ABSTRACT A NEW FORM OF INTERNATIONAL MIGRATION: THE UNITED STATES MILITARY RETIREE IN CENTRAL AMERICA By Phillip H. Allman, Jr. The purpose of this study is to: 1) examine a develop— ing United States military retiree migration into Central America; 2) determine where the migrants are concentrating; 3) ascertain why the migrants chose specific nations and sites for settlement; 4) identify the main characteristics of the migrants; 5) explain how the migration is taking place; and 6) point out the consequences of the migration. International retirement migration was investigated by use of several models. A military retirement system model was developed to portray the active military establishment as a training institution and producer of potential foreign migrants. Using the concept of this model, a retirement migration schema was designed as a framework for examining retirement migration in all real world situations. Addi- tional models, a migrant circulation system and an informa— tion exchange system, reflect the specific movement pattern and information exchanges of this migrant group. These models were applied in the Central America case. Phillip H. Allman, Jr. Data were obtained during the 1975-76 academic year by personally administering a seven-page questionnaire to the military retiree population in Central America. During this field survey data were also collected from governmental organizations and other institutions. These data, combined with information gained by personal observation, formed the bases for comparative and interpretive analyses. The study surveys the history of United States military personnel in foreign areas, emphasizing the relation of these movements to their propensity to migrate after retire- ment. It describes and explains the world distribution of military retirees and then focuses on Central America as a new amenity-related retirement area. In Central America only Costa Rica, with 70 per cent of the military retiree population, was found to be impor- tant as a retirement area. Sixty per cent of the retirees are concentrated around San Jose in the Meseta Central re- gion. In the remaining countries, most retirees reside in the primate cities, Guatemala City ranking second to San Jose. There was no evidence of retiree colonization or clustering within the major settlements. Correlation-re- gression analyses revealed that the distance factor accounts for little of the variation in retiree distributions. Examination of physical, economic, social, and politi- cal factors provided a means of differentiating individual nations. Costa Rica had the most advantages for retire- ment living, followed by Guatemala, Nicaragua, El Salvador, Phillip H. Allman, Jr. Honduras, and Belize. Favorable climate and low cost of living were the predominant reasons for migrant selection of Central America. Military retirees have a history of foreign moves; the median for those in Central America was five during active service. Sixty-one per cent had at least one previous re- tirement residence. The median residency period in Central America was 2.5 years, with 78 per cent having lived in the area less than five years. The migrants' median age was 55; all were Caucasian males; 81 per cent were married, with one-half having dependent children and 34 per cent married to native wives; 49 per cent were college graduates; and 88 per cent spoke Spanish. Most of the migrants were fully retired (61 per cent) and their median annual income was $8,900. More commissioned retirees (55 per cent) were found than enlisted, thus accounting for the relatively high med- ian income. . Most migrants had friends, relatives, or other mili- tary retirees in-country to assist them in the move. The latter case suggests a modified chain migration effect. Directional bias was present in the migration streams; 58 per cent of the retirees came from California, Florida, and Texas. Retirees represent imported capital, e.g., in Costa Rica their annual income amounts to $1,000,000. A NEW FORM OF INTERNATIONAL MIGRATION: 'PHE UNITED STATES MILITARY RETIREE IN CENTRAL AMERICA By Phillip H. Allman, Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geography 1977 $107073 © Copyright by Phillip Hugh Allman, Jr. 1977 'To Daisy... for sharing the sacrifices and satisfactions ii ACKNOWLEDGMENTS The preparation of a study of this nature entails personal contact with a wide range of individuals from whom information, advice, and assistance are required. Recogni- tion of all these persons obviously is an impossibility. Consequently, I would like to extend my appreciation to all those individuals who are not specifically named, but who shared in this research experience. This especially applies to the military retiree population of Central Amer- ica, without whose friendliness and cooperation this re- search could not have been completed. I am particularly indebted to Dr. Robert N. Thomas, my graduate committee advisor and dissertation director, who stimulated my interest in Central America and migration theory. His guidance and devotion of much valuable time were instrumental in my completing this task. Other mem- bers of my committee whom I wish to thank for their advice and support are Dr. Clarence L. Vinge and Dr. Clarence W. Minkel of the Department of Geography, Dr. James Zuiches of the Department of Sociology, and Lt. Col. Francis L. Brokaw of the Department of Military Science. The courtesies and assistance extended by many persons representing various organizations and institutions during 111 my year in Central America contributed immeasurably to the data gathering procedure. In Guatemala, those individuals especially helpful were Ing. Jose Lopez Toledo of the Office of Geographic Studies, Ministry of Public Works; Col. (Ret.) James Norman of the American Chamber of Commerce; Mr. Romeo Routhier, Adjutant of the American Legion post; Mr. Alexan- der Gregg of the Inter-American Geodetic Survey; and Sgt. Richard Allen of the U.S. Defense Attache Office. In El Salvador, Honduras, and Nicaragua the U.S. Defense Attaches, the Commanders of the U.S. Military Groups, and the repre— sentatives of the Inter-American Geodetic Survey, along with personnel of the government tourism institutes, pro- vided much valuable information. In Costa Rica, the most significant of the countries surveyed, special appreciation is extended to Col. (Ret.) John J. Caulfield and Cdr. (Ret.) Wade Harding of the Costa Rica Retired Officers Associa- tion; Mr. Floyd (Pete) Rader, Adjutant of the American Le- gion post; Mr. Robert Senter, U.S. representative of the Inter-American Geodetic Survey; Mr. Carlos Lara, Director of the Costa Rican Tourism Institute; and Lt. Col. Robert Trdla and SFC Arthur Martinez of the U.S. Military Group. Special recognition is also given to the latter organiza- tion for allowing me to use their facilities as a base of operations in the San Jose area. Credit is due the Departments of Army, Navy, and Air Force, and the Marine Corps; the Retired Officers iv Association; and the Army Times Publishing Company for much of the general and statistical information on military re- tirees. Mr. John Beach, editor of the Harrison Post, Fort Harrison, Indiana, was also helpful in obtaining current statistics on Army retirees. Mrs. Edith Potts unselfishly gave her time and attention to a reading of the final manu- script. Final appreciation is given to my wife, Daisy, for her understanding and patience throughout my graduate pro- gram; her sharing in the travels and trials of data col- lection in Central America, including the major Guatemala earthquake experience in February 1976; and her secretarial and editorial assistance from beginning to end. TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . xi LIST OF FIGURES . . . . . . . . . . . . . . . . . . xiii Chapter I. INTRODUCTION . . . . . . . . . . . . . . . . 1 Statement of the Problem . 4 Purpose of the Study . . . . 4 Contribution to Theory on Foreign Retirement Migration . . . . . . 5 Usefulness of the Study . . . . . 5 Objectives . . . . . . . . . . . . . . . . 6 Migration System and Military Retiree Distributions . . . . . . . . . . 6 Military Retirees . . . . . . . . . . . 7 Migration Decision . . . . . . . 7 Networks . . . . . . . . . . . . . . . 9 Settlement Patterns . . . . . . . . . . ll Adjustment and Effects . . . . . . . . ll Hypotheses . . . . . . . . . . . . . . . . 12 Methodology . . . . . . . . . . . . . . . 13 Personal Interest in the Subject . . . 13 Selection of the Study Area . . . . . . 15 Data Requirements and Sources . . . . . 17 Data Gathering Techniques . . . . . . . 18 Literature Search . . . . . 19 Preliminary Data Collection . . 19 Questionnaire Preparation . . 22 Field survey 0 O O O O O 0 0 O O O O 23 Data Analysis . . . .i. . . . . . . . . 26 vi Chapter Page II. MILITARY RETIREE CONCENTRATIONS . . . . . . . 28 Retirement Migration System . . . . . . . 28 Military P0pu1ation . . . . . . . . . . 34 Historical Bases of Foreign Exposure . . . . . . . . 35 Recent Foreign Military Commitments . . . . . . . . . 37 Military Veterans . . . . . . . . . 38 Military Reserves . . . . . . . . . 41 Career Personnel . . . . . . . . . . 42 Military Retiree P0pu1ation . . . . . . 48 Employment Versus Full Retirement . 49 Retiree Numbers . . . . . . . . . . 51 World Distribution of Military Retirees . 54 Total Numbers . . . . . . . . . . . . . 55 Total Pensions . . . . . . . . . . . . .61 Significance . . . . . . . . . . . . . 64 United States . . . . . . . . . . . . . . 64 Distribution . . . . . . . . . . . . . 65 Location Choice . . . . . . . . . . . . 68 Foreign . . . . . . . . . . . . . . . . . 72 Distribution . . . . . . . . . . . . . 73 Location Choice . . . . . . . . . . . . 76 Migration Potential . . . . . . . . . . 80 Middle America . . . . . . . . . . . . . . 82 General Distribution . . . . . . . . . 82 Mexico and Panama . . . . . . . . . . . 85 Distribution . . . . . . . . . . . . 85 Historical Perspectives . . . . . . 89 Incentives and Deterrents . . . . . 92 Central America . . . . . . . . . . . . 96 III. FACTORS RELEVANT TO CENTRAL AMERICAN RETIREMENT MIGRATION . . . . . . . . . 100 Place Utility . . . . . . . . . . . . . . 100 Location Choice Theory . . . . . . 101 Applicability in International Retirement Migration . . . . . . 101 vii Chapter Geographic Factors . Environmental Conditions Surface Configuration Climatic Conditions . . . : : Natural Hazards . . . . . . . Accessibility to the United States Effect of Distance . . . . . . Intervening Opportunities Stage Migration . . . . Political Factors Government Policy . Special Retiree Provisions Costa Rica . Other Central American .Nations. Tourist and Resident Provisions Form of Government . . . . . Stability of Government . . . Immigration and Customs Inconveniences . . . . . Socio-Economic Factors . . . . . . Latin Way of Life and Attitudes . Amenities 0 O O O O O O O O O O 0 Cost of Living . . . . . . . . . Economic Opportunities . . IV. MIGRANT CHARACTERISTICS . . . . . . . Character of the Migrants . . . Demographic . . . . . . . Socio-Economic . . . . . . . . . Personal Interests . . . . . . . seleCtivj-ty O O O O O O O O O O O O 0 Age, Sex, and Race . . . . . Language . . . . . . . . . . Retirement Income . . . . . Military Service . . . . . Propensity to Migrate . . . . . . . . Movement During Military Service . Previous Retirement Migrations . . viii Page 103 105 106 110 113 114 115 116 117 118 119 119 124 125 126 127 128 131 132 133 137 138 139 141 141 142 145 153 154 155 155 155 156 157 157 158 Chapter Migration Decision . . . . . Positive Reasons for the Migration Costa Rica . . . . Other Central American Nations . Negative Factors Influencing the Migration . . . . . . . . Role of the Amenities . . Importance of Place Perception Degree of Permanency . Analysis of Area Satisfaction . Cyclic Nature of the Migration Evidence of Seasonality . . . . Effects of Kinship and Friendship Ties . . . . . . . . . V. SPATIAL SIGNIFICANCE OF THE MIGRATION . Migration Channels and Information Flows Migrant Circulation System . . Case Study . . . . . Directional Bias and Chain Migration Information Exchange System . . . Influence of Information Sources Impact: The Case of Costa Rica . Networks . . . . . . . . . . . Transportation Networks . External Transportation Systems Airways . . . . . . Highways . . . . . Seaways . . . . . . . . . Internal Transportation Systems Railways . . . . . . . . . . Waterways . . . . . . . . . . Airways . . . . . . . . . Highways . . . . . . . . Social Networks . . . . . . . . External Contacts Social Activities . . . . Economic Activities . Political Activities ix Page 160 161 162 162 162 166 167 169 169 172 173 175 177 177 177 180 184 186 189 191 193 193 194 194 196 197 197 199 199 201 201 203 204 204 205 206 Chapter Page Internal Contacts . . . . . . . . . 