structure with a significantly smaller share of youth (0-14 yrs.), 25 versus 31 percent, and significantly larger shares of working age (15-64 yrs.) and old (65+ yrs.) cohorts, respectively, 68 and 64 percent and 7.4 and 5.2 percent.  Because of their stronger economies and levels of affluence, they possess somewhat larger labor forces and retiree segments.  As a result, median age in the non-sovereign islands (31 yrs.) is roughly 6 years older than in sovereign islands (25 yrs.).


The dependents also exhibit greater progress along the demographic transition from high to low birth and death rates that all modernizing societies pass through.  To illustrate, the non-sovereign islands average significantly lower crude birth rates, 17 versus 21 per 1,000 population for the independents.  This is partly a function of their relative affluence.  Likewise, their total fertility rate averages significantly fewer children (2.1 vs. 2.7) per women of child-bearing age.  Such behavior is a function of their greater socio-economic modernization, better health care and lower infant mortality.  They also exhibit significantly lower crude death rates, 5.4 versus 6.7 per 1,000 population, partly due to their older age structure and perhaps also associated with their 50 percent higher population density (261 vs. 173 persons/km2).  Density in small islands is a surrogate indicator of greater urbanization and medical access (McElroy and de Albuquerque, 1995).


Finally, the dependents demonstrate a lower average rate of natural increase (birth minus death rates) but a higher population growth rate than their sovereign counterparts.  This derives principally from their very different net migration rates.  On the one hand, the non-sovereign islands average an immigration rate of 2.25 persons per 1,000 population.  This inflow most likely of working age migrants is due to their expanding economies based on tourism, related construction and offshore finance, all relatively labor-intensive industries.  As a result, such islands have passed through the so-called migration transition (McElroy and de Albuquerque, 1988) whereby former labor exporters become labor importers.  On the other hand, the sovereign islands as a group show an average emigration rate of -5.66 persons per 1,000 population, largely a function of their relatively slower-growing economies and chronic labor surplus.  This demographic characteristic of out-migration is perhaps the most telling indicator that discriminates the independents from their more affluent and dynamic dependent neighbors.



One of the anomalies of the postwar world has been the visible slowdown in the movement toward independence, particularly among small island polities.  Although off-hand explanations for this propensity for dependence might suggest political inertia and/or the fear of marginalization in a globalized economy, closer inspection reveals deeper and more complex determinants.  The overwhelming conclusion of this study is that such non-sovereign states have become increasingly cognizant of the substantial socio-economic benefits associated with affiliation, have exercised and expanded their resource of jurisdiction to exploit those advantages, and have successfully carved out a niche in the world system as tourism and offshore finance service providers for metropolitan clients.


The economic linkages afforded by dependent status are significant.  They include: preferential trade, migration and citizenship arrangements, access to metropolitan capital markets and specialized labor expertise, the subsidized provision of key transport and communications infrastructure essential for the success of the two primary engines of insular economic growth—tourism and offshore finance—plus a host of other industry specific concessions and common customs and standards that facilitate commerce.  Because of these visible, concrete, day-to-day benefits, over the past 15 years small-state islanders have repeatedly voted to hold on to the status quo.


A recent stream of island research has confirmed the scientific basis for their persistent choice to retain metropolitan linkages and the favorable benefits of the political economy of dependence.  A variety of authors have found non-sovereign islands significantly superior in income levels and economic growth compared to their sovereign neighbors.  They have also identified the sources of their superiority tied in general to metropolitan economic linkages, and in particular to higher levels of aid and a policy orientation toward services and finance to capture the forces of sustained international tourism growth and the ubiquity of global capital movements.


This study empirically extends that research by detailing the socio-economic and demographic profiles of 25 tropical islands in the Caribbean and Pacific:  16 dependents and 19 independents.  Despite the small sub-samples, the analysis constructs statistically distinct contours that clearly favor the dependent islands.  In addition to their higher levels of per capita income and electricity consumption, islanders in the non-sovereign states live longer, are more literate and experience much lower infant mortality and health care.  Because of their relative affluence, not surprisingly they demonstrate greater demographic maturity in terms of lower fertility and mortality.  According to McElroy and de Albuquerque (1995: 176), their achievement “is all the more remarkable given the fact that a generation ago the performance indicators were decidedly reversed in favor of the larger soon-to-be sovereign islands.”

In summary, in contrast to the independent islands, the dependents have more successfully restructured their economies from income-inelastic colonial staples to income-elastic services, have progressed further along the demographic transition from high to low birth and death rates, and in the process passed through the migration transition from chronic labor exporting emigrant societies to dynamic, labor-importing immigrant societies.  On the heels of these momentous socio-economic and demographic changes, they have come to represent a new, successful, insular development case—the small, service-driven dependent island economy.  For these reasons, their propensity for dependence persists.  



ALDRICH, R. and CONNELL, J. (1997)   The Last Colonies.  Cambridge, UK:

     Cambridge University Press.


ARMSTRONG, H. W. and READ, R. (2005)   “Insularity, Remoteness, Mountains and

     Archipelagos: A Combination of Challenges Facing Small States?”  Asia Pacific

     Viewpoint (forthcoming).