206 Social Activities . . . . . . . . 207 Economic Activities . . . . . . . 212 Political Activities . . . . . . 213 Interrelationships and Connectivity . . 214 Settlement Patterns . . . . . . . . . . . 216 Urban Versus Rural . . . . . . . . . . 216 Primate Cities . . . . . . . 218 Migrant Clustering or Dispersal . . . . 219 Effect of Distance in Costa Rica . . . 224 VI. SUMMARY AND CONCLUSIONS . . . . . . . . . . . 227 Summation . . . . . . . . . . . . . . . . 227 Military Retiree Concentrations . . . . 227 Factors Relevant to Central American Retirement Migration . . . . . . 228 Migrant Characteristics . . . . 232 Spatial Significance of the Migration . 237 Migration Theory . . . . . . . . . . . . . 241 Application to Existing Theory . . . . 241 Development of a Retirement Migration Theory . . . . . . . . 242 Application . . . . . . . . . . . . . . . 243 Socio-Economic Aspects . . . . . . . 244 Planning and Policy Formulation . . . . 245 APPENDICES A. MILITARY RETIREE MIGRATION HISTORY QUESTIONNAIRE . . . . . . . . . . . . . 247 B. U.S. MILITARY STRENGTH OUTSIDE THE UNITED STATES . . . . . . . . . . . 254 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . 255 11. 12. 13. 14. 15. 16. 17. 18. LIST OF TABLES Military Retirees in Central America Composition of the Interviewees . Per cent of Military Non-disability Retirees Who Have Never Been Employed . Military Retiree Strengths FY 1972-1975 . Percentages of Military Retirees in Foreign Areas . Numbers of Military Retirees by Foreign Countries . Numbers of Military Retirees by FY 1975 . . States Military Retirees by Geographic Region . Retired Military Personnel Pensions, Coefficients of Correlation Between Variables (r) Significance of the Beta Values Ranking of Retirees by Country 1973 Per cent of Retirees in Mexico by States Area and Population Comparisons . Average Monthly Temperatures Retirement Laws for Foreign Nationals . Retiree Status in Central America . Per Capita Income . xi Page 23 27 50 52 55 56 60 62 63 70 71 75 88 104 112 120 128 140 Table Page 19. Age at Retirement from the Military . . . . . 142 20. Demographic Composition of the Migrants . . . 144 21. Migrant Socio-Economic Characteristics . . . 146 22. Migrant Expenditure of Time in Average Week . . . . . . . . . . . . 154 23. Previous Foreign Retirement Residences . . . 159 24. Reasons for Migration to Costa Rica . . . . . 163 25. Reasons for Migration to Other Central American Nations . . . . . . . 163 26. Reasons for Leaving Previous Retirement Location . . . . . . . . . . 165 27. Previous Latin America Military Assignments . 169 28. Reasons for Desiring to Leave the Host Country . . . . . . . . . . . . . 171 29. Migrant Origins by States . . . . . . . . . . 184 30. Importance of Information Sources . . . . . . 190 31. Internal Transportation System Comparisons . 200 32. Retiree Residence Location (Per cent) . . . . 217 xii 10. 11. 12. 13. 14. 15. 16. 17. 18. LIST OF FIGURES Page Central America . . . . . . . . . . . . . . . 16 Travel Route in Central America . . . . . . . 24 Retirement Migration System . . . . . . . . . 30 Retirement Migration Schema . . . . . . . . . 31 Armed Forces Strength Trends for Selected Years . . . . . . . . . . . . 39 Time-Space Path Scheme . . . . . . . . . . . 44 Time-Space Paths of a Typical Military Family . . . . . . . . . . . . 46 Annual Military Retirement Trends (1964-1975) . . . . . . . . . . . . . . 53 Military Retirees in the United States (1975) . . . . . . . . . . . . . . . . 66 Military Retirees in Foreign Areas (1976) . . 74 Military Retirees in Middle America (1976). . 84 Military Retirees in Mexico (1973) . . . . . 87 Military Retiree Locations in Central America (1976). . . . . . . . . 90 Retirement Location Choice . . . . . . . . . 102 Physical Map of Central America . . . . . . . 107 Migrant Circulation System . . . . . . . . . 179 Information Exchange System . . . . . . . . . 188 External Transportation System . . . . . . . 195 xiii Figure Page 19. Main Internal Transportation Systems . . . . 198 20. Internal Contact Network . . . . . . . . . . 208 21. Network Interrelationships . . . . . . . . . 215 22. Military Retirees in the Meseta Central of Costa Rica . . . . . . . . . 221 xiv CHAPTER I INTRODUCTION The population geographer is vitally concerned with the spatial relationships of population phenomena. Accord- ingly, the process of international migration receives spec- ial attention for study because of its obvious spatial char- acteristics (Demko, £3 31., 1970, p. 268). However, since migration is one of the more complex of the factors con- sidered in population dynamics, its study and the formula- tion and testing of new concepts and theories is difficult. The absence of adequate and reliable data for study is es- pecially noticeable (Hollingsworth, 1970, pp. 5-7 and 16- 19). Consequently, many of the older laws and theories of migration cited in the literature (including Ravenstein's laws of migration, written in 1885) are still considered valid theories. Although Ravenstein's laws of migration have endured, they have been challenged at times and some modifications made of them. Most notable is Lee's (1966) work, which reiterated Ravenstein's theories and introduced a migration schema associated with migrants' areas of ori- gin and destination, intervening obstacles to migration, and personal factors affecting individual migrants. These general theories, which are principally of a descrip- tive nature, are useful in forming frameworks for the study of international migration. Historically, international migrations have been ex- amined by measuring the numbers of people moving between specific nations. Moreover, these movements have been analyzed to determine whether they were of a voluntary or involuntary nature and to discover the associated problems and reasons for the migrations (Maselli, 1971). In many of these analyses the movements were categorized and exam- ined according to a defined migration typology. A typology most used today is one developed by Petersen (1958) which distinguishes between forced and free movements of an inno- vative or conservative nature by groups of peOple or indi- viduals. Consequently, major movements of international migrants are identified in accordance with these typolo- gies. The effects of these migrations upon the generating and recipient nations are subsequently investigated (Kirk, 1958) . In Central America, during the colonial period, the major streams of immigrants were received from Spain and Africa, followed by less significant immigrations of Euro- peans and other colonizing groups. The interactions of these migrants with the indigenous Indians and natives formed the socio-cultural complex of the various Central American nations. In recent times international migration in the area has declined and is of little significance in comparison with that of the colonial era. Furthermore, international migration is now restrictive, with most mi- grants admitted in the interests of national development (Neiva, 1965). As a consequence of these current limitations, it is apparent that more attention should be devoted to an under- standing of free and innovative international migrations of a less spectacular, but yet of possible future signifi- cance in Central America. Included in this category are American military retirees who have migrated south of the border to the more favorable climates, recreation areas, and for other amenities of the Latin American nations, with principal concentrations for retirement in Mexico (Sleed, 1974, p. 103). The Mexican government, however, has become concerned about a potential build-up of upper middle class American communities in certain regions, par- ticularly the Guadalajara area, and as a consequence may restrict future American immigrations (Times of the Amer- icas, 1973, p. 12). The Central American nations offer American military retirees similar benefits to those found in Mexico and most of these nations have recently issued government decrees which guarantee foreign pensioners certain in-country bene- fits.1 Moreover, this information is being disseminated 1All the Central American nations with the exceptions of Honduras and Belize have enacted special laws which are beneficial to qualified retirees. in the United States through mass media channels (Turner, 1973; Frisbee, 1973, p. 22). As a consequence, the favor- able conditions for retirement in Central America have al- ready been acknowledged by a segment of the United States populace, including an increasing number of United States military retirees (Harding, 1973a, pp. 30-33). Some of these retirees have recently established residence in Cen- tral American nations, especially Costa Rica, and it ap- pears that many more are now entering this new retirement stream (Harding, 1973b). Statement of the Problem The emergence of this new form of international migra- tion into Central America focuses attention upon a sub- ject which lends itself to investigation by the population geographer. Not only does it offer an opportunity for a spatial analysis of the retirement migration process, but provides a regional laboratory for testing theoretical as- pects of this phenomenon. Purpose of the Study The specific purpose of this study is to examine the developing United States military retiree migration into Central America and attempt to answer questions as to where it is concentrating, why the migrants choose specific na- tions and locations to settle, what are the characteristics of the migrants, how is the migration taking place, and what are the consequences of the process. Contribution to Theory on Foreign Retirement Migration There is no previous study in the international mi- gration literature which deals with retirement migration, nor has anything of a professional geographic nature been published on military retiree migration. This research represents a pioneer effort in this respect and should con- tribute toward a better understanding of international mi- gration from spatial and socio-economic viewpoints. More importantly, it might prove useful as a contribution to a theory on foreign retirement migration. Usefulness of the Study The study should prove of use to United States citi- zens contemplating retirement in Central America by provid- ing an overview of migration channels, retirement settle- ment areas, the causes for and consequences of foreign re- tirement, and information on the activities of their fellow Americans. Officials and planners of the host nations should find the study of use in identifying the effects of Ameri- can retirees on their social and economic structure and for planning purposes in attracting additional members of this group. Objectives The general objectives of the study are to gain an understanding of the spatial processes involved and the spatial structures formed as a result of American military retirement migration to Central America. The determinants of the migrations and the resultant spatial implications will be given special attention as explanatory factors in the migration processes. Moreover, the socio-economic ef- fects of the migration on the migrant group and the nations concerned will be analyzed. Finally, a theoretical retire- ment migration schema will be deve10ped as a framework for examining the Central American model and its applicability to other world environments will be considered. Migration System and Military Retiree Distributions The process of international migration includes a complex set of dynamic and interacting forces which can be best understood and treated in a conceptual framework. An initial objective of the study, therefore, is the formula- tion of a migration system model to be used in analyzing and discussing component parts of the migration. Move- ments envisaged in this migration model produce and modify retiree spatial distributions. Consequently, the depiction of these spatial concentrations is a basic objective of the study. Military Retirees United States military retirees are a select retire- ment age group which represents a segment of American so- ciety that has great potential as a generating unit of in- ternational migrants. These retirees now number over one million2 and as a consequence could have a significant im- pact on specific retirement areas. They generally repre- sent a trained, educated, and experienced international migrant group because of their many years of foreign travel and living. The fact that military personnel and their families have been conditioned to frequent moves and ex- posed to foreign areas contributes toward their propensity to migrate after retirement. Information regarding the characteristics of military retirees, including that of a demographic, social, and economic nature, is important in analyzing the selectivity process in migration and the de- gree of permanency of the migration act. An examination of these characteristics will be a major objective of the in- vestigation. Migration Decision Most international migrations, whether by groups or individuals, have been precipitated by an overriding 2At the end of fiscal year 1975 the total number of military retirees was 1,054,923. Projections at this time called for 1.3 million in the next decade, with a peak of 1.4 million expected by 1995 (Lien, 1976a, p. 16). condition, often economic, social, or political, which led to the decision to move (Petersen, 1955; Swatzky, 1971; Hollingsworth, 1970). According to Cox (1972, p. 75), the most significant factors to consider in an analysis of in- ter-nation migration decisions are those pertaining to eco- nomic opportunity, social amenities, and movement costs. He also believes that language differences and political boundary barriers also play an important and sometimes a restrictive part in the movement. Various migration studies have been undertaken using these factors to explain migration decisions and move- ments. Moreover, most entail the use of socio-economic factors and the application of statistical tests to explain migration behavior (Kariel, 1963; Tarver, 1961). More re- cently, however, new explanations as to why people migrate have been based on behavioral conditions and relationships between individuals and environmental forces (Wolpert, 1965, 1966; Golant, 1971). Amenities as a significant factor in migration studies was given particular attention in explaining migration by Ullman (1954). Although a consideration of all these factors is im- portant in the migration decision process of military re— tirees, that factor pertaining to amenities appears domi- nant in relation to international retirement migration. Military retirees are economically independent or semi- independent (retirement pensions) and can choose a retire- ment area based on its attractiveness (amenities) rather than upon economic conditions or employment Opportunities. Their main concern in the selection of a foreign area for retirement, therefore, would probably be the amenities as— sociated with pleasant living conditions such as climate favorable to health and recreation, beautiful landscapes, an acceptable social status, and special rights and privi- leges granted by the host nations. They are more apt to choose those foreign nations which offer the most advantages for favorable retirement. Consequently, a most important objective of the study will be an analysis of data concern- ing the amenities offered by the Central American countries which influence the decision on retirement location. Networks An understanding of the structure and process of mi- gration and information networks is of particular impor- tance in spatial analysis. Migration and information flows have been described and explained by various methods. Rav- enstein (1885) established a relationship between distance and the propensity to move in his early published laws of migration. Gravity models which generally employ a dis- tance-decay component have been widely used as a quanti- tative measure, since they exhibit locational regularity and can be explained in terms of least effort (Zipf, 1946). The intervening opportunities model introduced by Stouffer (1940) has received some use and other studies employing 10 modifications of these models as well as some new ideas have also been utilized to explain and predict migration patterns (Ter Heide, 1963; Haenszel, 1967; Bunge, 1971). In many studies, particularly those dealing with in- ternal migrations, these methods have proven useful in ex— plaining migration patterns. In the case of inter-nation movements, however, these methods become more difficult to apply and sometimes are completely inapplicable. It is believed, therefore, that international migrations of the type envisaged in this study can best be understood by considering them as deviant migrations (those not fitting the normal distance-decay pattern) and looking for explan- ations in terms of components such as the acquisition of information through personal or mass media channels (Cox, 1972, pp. 79-81). Information exchange is the critical factor, then, since it is necessary for establishing net- work linkages. Considering the migration network in terms of information linkages based on the locations of favor— able retirement areas, it is conceivable that a number of human interaction networks would develop. A major objec- tive, then, is to identify and examine the migration chan- nels, information flows, and networks; and to evaluate their importance in the migration process. 11 Settlement Patterns Once migrants enter a foreign country, they tend to distribute themselves in various patterns. Their spatial arrangements in the host nations, whether they are clus- tered. scattered. mixed military retiree settlements with other United States migrants. urban or rural oriented, or mixed settlements of United States migrants with nationals, are important aspects of the migration phenomenon. These arrangements can influence the rate and amount of inter- action between the migrants and nationals. Internal link- ages and communications flows are sustained through various social, economic, and political affiliations and activities in the respective nations. This aspect of the study will entail answers to such questions as the following: Where are United States military retirees located within each host nation? Are their settlements identifiable? Do the retirees mix with the nationals, or do they remain sepa- rate? Are the retirees dispersed among the existing na- tional settlements or are they established in nucleated United States settlements? The primary objective in this respect is to spatially locate and explain migrant settle- ment patterns. Adjustment and Effects Migrant adjustment and adaptation to a new environ- ment, and the socio-economic effects of the migration on the 12 host communities as well as on the migrants themselves, are important considerations in determining the success and continuation of the migration (Jones, 1956). This leads to the question of whether or not the migrant will succeed or fail in his endeavor to adjust to his new en- vironment and subsequently affect the permanency of the migration act. In looking at these factors, principal ob- jectives are to determine the influences of relatives and friends and national hostilities on the migrant's persever- ance in the host nations. Hypotheses In order to accomplish these objectives the following hypotheses have been formulated for testing in the study: 1. Conditioned to frequent moves and exposure to foreign areas, military people have a high propensity for foreign migration after retirement. 2. Military retiree migrants share the follow— ing characteristics: a. Median age in the mid-fifties b. Married c. Highly educated d. Substantial retirement income e. Similarity of previous residence f. Propensity to return to the United States 3. In the migration decision: a. Site selection is directly related to the amenities offered by the host nations. b. Climatic conditions are of prime importance. c. Cost of living is of secondary importance. d. Previous exposure to, or residency in, Latin America are important considerations. 13 e. Place differentiation is dependent upon the exchange of information and the com- munications linkages between the United States and the host nations. 4. In the host nations: a. The retirees locate mainly in the central cities or adjacent areas connected by the major routes of communication. b. The numbers of retiree migrants vary in— versely with the distance from the capital or major cities. 0. Retiree clustering is directly related to housing and recreational developments. Methodology It is believed that significant research in the social sciences can best be accomplished by forming theoretical concepts which then can be empirically tested and applied to the real world. Geography, in particular, lends itself to these processes because of its inherent methodological approach to solving social problems, i.e., Spatial organi- zation applications, regional analysis, and field survey techniques. In an examination of the American retirement migration structure in Central America, all appropriate methods for accumulating information and data were utilized; however, field surveys were the main source of primary data. Personal Interest in the Subject With domestic social problems, especially those exist- ing in our major cities, in need of geographic solutions, 14 it is natural to question the selection of a subject deal- ing with foreign areas and a group composing a relatively small segment of our society. The issues of impact and significance will of course be developed and explained in the subsequent analysis. For the present, it appears per- tinent only to point out my interest in the subject and qualifications for its investigation. As a retired Army officer with over twenty years of active military duty, I have served in several foreign re- gions, notably Asia, Europe, and Africa. During these over- seas assignments I became aware of living conditions and attitudes of foreign nationals regarding American military visitors in their respective countries, as well as the im— pact of the international moves on the American military personnel themselves. Thus, having been exposed to these foreign experiences and sharing many of the same character— istics, beliefs, and institutions of military peOple, I feel qualified to undertake a study of an international nature dealing with this group. Since retirement from the military, I have been teach— ing college level geography in Michigan and concomitantly, through graduate study, have acquired an interest in migra— tion theory and in Latin America as a regional field of study. Furthermore, being a military retiree and having chosen, from a climatic standpoint, an unfavorable area for retirement living, the questions of where military retirees 15 settle, and for what purposes, often come to mind. Given particular consideration in this questioning process are those foreign areas which present a combination of climates and social amenities for retirees. Consequently, the desire to combine these interests and qualifications into a rele- vant research area led to the decision to search for a prob- lem in international migration dealing with American mili- tary retirees. Selection of the Study Area Central America, excluding Panama, was chosen as the research area for several reasons (Figure 1). It lies within the Latin America sphere of interest, has perenially favorable climates, offers special privileges to foreign retirees, and is accessible to the United States by direct highway connections. Mexico, a pOpular retirement area for many years, also satisfies these requisites, and certain other areas of Latin America provide advantageous conditions that attract foreign retirees. Central America, however, although a small geographic entity, is composed of several politically distinguished nations, a situation which facil- itates the researcher's task of comparing and contrasting their retirement attributes. Most significant in the se- lection process was the receipt of information through var- ious media, mass and interpersonal, regarding an influx of American retirees into Costa Rica beginning in early 1973. 16 mofipoe< Hmswcoo .H onswfim coo _ _ \s 0%” D d Ohm com 1 _ _ _ _ . «enema .. oou oo. o .5 A mu4_z . N some _ 1 numoo z oo. IIGO- :c(:: c n o o o I . o H u a o a a I. a o w «unannouz .cmonnfismo \/ L. fioomSHmmH \\ Ill. th/II \L Vsa \\ was: . 0 . \\ Uflom d . \ A '09- \ ‘ s on O 00% «H named—.6 \. \ _, .. _ u c I / ab MD 0038: g yl...ll.