ARMSTRONG, H. W. and READ, R. (2000)  “Comparing the Economic Performance of

     Dependent Territories and Sovereign Micro-States,”  Economic Development and

     Cultural Change, 48: 285-306.


BALDACCHINO, G. and MILNE, D. (eds.)  (2000)  Lessons from the Political Economy

     of Small Islands: The Resource of Jurisdiction.  New York: St. Martin’s Press.


BELLER, W., D’AYALA, P. and HEIN, P.  (1990)  Sustainable Development and

     Environmental Management in Small Islands.  Paris: UNESCO-Parthenon.


BERTRAM, G.  (2004)  “On the Convergence of Small Island Economies with Their

     Metropolitan Patrons,”  World Development 32(2): 343-364.


BERTRAM, G. and WATTERS, R. F.  (1985)  “The MIRAB Economy in South Pacific

     Microstates,”  Pacific Viewpoint, 26(3): 497-519.


CALDWELL, J.C., HARRISON, G.E. and QUIGGIN, P.  (1980)  “The Demography of

     Micro-states,” World Development 8(12): 953-968.


CENTRAL INTELLIGENCE AGENCY  (2004)  The World Factbook.  Washington,

     DC: CIA.


CONNELL, J.  (2001)  “Eternal Empire: Britain’s Caribbean Colonies in the Global

     Arena.”  In Ramos, A. G. and Rivera, A. I., (eds.), Islands at the Crossroads,

     Kingston, JA, Ian Randle Publishers: 115-135.


CORBIN, C.  (2001)  “Direct Participation of Non-Independent Caribbean Countries

     In the United Nations: A Method for Self-Determination.”  In Ramos, A. G. and

     Rivera, A. I. (eds.), Islands at the Crossroads, Kingston, JA, Ian Randle Publishers:



DANIEL, J.  (2001)  “The Construction of Dependency: Economy and Politics in the

     French Antilles.”  In Ramos, A. G. and Rivera, A. I. (eds.), Islands at the Crossroads,

     Kingston, JA, Ian Randle Publishers: 61-79.


DOMMEN, E. and HEIN, P.  (1985)  States, Microstates and Islands.  London: Croom



GIACALONE, R.  (2001)  “The Political Status of Curacao at the End of the Twentieth

     Century.”  In Ramos, A. G. and Rivera, A. I. (eds.), Islands at the Crossroads,

     Kingston, JA, Ian Randle Publishers: 95-105.


HINTJENS, H. M.  (1997)  “Governance Options in Europe’s Caribbean Dependencies.”

     Round Table, 344(October): 533-547.


LAMPE, A.  (2001)  “The Recolonisation of Aruba.”  In Ramos, A. G. and Rivera, A. I.

     (eds.), Islands at the Crossroads, Kingston, JA, Ian Randle Publishers: 106-114.


McELROY, J. L.  (2005)  Small Island Tourist Economies across the Life-Cycle,”

     Asia Pacific Viewpoint (forthcoming).


McELROY, J. L. and MORRIS, L.  (2002)  African Island Development Experiences:

     A Cluster of Models,” Bank of Valletta Review, 26(Autumn): 38-57.


McELROY, J. L. and MAHONEY, M.  (2000)  “The Propensity for Political Dependence

     in Island Microstates,” Insula, 9(1): 32-35.


McELROY, J. L. and De ALBUQUERQUE, K.  (1995)  “The Social and Economic

      Propensity for Political Dependence in the Insular Caribbean,” Social and Economic

     Studies, 44(2/3): 167-193.


McELROY, J. L. and De ALBUQUERQUE, K.  (1988)  “Migration Transition in Small

     Northern and Eastern Caribbean States,” International Migration Review, 22(3):



MILES, W. F. S. (2001)  “Fifty Years of ‘Assimilation’: Assessing France’s Experience

     of Caribbean Decolonisation through Administrative Reform.”  In Ramos, A. G. and

     Rivera, A. I. (eds.), Islands at the Crossroads, Kingston, JA, Ian Randle Publishers:



POIRINE, B.  (1998)  “Should We Hate or Love MIRAB?” The Contemporary

     Pacific, 10(1): 65-105.


RAMOS, A. G. and RIVERA, A. I.  (2001)  Islands at the Crossroads: Politics in the

     Non-Independent Caribbean.  Kingston, JA: Ian Randle Publishers.


RIVERA, A. I.  (2001)  “Rethinking Politics in the Non-Independent Territories.”

     In Ramos, A. G. and Rivera, A. I. (eds.), Islands at the Crossroads, Kingston,

     JA, Ian Randle Publishers: 160-179.


UNITED NATIONS  (2000)  “Second International Decade for the Eradication of

     Colonialism,” Resolution 55/146.  New York: UN General Assembly.  <http://>.


UNITED NATIONS  (1970)  “Declaration on Principles of International Law . . .,”

     Resolution 2625 (XXV).  New York:  UN General Assembly.  <http://daccessdds.un.



WORLD TOURISM ORGANIZATION  (2004)  Compendium of Tourism Statistics:

     Data 1998-2002.  Madrid, SP: WTO.