-ll._ 17 This information was the catalyst that prompted my prelim- inary investigation of this new activity in Costa Rica and which further influenced my decision to look at the phenom— enon in the larger perspective of Central America. Panama was not included because the strong influence of the Canal Zone on American activities in the Republic of Panama could distort results of the study. Data Requirements and Sources To deve10p, describe, and explain the overall migra— tion process in the study area, a wide range of data is re- quired, i.e., pertinent statistics on the active military force, the military retiree base, and the Central American military migrant group. Moreover, information on the nature of the linkages between these groups, the migration and communications networks, and other external and internal functional relationships is vital in the collection effort. Although certain information, including population figures and socio-economic information for the active military or- ganization and the military retiree base in the United States, are important requirements, the military retiree migrants in Central America are of principal interest in the collection of primary information. Consequently, data categorized as follows are required on United States mili- tary retirees in the host nations: 18 Personal data (demographic, social, and economic) Financial data Military service data Military service migrations Retirement migrations Migration decision data Place location data Permanency and area satisfaction data mflmtfltht-l Other required data include information on the actual locations of retirees in the host nations and the amenities offered to them, especially those pertaining to incentives for settlement such as importation of duty-free items, fav- orable taxes, and other special allowances for retirees. The chief sources of data are the military retirees located in the Central American nations. Other important sources include United States government personnel, both in the United States and the host nations; officials of military affiliated organizations; Central American govern- mental representatives; and other American personnel resid- ing in Central America who have an intimate knowledge of the area. Data Gathering Techniques Data for the study were obtained by searching the literature, contacting United States government officials ;personally and through correspondence, and personally in- terviewi ng the migrants . 19 Literature Search A survey of the literature regarding all aspects of Central American retirement areas and communities, real estate ventures, and other schemes which may develop into retirement areas was completed prior to the field investi- gation. Published material pertinent to military retirees and their migration histories was reviewed and abstracted as appropriate. Moreover, a thorough search was made on a country by country basis to determine the existence of in— formation on special government provisions and decrees for foreign retirees, as well as other data of a political, economic, and sociological nature which might affect re- tirees. During the literature search stage, references pertaining to methodologies that could be useful in the study were also noted. As anticipated, little published data on this subject were uncovered. Preliminary Data Collection In order to obtain information on the numbers and locations of American military retirees residing in Central America, the following Department of Defense agencies were visited: 1. Headquarters, Department of the Army U.S. Army Military Personnel Center Retired Activities Branch Alexandria, Virginia 2. Headquarters, Department of the Navy Bureau of Naval Personnel Retired Personnel Support Section Washington, D.C. 20 3. Headquarters, United States Marine Corps Personnel Affairs Branch Retired Activities Section Washington, D.C. 4. Headquarters, Department of the Air Force Air Force Military Personnel Center Randolph Air Force Base, Texas 5. Headquarters, Department of the Army U.S. Army Finance and Accounting Center Pay Services Division Indianapolis, Indiana In each case discussions with appropriate officials at these agencies revealed that names and addresses of individual retirees could not be released because of the implementation by the Department of Defense of privacy laws. Information was obtained from the agencies, however, which gave total numbers of retired personnel living in Middle and South America. Since Army Post Office (APO) addresses were given in some of the listings, complete country breakdowns were not possible. Consequently, all of the agencies personally contacted agreed to accept and staff a letter from me which would request additional in- formation on this subject, in accordance with provisions of privacy laws. The Retired Officers Association in Washington, D.C. was also visited to inquire about the availability of list- ings of retired officers located in foreign nations. Some helpful material was obtained from this source, and the editor of the Association's journal accepted for publica- tion a personal plea to members for information regarding 21 my research interest. Responses, however, were disappoint- ing, with most coming in the form of requests for copies of the study, when published. Another article published by the Army, Navy, and Air Force Times of Washington, D.C. produced similar results. Following these visits, correspondence was initiated with the previously enumerated Defense Department organi- zations, plus the finance and accounting centers of the Armed Forces. These letters requested the release of names and addresses of retired personnel in Central America, and contained a full justification for the information, stating the purpose and use of the study. A listing of all foreign nations where retirees reside and the total numbers living therein was also requested. Letters were drafted to the librarians of the following institutions seeking informa- tion or references pertaining to United States military re- tirees in Latin America, and any other background material and general coverage on the area: 1. Industrial War College Fort McNair, Virginia 2. National War College Fort McNair, Virginia 3. Inter-American Defense College Fort McNair, Virginia 4. U.S. Army Library The Pentagon Washington, D.C. All replies from the Defense Department offices cited the Privacy Act of 1974, which prohibits the disclosure of 22 addresses to any person without the consent of the indi- vidual involved, as a justification for denying the request. In all instances, however, they did provide a listing of foreign nations where retirees reside, along with the total number of retirees in the respective nations. Nothing of a positive nature was received from the Defense librarians, two of the four failing to reply. Since the retiree locational data could not be ob- tained from the referenced sources, it was decided to use the total number figures provided by the Defense Department as a general guide for each nation and to proceed with the field survey, utilizing in-country sources to later iden- tify and locate the retirees (Table l). The principal con- tacts in these cases were United States government and foreign government officials in each country. Questionnaire Preparation Prior to commencing the field survey in Central Amer- ica, a questionnaire was designed that would satisfy the data requirements listed previously (Appendix A contains a copy of the questionnaire). Originally, the question- naire was approximately one-third longer. Its pre-testing in Mexico the summer before its use in Central America, how- ever, proved advantageous by identifying questions, particu- larly those dealing with active military assignments, which could be eliminated, thus reducing its length. This pre- testing also provided constructive criticism on clarity and 23 Table 1. Military Retirees in Central America W Air Country Army Navy Force Marines Total Belize 5 - - - 5 Costa Rica 24 13 1 3 41 El Salvador 1 l 2 - 4 Guatemala 1 5 3 2 11 Honduras 3 3 - - 6 Nicaragua 2 - — l 3 Total 36 22 6 6 70 Source: Data received in 1975 and 1976 through personal correspondence with the separate service depart- ments of the Department of Defense. the arrangement of questions. The questionnaire was the principal instrument used in the data collection process and the basis of information for most of the analyses. Field Survey The field work and military retiree interviewing in Central America were conducted during the 1975-1976 academic year. The travel route from the United States on a country to country basis consisted of travel by private car through Mexico, Guatemala, El Salvador, Honduras, Nicaragua, Costa Rica, and Panama, returning to the United States in the re- verse order, and including Belize after departing Guatemala (Figure 2). 24 mofisoa< Hmspsoo cw opsom Ho>ms9 .N osawum _ 1 com com - 4 % q J cow 0 _ 0 2.2.0 I. r. $6 2.2:? /. mum... 5.60 2.2.0 ace-oz 00.... 100— Z mm_._._o turbo O 2.2.20 v 2623! quo 43:40 O 5. 36.228 3.03.6 (v...) I (JV a: . a .. 38.. . : .224 .aooON 2.3» on... 100- \ so 7. \ .Illlun r/ .22... a .38 u k .110 ‘§N-l I..- lull II— 000 000 s 00¢ . — . L 25 Since the total military retiree migrant population in Central America was the target for the field investiga- tion, the principal problem in each nation was to identify and locate every person meeting this criterion. Once this was accomplished, the process of making contact and com— pleting the questionnaire interview could then be undertaken. The approach taken to obtain this information was generally consistent within the respective nations. First, contacts were made with appropriate American organizations, including the American Embassy, Office of the Defense At— tache, United States Military Group, Inter-American Geo- detic Survey, American Legion, American Chamber of Com- merce, and the Retired Officers Association. Of this group, a Retired Officers Association chapter was found only in Costa Rica, and American Legion chapters in Guatemala and Costa Rica; whereas all other organizations appeared in each nation except Belize, where only the American Embassy is present. The project and its potential usefulness was explained to representatives of these organizations and as a consequence of their receptiveness, most furnished some names of military retirees. The standard procedure was then to contact each known retiree for an interview and at the same time query him regarding the identification of other retiree acquaintances. Certain of the Tourist Bureau officials of the host nations were also most helpful in identifying retirees and 26 providing other pertinent material on retirement conditions. fflhis was especially true in Costa Rica and Guatemala. Through the utilization of these sources, it is be- lieved that all, or at least the vast majority, of military :retirees were identified and subsequently contacted if phys- ically present in-country. This investigative method iden- tified larger numbers of retirees in Central America than had been reported in the Defense Department figures (Table 2). Concurrent with the interviewing procedures, general field observations were made and a variety of other sources contacted to provide a broad range of information useful in the overall analysis of the study. Data Analysis The data contained in the questionnaires, combined with other collected field data and supplemented by perti- nent secondary and tertiary sources, form the base for the analyses in the succeeding chapters. 27 .mampou ca uoesaocfi uOZu .mmoapom 0: gas? was mad: an poumawcoofi omona no .wmoupom caocx pmmH Sony manuhmmou psocmEnmm mo omsmoon popompcoo on won canoe on? mochawoh cofimfiamsan .oow>umm memos Hmuo>om saws mumfi>somon hhduaaaa ..o.fl mmsoflmcon hhdpfiaas you pofiHHHmnd nos who; on; msomumnm .mcomaos mmocfimsn so .msflnnonm .Hmoaooa .coapmom> you 080: Scam head no hsucsoo mo ado maauaaoasou ohms menu omsmoon pmaonwsoo on yo: oasoo 0:3 moohfluou poHMfiHm=G* baa m cm oflhv mm vHH Hapoe m H oAHv N m mamasmofiz m N m measoeom Hm n ma mamaopmso m e v goua>Ham Hm ems m we came «a as «one «pmoo H a oufiflom eaves pomsmom neopaooq moanaMaHm *pouonpcoo mama>hopcn mapcsoo poz uoz poz umumHnaoo m0030u>umpnm 059 m0 sofiwfimonaoo .N oHadB CHAPTER II MILITARY RETIREE CONCENTRATIONS Before focusing on the distribution of military re- tirees in Central America, it is necessary to look at their worldwide distribution and the processes which formed con- centrations of retired migrants. Consequently, this chap- ter is concerned with a description of the general retire- ment migration system and its component populations, an introduction to the migration schema used as a framework in the study, and the domestic and foreign distribution of military retirees. Retirement Migration System The foreign retirement locations of military retiree migrants are not ones of random selection, but the result of choice based on previous exposure to foreign areas or information about those areas accumulated during military service. In this respect, it is believed that the active military organizations (Army, Navy, Air Force, and Marine Corps) function as training institutions for international retirement migration. This concept of a military 28 29 retirement migration system is depicted in Figure 3. The model employed here is one of components which direct and control the flow of military persons into foreign retire- ment areas. In this system the active military organization is a producer component which places military retirees into a donor component representing military retirees in the United States, and the source of potential international migrants. Certain of these retirees migrate to foreign nations (recipient component), thus becoming international migrants. Interaction among the components of the system is maintained through the transfer of information. The flow of persons and the exchange of information in the sys- tem form circulatory movements which create a chain migra- tion effect, sustaining the system's operation. With this model as a base,and considering it as the generating mechanism for potential international migrants, the donor and recipient elements of the model (military re- tirees) were used as the main components to formulate a £3- tirement migration schema (Figure 4). This schema was de- signed as a framework for analyzing and explaining the com- ponent parts of the migration process, both static and dy- namic, and to insure that all important aspects of the phenomenon were given full consideration in the study. Ma- bogunje (1970) introduced this method for studying migration when.he utilized a systems approach to understand and de- velop theory on rural to urban migration, especially that 30 PRODUCER Active Military Organization Training Institution for International Migration DONOR Potential International Migrants Military Retirees in the United States J RECIPIENT International Migrants .(_.._.J Military Retirees in Foreign Nations Figure 3. Retirement Migration System uotiemJOJUI (AOL! J Environment (Social Conditions) 31 Environment (Physical Conditions) I Stimuli . ———————— 1 ------- 1 I J Potential l I Migrant . l I l | DONOR COMPONENT l l- 8 ~ — »| I 0:1 1 Origin I I Contgotlls orig 1 n Adjustments I Migration Decision Positive Negative Feedback Feedback Migration Channels Destination Destination '1 Controls Destination Adjustments RECIPIENT COMPONENT l A J Migrant1~— A ’ l J : (suctntpuoo IVOIlttod)l iuemuoxtnug Environment (Economic Conditions) Figure 4. Retirement Migration Schema 32 pertaining to Africa. Figure 4, therefore, uses the ideas expressed by Mabogunje, although modified to fit the spe- cific interests of this study. The schema illustrates the functions of migration ele- ments and the processes involved in migrant movements, as well as the relationships of the component parts to each other and the outside environmental effects. Within this scheme, the donor and recipient components, along with their respective sub-components, represent specific populations located within a spatial context. Acting on the inhabitants of these populations are various internal and external fac- tors which may create or retard movement. In theory, the potential migrant, located at the place of origin, is sub- jected to a number of control and adjustment mechanisms, such as family and community interests and other socio—eco- nomic relationships of either a positive or negative nature. Concomitant with these internal functions, physical, politi- cal, economic, and social environmental influences (stimuli) also reach the potential migrant, thus affecting his desire to stay or move. If his decision is to move, he must eval- uate a number of geographic places in terms of favorable and unfavorable conditions related to the purpose of the migration and then make a decision as to the best specific Place. Once the migration decision is made, he becomes a migrant and follows particular migration channels to the Selected destination. At the destination, within the 33 recipient component, the migrant again faces controls which encourage or discourage his stay; i.e., national laws and statutes, local customs, etc., and is subjected to adjust- mggt_mechanisms, such as social organizations (church, clubs, etc.), that affect the success or failure of the mi— gration. Here the migrant is again influenced by environ- mental conditions (physical, political, economic, and so: cial) and, in combination with those factors considered under the recipient component, he evaluates his new habitat and transmits information as negative or positive feedback to the potential migrant of the donor component. The flows of people and information in this model are not uni-direc- tional, but circular. As a consequence, it is obvious that the migrant, if disaffected with his new environment, could again become a potential migrant in search of a new retire- ment site. On the other hand, if fully satisfied, he be- comes a carrier and transmitter of positive information which might influence significant directional migration flows. In any case, the movements of people between the donor and recipient components create structural growth or decline in the respective populations, along with the at- tendant spatial changes of human distribution and infra- structural patterns on the landscape. This model, then, fully illuStrates the basis for spa- tial interaction by military retirees. The complementary factors of supply and demand are represented by military 34 retirees as potential migrants in the donor component, and by the retirement opportunity areas shown as the destina- tion in the recipient component. Intervening opportunities, whether present or not, are implied in the model as other foreign retirement places that the retiree is exposed to in executing the move from the place of origin to destina- tion. Finally, the transferability, or friction of dis- tance, factor is considered as the actual migration path or channel of the military migrant between the origin and destination, and is measured in terms of geographic dis- tance, time, and/or money. The environmental conditions of the model, inferred as domestic and foreign, provide the stimuli responsible for the retiree's migration decision and the activation of the spatial interaction, thus creat- ing spatial forms and structures. The specific military population involved and the distribution of its relevant component (military retirees) are now appropriate for dis- cussion. Military Population For background information on military retirees, in- cluding their numbers, characteristics, and learning ex- periences, it is essential to survey the United States mil- itary population, past and present, from which this retiree population was generated. 35 Our present military structure was created by the National Security Act of 1947, which established the De— partment of Defense and its three separate armed services, the Department of Army, Navy, and Air Force, with the Mar- ine Corps under the jurisdiction of the Navy Department. Prior to this reorganization, the American military forces consisted of an Army, a Navy and Marine Corps, all nation- alized in 1776. These services fluctuated in size through- out their history, according to domestic and foreign re- quirements and commitments. Historical Bases of Foreign_Exposure In the first hundred years of America's existence there were few periods of peace; however, the military and naval commitments of the United States then were confined mostly to warfare and military actions on or near our home front; i.e., the War of 1812, the Mexican War, the Civil War, and the Indian Wars of the West. Other than the Mex- ican War, fought on soil south of our border, the first major overseas military operations conducted by United States forces occurred in the 1898 Spanish-American War fought in Cuba, Puerto Rico, and the Philippines with a military force of 210,000 persons (Hotchkiss, 1975, p. 20). Since 1900 American military forces have been involved 1n.four major wars and numerous expeditions, interventions, ttnd small wars encompassing a major portion of the globe. IDuring the early 19005 singular or joint military task E! ‘2 36 forces were engaged in fighting or intervention Operations in Puerto Rico, Cuba, the Philippines, North China, Panama, the Dominican Republic, Honduras, Haiti, and Mexico. Our first major war, World War I, was fought in Europe with an American expeditionary force of more than two million per- sons. American military action extended from April, 1917 to November, 1918, though American trOOps remained as an occupation force in Germany until 1923. Between World War I and the advent of World War II in 1941, America went through a period of isolationism in which military activities were confined to a few interventions (Nicaragua, Haiti, and China) to protect the lives and interests of Americans abroad. Over 16 million Americans were brought into the armed forces during World War II, in response to a major war effort that exposed them to countries in every conti- nent. Following the victory in 1945, many Americans re- mained in Europe and Asia as occupation forces, despite military strength reductions to about 2 million in 1946. When the Korean War began in 1950, American military forces numbered just under a million and a half persons, the low- est level since pre-World War II. However, World War II- experienced reserve units reinforced the regular forces tind by the war's end over five million persons had served 1:1 that conflict. During the period between the Korean War and the beginning of the Vietnam buildup in 1965, mil- 1tary manpower strengths averaged about 2.5 million. 37 These were considered minimum manpower levels to satisfy home defense and strategic reserve plans, to provide ade- quate forces to meet overseas commitments in accordance with combined and bi-lateral defense agreements, and to man worldwide military assistance and advisory groups. During the height of the Vietnam War in 1968, armed forces per- sonnel peaked at about 3.5 million in support of the war effort and other defense needs. At that time, about 1,200,000 military persons were serving overseas, including some 600,000 in Southeast Asia, approximately 300,000 in Europe and South Korea, and the remainder distributed throughout other areas of the world considered to be of importance to the welfare and defense of the United States. In terms of foreign places, military personnel were located at 343 major and 1,927 minor bases. Japan alone maintained 48 major military installations and the Philippines 12. Bases in West Germany numbered 146, Korea contained 55, Okinawa 18, Italy 9, Spain 8, and Greece, Turkey, and Tai- wan each had 3 (Donovan, 1970, pp. 2-11 and 52-54). Recent Foreigg_Militagy Commitments The wind-down and conclusion of the Vietnam War re- duced armed forces strengths to near post-Korean War levels. As of December, 1975, the active military force consisted of 2,085,000 persons, with 767,000 soldiers, 525,000 sail- ors, 193,000marines, and 600,000 airmen (DOD, 1976, p. 21). 38 Figure 5 shows Armed Forces strength trends by the separate services for selected years. By the end of 1975, the overseas military nampower strength totaled 481,000, distributed as follows (DOD, 1976, p. 34): 306,000--Western Europe and related areas (217,000 in Germany and 27,000 in the fleet) 129,000--Western Pacific and related areas (48,000 in Japan, 41,000 in South Korea, and 22,000 on ships of the fleet) 46,000-d0ther geographic commands and foreign installations (See Appendix B for a detailed breakdown) American armed forces personnel overseas, then, represented about one-fourth of the total active military establish- ment and were stationed,on a rotational basis, in those foreign areas traditionally considered of strategic im- portance to the United States. Since there have been no major changes in strength or deployment of troops since 1976, it can be assumed that mobility and the opportunity to visit foreign places continues to be a way of life for the American serviceman. Military Veterans From the brief history of American military involve- ment in international affairs, it is obvious that many mil- lions of American men and women gained knowledge of foreign areas that in modern times no other nation could possibly match. Most of these service persons were not professional 39 NUMBER OF PERSONS IN MILLIONS ISIS I933 |945 I948 I952 I960 IDS! IS75 ARMY _—NAVY O MARINE CORPS """ AIR FORCE Source: Department of Defence l976 Figure 5. Armed Forces Strength Trends for Selected Years 40 career people, but simply those citizens who responded to the call in time of need. Consequently, the majority were released after relatively short periods of service, became additions to the list of military veterans, and returned home to families and jobs, though some continued their mil- itary associations by joining the military reserve forces or by membership in veterans organizations. At the end of 1975 America had 29.4 million living veterans, nearly 14 per cent of the total United States population, with over one-third of them living in the states of California (3.2 million), New YOrk (2.5 million), Penn- sylvania (1.8 million), Texas (1.6 million), and Illinois (1.6 million) (TROA, 1975, p. 41). Some 225,000 veterans were living outside the United States, 159,000 of whom re- sided in Puerto Rico. Most of these were veterans of World War II (13.5 million). Korea provided 5.7 million, those from World War I numbered just under a million, and the majority of the remainder were veterans of the Vietnam War era (7.9 million). Out of this sizeable segment of the American popula- tion has evolved a varied number of veterans organizations, foremost of which is the American Legion, with a 1976 mem- bership of 2,504,794. American Legion members living abroad in over 30 countries and territories at that time accounted for 17,000 of the total number of Legionnaires. American Legion posts in Latin America are found in Puerto Rico, 41 the Virgin Islands, Bermuda, the West Indies, Mexico, Guatemala, Costa Rica, Panama Canal Zone, and Peru (Amer- ican Legion, 1976, p. 30). Other significant veterans groups include the Veterans of Foreign Wars (1,800,000 mem— bers), which consistently lobbies for worldwide deployment of United States armed forces, and the Disabled American Veterans (245,000), many of whose members are eligible for military disability pensions (Donovan, 1970, p. 29). For purposes of this study those disabled veterans who receive full disability allotments qualify as pensioned military retirees, since the monetary amounts received are usually sufficient to sustain retirement. As of July, 1976 there were approximately 160,000 retirees in this category. Military Reserves Among those veterans who have served in the Armed Forces are persons that have opted to continue their affil- iation with the military establishment in a semi-active capacity after release from active service. These are the veterans assigned to the nation's military reserve forces, many of whom remain with those units for a sufficient per- iod to qualify them for partial military pensions upon reaching age 60. The reservists who have fulfilled these requirements are included in the United States Government's military retiree community, and, as a consequence, are also eligible for inclusion as retirees in this study. By the end of 1970, about 100,000 reservists had retired under the 42 reserve program. Since most reservists began their military service in World War II, and the average age of service- men at that time was 25, it has been estimated that re- serve retirements will peak in 1986, with a total of 136,980 (Army'Times, 1970, p. 13R). The Reserve Forces of the United States in June, 1976 totaled 822,992 service personnel. By individual armed service elements, these reservists were distributed as follows (Lien, 1976b, p. 15): Army ' National Guard. ...... . .......... 362,330 Reserve....... ........... .......194,611 Navy Navy Reserve.......... .......... 97,051 Marine Corps Reserve ............ 29,638 Air Force National Guard............ ..... . 90,992 Reserve......................... 48,370 Since the termination of selective service in 1974 this component of America's armed forces has lost much of its attractiveness for veterans of the Vietnam War era and for new recruits. Consequently, an organization which as re- cently as 1967 had almost 2.8 million members now has fewer than one million, and will have little future effect in significantly increasing the number of military retirees. Career Personnel The most important element of the military population, especially in terms of ultimately producing military re- tirees, is that group of tenured career military personnel 43 who serve for 20 to 30 years, or longer, in the active ranks of the armed forces. These service people, through frequent moves between domestic and overseas stations dur- ing their military careers, have been exposed to a variety of foreign areas, cultures, and customs. Before World War II, American career soldiers served principally at iso— lated posts in the continental United States. That war, and the resulting revolutionary changes in peace-time Amer— ican defense policy, however, brought about new missions for the armed forces that have subsequently required pro- fessional military persons to serve from about one-third to one-half of their careers abroad. In many of these for- eign moves, not only are the servicemen affected, but also their immediate families. International migratory patterns of career military people, particularly those reflecting alternate tours at foreign and domestic stations, can be graphically illus- trated by use of time-space path techniques. This metho- dology, developed by Hagerstrand (1969), was to investigate individual and family migration paths in Sweden. The gen- eralized scheme (Figure 6) illustrates a time-space path in which an individual or a family moves through specified periods of time at various geographic stations. In Figure 6, x, y, and g_are geographic stations from which spatial movements take place as indicated by the directional arrows. At each station, the heavy vertical lines (Time 1, Time 2, 44 TIME TIME3 Source! Hagerstrand, Toreten (I969). "On the Definition of Migration", Yearbook at Population Research in Finland, III63-72. Figure 6. Time—Space Path Scheme 45 and Time 3) represent the period of time spent at that place by the mover. Obviously, the movement history of individuals, families, or groups can be readily portrayed in this manner. Using Hagerstrand's methodology, the international movements during the career of a typical military family are illustrated by Figure 7 to further support the nomadic nature of military persons. Migratory events were measured on the basis of stations representing either the continental United States or a foreign area (Figure 7). No internal migration patterns are shown. The time interval extends from the birth date of the military man to one year after military retirement (1924-1970). Solid vertical lines rep- resent the lifeepath of a man while broken vertical lines are used for a 32532. In the family scheme, children are located to the right of the parents and children births are designated by an'§ (son) or‘D (daughter). The principal m§2_and 2122 time lines are the heavy lines designated M (man) and 1 (wife). Horizontal lines are migratory move- ments either out of or into the United States, depending on the direction of the arrows. An analysis of the time-space paths of this family reveals that the man entered the military service at age 18, remained in the United States for one year, moved overseas for two years (World War II), then returned to the United States. The woman separated from her family at 46 ' Retirement I970 P I > COIIege .. II II II I965- 77' > s“ -I E r >College - I < I I J I960 - J, I > - I I I 1 \ - 1:] II _ ISSSr - > 7] “*IH , 7' Q J ISSOF- : _ I I ”7‘ I945 ' *ws 4‘ ‘ Marriage 4 ' .. > ] II I Begin Military Service II .940 - l I: - l I I ll ' II I935” I M-M°" II -I I w- wm : S- San I. I 0- Daughter II I930 *- l :: n I I | I I$4- ISZSP I 5 I -I :31 INTERNATIONAL MOVEMENTS l Figure 7. Time-Space Paths of a Typical Military Family 47 age 18, met the military man that same year and was subse- quently married. A son was born the following year and the family remained together in the United States for the next five years. In 1951 the man migrated overseas alone (Korean War), returned a year later, and a daughter was born the following year. After two years together in the United States, the entire family moved to a foreign area during the period 1954 to 1957, then returned to the United States. After three years in the United States, the man was again separated from his family during a two year for- eign assignment. Upon return, the family remained together for two years and during the third year the son left for college. In1965 the man, wife, and daughter again de- parted the United States for a foreign tour of duty and did not return until 1969, when the man retired from mili— tary service. A year later the daughter left for college. Thereafter, with only the man and wife remaining in the household, and receiving a lifetime government pension, they are prime prospects for future foreign retirement migration. It is recognized that professional American military persons are not alone in undertaking these foreign moves. State Department foreign service personnel and many other career governmental employees, as well as numerous execu- tives and technicians of United States multi-corporation organizations spend large portions of their lives overseas. 48 Some also, quite surely, settle in their favorite foreign habitat after retirement. In terms of significant numbers involved and long period commitments, however, there is no other society or group that equals those of the career military establishment. The learning experiences and training received during a military career are long lasting ones, particularly those pertaining to comradeship, job or mission accomplishment, teamwork, and the ability to accept and apply the creeds and traditions of service life. This experience of sharing a common life style may lead to the desire of overseas servicemen to seek out common social institutions. Military Retiree Population For the military careerist, after having served 20 to 30 years in a profession that has provided excitement with each change of duty assignment, the opportunity to meet new friends and frequently renew Old friendships, and to travel throughout the world, retirement calls for a major adjustment. The unique aspects of military life, profes- sionally and socially, make complete separation from this environment difficult. Consequently, many retirees attempt to retain their military ties after retirement by locating near military establishments or in communities where they can relate with others who have shared similar experiences. Military posts and stations also offer military-related 49 job opportunities for those searching out new careers. In addition, military installations provide various facili— ties, such as medical, commissary, post exchange, etc., which retirees are authorized to use. These phenomena have caused retirees to cluster around military installations, including foreign bases. This is especially noted where major bases are found close to large cities, further en- hancing post-retirement employment opportunities. For ex- ample, large retiree populations are found in the areas around Washington, D.C.; San Antonio, Texas; and San Diego, California. On the other hand, some retirees return to their hometowns, accept jobs far from military facilities, or seek out remote retirement sites. For those retirees not working or seeking employment, an active and interest- ing way of life for many has been maintained through repet- itive retirement moves, both domestic and foreign. Employment Versus Full Retirement The transition from military to civilian life has an impact on most retirees not only psychologically, but more importantly, in a financial way. With incomes reduced over 50 per cent in some cases, combined with early-age retirement and continuing inflation, after military service the majority of military retirees find it necessary to pur- sue a second career in order to maintain an acceptable standard of living. Table 3 illustrates that the younger a retiree is, the more apt he is to seek employment 50 Table 3. Per cent of Military Non-disability Retirees Who Have Never Been Employed. _===-= Age At Retirement Officer Enlisted 37 or less 2.2 1.9 38 - 39 2.1 1.7 40 - 42 2.8 2.7 43 - 45 5.0 3.2 46 - 48 7.1 4.9 49 - 51 12.1 6.6 52 - 54 16.8 10.4 55 - 59 24.5 16.8 60 or more 30.6 27.7 All ages 8.9 4.0 Source: Office of the Assistant Secretary of Defense upon retirement. Moreover, it shows that throughout the life cycle a higher percentage of enlisted personnel re- quire jobs than do officers. Although most retirees be- come employed even in the older age brackets, beginning at age 55 and above a substantial number of them opt for full time retirement. A recent survey by the Retired Officers Association, an organization of some 237,000 retired commissioned and warrant officers of the uniformed services, indicated that nearly 43 per cent of the members were fully employed, 11 per cent part-time employed, five per cent seeking jobs, and the remaining 41 per cent fully retired (Foster, 1976, p. 18). 51 A large proportion of retirees seek employment with the federal government. In 1976 almost 142,000, or 14 per cent of all military retirees, were working in federal jobs. This represents about five per cent of the total government civilian workers. The largest employer, as could be expected, is the Department of Defense with about 78,000 retirees. Moreover, the Postal Service employs some 34,000, the Veterans Administration nearly 8,000, and sev- eral other agencies hire a few thousand (Lien, 1976c, p. 18). Some of these retirees fill positions with these agencies at overseas locations. Retiree Numbers The total military retiree population in mid-1975 was 1,043,923 persons, 17,649, or 1.7 per cent, of whom were located outside of the continental United States.3 Of the total number, 363,701 were Air Force, 354,031 Army, 272,381 Navy, and 53,810 Marines. Compared with the total United States population, less than one half of one per cent 3The Coast Guard, under Navy jurisdiction during war- time, although not a separate armed force of the Department of Defense, is one of the seven uniformed services of the United States and provides a retirement pension for its per- sonnel similar to that of the Navy. At the end of 1976 it had an estimated 17,500 persons on its retirement rolls (Lien, 1976c, p. 16). Commissioned officers of the National Oceanic and Atmospheric Administration (NOAA) and the Pub- lic Health Service (PHS) are the remaining members of the uniformed services. Pay, benefits, and retirement plans for these members correspond with those of the commissioned personnel of the armed forces. The retirement numbers for the few members of these groups are: NOAA-120 and PHS- 1,400 (Lien, 1974, p. 19). 52 of that population is retired military. In terms of over- all average retired pay, in mid-1974 military retirees re- ceived $452 per month; whereas, separately by officer and enlisted categories, the rates were $691 and $339 respec- tively. Table 4 provides total military retiree strength figures, plus those of the separate services for the fis- cal years 1972 through 1975 (Army Times, 1972-1976). Table 4. Military Retiree Strengths FY 1972-1975 1972 1973 1974 .1975 Army 327,037 327,674 343,263 354,031 Air Force 274,744 310,161 333,948 363,701 Navy 253,116 248,776 256,677 272,381 Marines 46,997 48,661 52,061 53,810 Total 901,894 935,272 985,949 1,043,923 Total 1976 Strength -- 1,130,225 (Estimate) Source: Army Times, 1972-1976 Although Table 4 shows a continuing annual increase in the total numbers of retirees, the annual input from the active military services is now declining (Figure 8). As indicated in Figure 8, yearly military retirements peaked in 1971, 79,252 retirees being added to the rolls that year. Disability retirements resulting from the Viet- nam War were responsible for this earlier than expected peaking, which according to previous estimates should have come about in 1974 or 1975, 30 years after the height of 53 4b on O O I NUMBER OF RETIREES IN THOUSANDS IS I «* I, \‘\ zo - \ / ‘~ \v ° \ ‘0\ I” / \. \\\ ‘9 ”I e \ \\‘ '- 0”’ / .\ ‘\\ .-—-. ’1 . ____,\_..\_ "' \-\,-,7“~9551°.E'3—-J \--——'-* '0 '- ---—-"’” \‘\\ I” ‘~-,_ ‘-4 J l I J I l I l I J I965 I967 l969 I9TI I973 l975 FISCAL YEAR Source: Army Times I975 Figure 8. Annual Military Retirement Trends (1964-1975) 54 the military buildup of World War 11. Future normal re- tirement figures, estimated on the basis of the current size and retention rates in the active armed forces, call for about 47,000 retirements annually, including almost 8,000 officers and 39,000 enlistees., Of this number, ap- proximately 40,000 would be non-disability and 7,000 dis- ability retirements (Army Times, 1975, p. 19). Despite the evident future annual decline in retirees, the overall total numbers continue to increase and the active military establishment remains of sufficient size to consistently supply a sizeable number of military retirees who could be- come international retirement migrants. World Distribution of Military Retirees In a broad spatial context, American military retirees are found within either a domestic or foreign environment. In the migration system and schema (Figures 3 and 4), the donor component represents the domestic environment; i.e., the fifty states; whereas, the recijient component is rep- resentative of all other locations. It is, therefore, per- tinent to examine the current distribution of the military retirees according to these criteria, beginning with a worldwide overview. 55 Total Numbers In terms of the number of military retirees located in the United States versus those living in foreign areas, there there is no question as to which predominates. Since mid- year 1973 the ratio of military retirees living outside the United States to those in the United States has been approx- imately two out of every hundred retirees (Table 5). Table 5. Percentages of Military Retirees in Foreign Areas W —:_ ___,———_______—— Year Total Retirees No. Outside U.S. Percent 1973 935,272 16,627 1.8 1974 985,949 17,440 1.8 1975 1,043,923 17,649 1.7 Source: Army Times, 1973-1975 According to 1975-76 statistical data obtained from the personnel departments of the armed services, military retirees were found in most areas of the world (Table 6). The total number of retirees in this table (22,997) differs substantially from that given in Table 5 for fiscal year 1975 (17,649), since it is a more detailed compilation of known numbers at different end of month periods in late 1975 and mid 1976. The breakdown by separate states of the United States for fiscal year 1975 is given in Table 7. Aggregating the figures from these tables categorically by major geographic regions provides an abbreviated form for 56 Table 6. Numbers of Military Retirees by Foreign Countries =.===.__.=._=._=.{_ ::3===.=======.======= Air Marine Country Army, Navy_f Force Corps Total Afghanistan 2 2 Am. Samoa 26 73 99 Antigua l 1 Argentina 2 4 l l 8 Australia 68 113 8 22 211 Austria 75 5 12 1 93 Azores Is. 8 8 Bahamas 6 12 3 1 22 Barbados 2 2 Belgium 49 10 14 73 Belize 5 5 Bermuda 3 6 4 l 14 Bolivia 3 1 4 Botswana 1 1 Brazil 11 10 9 l 31 Canada 161 129 6 23 319 Canal Zone 272 110 14 8 404 Ceylon (Sri Lanka) 1 1 Chile 2 2 China 16 16 Colombia 6 5 5 16 Costa Rica 24 13 1 3 41 Cuba ,1 1 Cyprus 1 l 2 Czechoslovakia l 2 3 Denmark 15 7 3 25 Dominican Rep. 4 l 1 6 Ecuador 8 7 2 17 Egypt 2 l 3 El Salvador 1 l 2 4 Ethiopia 5 3 1 11 Table 6 (cont'd.) 57 Air Marine Country Army, Navy Force Corps Total Fiji 1 1 Finland 1 1 l 3 France 175 31 17 2 225 French Guiana 1 1 Fr. Territory 2 2 Germany 2,916 29 712 12 3,669 Greece 30 12 70 4 116 Greenland 2 5 7 Guam 193 554 161 908 Guatemala 1 5 3 2 ll Haiti 1 l 1 3 Honduras 3 3 6 Hong Kong 9 l9 6 2 36 Iceland 2 2 India 1 2 3 Indonesia 8 7 l 2 18 Iran 132 25 3 20 180 Ireland 42 13 111 6 172 Israel 10 4 4 18 Italy 145 58 32 235 Ivory Coast 1 1 Japan 320 160 235 33 748 Jamaica 2 4 6 Jordan 4 l 5 Kenya 1 1 Korea, 80. 302 6 48 4 360 Kuwait 1 l ILaos 4 4 Lebanon 3 4 2 9 lLesotho l l JLiberia 2 l 2 5 58 Table 6 (cont'd.) W Air Marine Country_» Army Navy Force Corps Total Libya 3 3 Luxembourg 3 3 Malagasy l 1 2 Malaysia 1 2 1 4 Malawi 1 1 Mexico ~ 113 94 60 18 285 Morocco 6 2 2 2 12 Nepal 1 l Neth. Antilles 5 5 Netherlands 26 5 20 l 52 New Guinea 1 1 New Zealand 14 55 3 18 90 Nicaragua 2 l 3 Nigeria 1 2 3 Norway 1 3 6 l 11 Okinawa * 10 55 37 102 Pakistan 2 2 Panama 14 6 2 l 23 Paraguay 1 l 2 Peru 1 3 3 7 Philippines 1,240 3,161 100 21 4,522 Portugal 3 5 1 9 Port . Timor 2 2 Puerto Rico 5,694 257 523 53 6,527 St.Pierre & Miquelon l 1 Saudi Arabia 48 8 111 8 175 Sikkim l 1 Singapore 13 12 6 5 36 Society Is. 2 2 South Africa 6 3 2 11 Spain 129 85 268 19 501 Table 6 (cont'd.) 59 Country Army Navy ngze “33:3: Total Swaziland 3 1 4 Sweden 6 12 3 22 Switzerland 19 7 18 45 Syria 1 1 Taiwan 38 22 14 8 82 Tanzania 1 1 Thailand 108 3 205 8 324 Togo 1 1 Pacific Trust Terr. 29 4 28 61 Tunisia 1 2 3 Turkey 22 53 3 78 United Arab Emirates 1 1 United Kingdom 213 154 1,074 9 1,450 Uruguay 1 3 4 venezuela 7 3 2 12 Vietnam, So. 26 6 58 88 Virgin Is. 105 61 33 10 209 Western Samoa (Br.) 1 1 Yemen 1 1 Yemen, Republic 2 2 Yugoslavia 4 4 Zaire 2 2 Totals 12,984 5,4623 4,169 382 22,997 *Number not known--inc1uded with Japan aCountry breakdown for 328 Army Post Office (APO) and 432 Fleet Post Office (FPO) addresses not known EBOurce: Data received in 1975 and 1976 from the separate service departments of the Department of Defense. Table 7. Numbers of Military Retirees by States FY 1975 m State Total State Total Alabama 22,690 Montana 2,844 Alaska 2,824 Nebraska 5,704 Arizona 20,955 Nevada 6,963 Arkansas 12,287 New Hampshire 5,279 California 171,007 New Jersey 20,683 Colorado 23,295 New Mexico 9,269 Connecticut 8,502 New York 31,200 Delaware 3,199 North Carolina 29,814 Florida 88,779 North Dakota 1,237 Georgia 36,970 Ohio 23,312 Hawaii 7,205 Oklahoma 17,226 Idaho 3,364 Oregon 12,476 Illinois 21,033 Pennsylvania 29,184 Indiana 11,046 Rhode Island 6,510 Iowa 4,781 South Carolina 21,021 Kansas 10,076 South Dakota 1,832 Kentucky 12,288 Tennessee 21,464 Louisiana 16,790 Texas 87,736 Maine 5,424 Utah 4,272 Maryland 27,048 Vermont 1,790 Massachusetts 17,549 Virginia 56,094 Michigan 14,149 Washington 31,948 Minnesota 8,155 Washington, D.C. 5,371 Mississippi 11,736 West Virginia 5,206 Missouri 17,039 Wisconsin 7,935 Wyoming 1,713 Total 1,026,274 ‘ Source: Army Times, 1976. 61 global comparison (Table 8). Since the United States is included in the Anglo-America area, it dominates all the others. Otherwise, Asia, Latin America, and Europe appear somewhat equal in numbers, with Australia and Africa well below in the listing. Outside the United States, it is obvious that the principal numbers of retirees appear in territories of the United States or in those areas in which the United States has maintained major active military bases and installations. It is interesting to note that each foreign country (excluding United States territories) attracting the most retirees of a particular armed service is also the nation which contains a large number of active military personnel of that same service; i.e., Army--Ger- many, Navy--Philippines, Air Force--United Kingdom, Mar- ines-~Okinawa. Further attention will be given to the 10- cational factors in subsequent analyses. Total Pensions The total pensions dispersed to retired military per- sonnel in fiscal year 1973 amounted to $4,343,640,000. or this sum, $160,352,240, or 3.7 per cent of the total, was received by retirees residing outside the United States (Table 9). Comparatively, the dollar amounts accounted for in foreign areas equal just over three times that for the state of Michigan, or about one and a quarter of that received in the state of New YOrk. 62 Table 8. Military Retirees by Geographic Region Region Numbers Anglo-America 1,026,848 United States 1,026,274 Canada 319 Others 255 Asia & Pacific Areas 7,876 Philippines 4,522 Guam 908 Japan 748 Korea 360 Others 1,338 Latin America 7,446 Puerto Rico 6,527 Canal Zone 404 Mexico 285 Central America 70 (excl. Panama) Others 160 Europe 6,732 Germany 3,669 United Kingdom 1,450 Spain 501 Italy 235 Others 877 Australia 301 Australia 211 New Zealand 90 Africa 68 Total 1,049,271 Source: Based on data obtained from the separate services of the Department of Defense in 1975 and 1976. 63 Table 9. Retired Military Personnel Pensions, 1973 ..==: State Pzzgiins State nggiins Alabama $ 80,888,720 Montana $ 9,516,320 Alaska 7,786,080 Nebraska 20,330,320 Arizona 81,753,840 Nevada 22,060,560 Arkansas 43,688,560 New Hampshire 19,465,200 California 789,854,560 New Jersey 86,944,560 Colorado 96,028,320 New Mexico 37,632,720 Connecticut 36,767,600 New YOrk 126,740,080 Delaware 10,814,000 North Carolina 96,028,320 Washington, D.C. 34,604,800 North Dakota 3,460,480 Florida 381,950,480 Ohio 83,916,640 Georgia 134,958,720 Oklahoma 62,288,640 Hawaii 21,628,000 Oregon 44,121,120 Idaho 13,409,360 Pennsylvania 117,656,320 Illinois 79,591,040 Rhode Island 21,628,000 Indiana 40,228,080 South Carolina 83,041,520 Iowa 18,600,080 South Dakota 6,055,840 Kansas 35,902,480 Tennessee 65,749,120 Kentucky 42,390,880 Texas 363,782,960 Louisiana 59,693,280 Utah 16,869,840 Maine 20,330,320 Vermont 6,488,400 Maryland 133,661,040 Virginia 276,838,400 Massachusetts 78,293,360 Washington 133,661,040 Michigan 50,176,960 West Virginia 15,572,160 Minnesota 29,414,080 Wisconsin 29,846,640 Mississippi 39,795,520 Wyoming 5,623,280 Missouri 65,749,120 Total U.S. $4,183,287,760 Outside U.S. 160,352,240 Total Pensions to Source: Army Times, 1973 Retired Militarnyersonnel $4,343,640,000 64 Significance The socio-economic impact of these retirees on the host nations varies greatly, of course, based on the num- bers involved and the corresponding capital which they represent in the form of United States government pensions. In the case of highly populated and developed nations (Germany, Japan, England, etc.) American retirees are usu— ally widely dispersed and their presence among the populace is less conspicuous. Moreover, their pensions have little overall effect upon the economic bases of these states. Conversely, in the small less-developed countries (Costa Rica, Portugal, Greece, etc.) even a few hundred retirees could affect the social structure and strengthen the local economy, especially if clustered in retirement colonies or concentrated at particular sites. United States Most military retirees, as should be expected, normally settle in the United States, the donor component of the system (Figure 3). Since this component Operates as the provider of international retirement migrants, it is im- portant to examine its distributional aspects for a better ‘understanding of the overall functions of the system. 65 Distribution As of 30 June 1975, there were 1,026,274 military re- 'tirees residing in the United States, including Hawaii and Alaska (Table 7 and Figure 9). Of the total retirees, about 40 per cent were located in four states; i.e., California (171,007), Florida (88,779), Texas (87,736), and Virginia (56,094). The military significance of these four states is their association with the numerous military posts and bases of all the armed services which are found there. The attraction of military bases for retired personnel is the availability of facilities such as commissaries, post ex- changes, hospitals, clubs, recreation areas, etc., which remain open to the retiree. Consequently, those states offering numerous military facilities, combined with favor- able climatic conditions, job opportunities, and accept- able tax structures can expect to attract the most retirees. The states with fewest retirees are those in the north- ern interior areas of the United States where few military facilities exist, job opportunities are limited, and cli- mates are unfavorable for retirement life. Most obvious among these states are North Dakota (1,237), South Dakota (1,832), Wyoming (1,713), Montana (2,844), Idaho (3,364), Utah (4,272), and Iowa (4,781). In the states more climatically attractive for re- tirement, military pensioners have concentrated in and around specific urban areas that contain a complex of 66 Aeneas moenem enema: one am mooeaeom aaneeflas .m oanmam . _ _ $522.5; >53. Hooaeow own com 000. ‘Oon 00F 00¢ 000 b 000. h on: 5 coo? 5:6 I ooo_omsooo.on E ooo.onuooo.o~ @ ooo.o~..ooo.o_ E 086708.... a cookie D: 67 military installations. Most notable of these so-called "service towns" are Washington, D.C. and environs; San Diego, California; San Antonio, Texas; and Orlando, Flor- ida and vicinity. The Washington and Northern Virginia— Maryland suburban agglomeration combines major military facilities and numerous amenities with employment advan— tages that have made it the most popular retirement site for all the military services (50,000 retirees). For full retirement, central Florida, with Orlando and the surround- ing military bases as its nucleus, has drawn about 65,000 retirees from the four services. San Diego, which claims about 40,000 military retirees, has traditionally been recognized as the retirement home of naval personnel and many of the retirees domiciled there completed their ac- tive service at nearby Navy or Marine Corps installations. San Antonio has long been a favorite retirement site for thousands of Air Force and Army persons; the principal at- traction, in addition to the mild climate, is several major military bases located in or near its geographic boundar- ies (Gates, 1974, p. 25). There are, of course, many other popular sites in the United States where retired personnel have concentrated, and their numbers have affected the cor— responding communities socially and/or economically. In many instances, as in the special cases cited, these mili- tary retiree communities or retiree concentrations have developed from a relationship with nearby military install— ations. 68 Location Choice In the United States some military retirees, like most of their civilian retiree counterparts, settle in their hometowns or at the location of their last job site. More, however, have opted for the favorable climatic areas of the south and west, with special consideration given to those states containing major military facilities that provide services for retirees and offer employment oppor- tunities for the younger group. To measure the significance of the retiree-military installation relationship and to examine other factors which might account forthe spatial distribution of re- tirees in the United States, a correlation-regression anal— ysis was performed using the number of military retirees by states (Table 7) as a dependent variable and the follow- ing independent variables: 1. Per cent Population Change Between 1960-1970. It was bélieved thatPtHOSe areas in the United States which have experienced the highest popu- lation growth in the last decade could be expected to offer many economic and social advantages not found in the slower growing regions; therefore, military retirees, as a younger group than the typical 65 year-old plus, would seek out these areas. 2. Number of Major Armed Forces Installations. It was theorized that this factor wouId be a strong one for enticing military retirees into 4These data, as well as those for the other variables, except the Armed Forces Installation data, were obtained from the 1970 Census of Population (USBC, 1970), 69 particular states, since these installations represent both a fulfillment of certain eco- nomic needs as well as social ones by the military person. These bases normally con— tain a hospital or large dispensary, a com- missary, and a post exchange. They also of- fer other amenities such as service clubs, craft shops, theaters, recreational facilities, etc., as well as potential employment.5 3. Population Density (Persons per square mile). This variable was prOposed since it is bélieved that military retirees would search for a more tranquil life in rural areas. 4. Median Family Income. It would normally be expected that large numbers of retirees tend to concentrate in those areas having lower med- ian family incomes; whereas young people, who wish to increase their incomes, migrate into areas with higher income bases. Military retir— ees, because they are younger as a group than the average retiree and often have children to pro- vide for and educate, might be concentrated in the higher median income areas. 5. Per cent of the Civilian Labor Force Unem- lo Ed. It is generally assumed that those areas EavIng high unemployment percentages are less attractive to migrants. This was also the as- sumption used in the case of military retirees, the rationale again being their young retirement age and their pursuit of a second career. 6. Per cent of ngulation Urban. Urban popula- tions are more generally associated with greater employment opportunities and provision of ameni- ties than non-urbanized areas. Highly urbanized areas, therefore, could be expected to strongly attract migrants. The correlation-regression test provided data to com- pile a table of simple correlation coefficients (Table 10). 5These data were extracted from Handbook on Retire- pent Services, Headquarters, Departmefit'of’the Army'(USDA, 1975). 70 ooo.H MbN.o ooo.H www.0I b®N.cI coo.H mmv.on mam.OI mmm.on cmo.on mm¢.o omH.O coo.H voo.0I ooo.~ mmv.on moo.OI mbv.o 900.0 NON.O coo.H mem.ou eao.ou eoH.o moo.o .mmmqm Hou.o ooo.H cans: .Qom 8 oomquaoaD $ oaoocm .ooz huamcoo mco«anHNpmoH Omsazo .aom fl moomfiaom ARV MOHDdfioHd> “@0398 GOfiHflHOHHOU HO mHGOHOflHHOOU .CH OHDdH. 71 From this table, it is apparent that only one independent variable, Armed Forces Installations, shows a significant relationship (r = 0.932) and proportion of explained varia- tion (r2 = .868) with the dependent variable. A signifi- cance test of the r value revealed significance at the .05 level and therefore assumption of a linear association. A multiple correlation-regression analysis, combining all variables, gave similar results, with slightly higher R and R2 values. However, an examination of the Beta values (standardized regression coefficients), which represent the slope of the multivariate figure, revealed that again the only variable of significant explanation was Armed Forces Installations (Table 11). Table 11. Significance of the Beta values Variable Beta Value 1 % Pop. Change Between 1960-70 0.06198 2 Armed Forces Installations 0.93121 3 P0pu1ation Density 0.11319 4 Median Family Income -0.11130 5 % Civ. Labor Force Unemployed 0.01939 6 % Pop. Urban -0.00534 + It is evident that the Armed Forces Installations factor has remained dominant throughout the analysis. Con- sequently, these installations account significantly for ‘the spatial distribution of military retirees in the 72 United States. There are, of course, other factors of a socio-economic nature such as job Opportunities, climatic conditions, and the presence of amenities which could fur- ther explain the associations. For purposes of this study, the foregoing analysis indicates that retirees do concen- trate in and around military installations and organiza- tions where they can relate common experiences among them- selves and members of the active military organizations. Information about foreign retirement places, as well as the advantages and disadvantages of locating in many dif- ferent areas, is exchanged between these groups. Foreign Military retirees who make the decision to leave the United States, the donor component of the migration system (Figure 3), pass from that component as international re- tirement migrants into foreign nations, the recipient com- ponent of the system. These migrants, although accounting for only about two per cent of the total number of military retirees, have located throughout the world. As a percen- tage of the estimated 300,000 Americans retired abroad (Fischer, 1973, p. l), retired military persons represent about eight per cent. This is a significant proportion, considering that retired military members number just over a million. An examination of their general distribution is 73 important, therefore, to show the extent of retiree mobil- ity and for comparison with the Central American retirees. Distribution In 1976 approximately 23,000 military retirees resided in 115 foreign nations and territories, including those of the United States (Table 6). These retired persons are concentrated principally in three major areas: the West- ern Pacific region, Western Europe, and Middle America (Fig- ure 10). A ranking of the first twenty countries and ter- ritories by percentages of the total numbers of retirees living outside the United States, and the dominant mili- tary service for each place, is given in Table 12. Puerto Rico and the Philippines alone account for almost 50 per cent of the retirees; whereas, all twenty countries in the listing represent 94 per cent of the foreign total. Since Puerto Rico, Guam, the Canal Zone, and the Virgin Islands, all appearing in Table 12, are United States territories, as well as American Samoa and the Trust Territory of the Pacific (Figure 10), their interest to retired service per- sonnel is related mostly to factors similar to those con- sidered by retirees located in the United States. In the strict sense of foreign nation retirement, excluding the international migration between the United States and its overseas territories, the major military retirement coun- tries are the Philippines, Germany, the United Kingdom, 74 Aesoflv maoae amaoaom ea nooaaeom saaoaees .oH oaomfim St e288 3 20.2.2.8 32.8» 06. 3...: no 33a owns»: no 52:... . coo. ¢u>o . O 30.13 . av 0313 . O of an“ a . 75 Table 12. Ranking of Retirees by Country Rank Country Percent D;::3:2: 1 Puerto Rico 28.5 Army 2 Philippines 19.8 Navy 3 Germany 16.0 Army 4 U.K. 6.3 Air Force 5 Guam 4.0 Navy 6 Japan 3.3 Army 7 Spain 2.2 Air Force 8 Canal Zone 1.8 Army 9 So. Korea 1.6 Army 10 Thailand 1.4 Air Force 11 Canada 1.4 Army 12 Mexico 1.2 Army 13 Italy 1.0 Army 14 France 1.0 Army 15 Australia 0.9 Navy 16 Virgin Is. 0.9 Army 17 Iran 0.8 Army 18 Saudi Arabia 0.8 Air Force 19 Ireland 0.8 Air Force 20 Greece 0.5 Air Force Source: Data obtained from the Personnel Sections of the Army, Navy, and Air Force Departments and the Marine Corps, 1975-1976. 76 Japan, and Spain. Here again, the Philippines can be partly discounted as a foreign retirement area because of its long history as a United States territory, whose in- habitants were permitted to serve in the United States Armed Forces, mostly as mess attendants in the Navy or as Filipino Scouts in the Army. Most of those who served returned to their homeland after retirement. The dominant remaining nation in the ranking, then, is Germany (16 per cent) followed by several other nations of Western Europe and East Asia, countries in which the United States main- tains significant military forces in accordance with bi- lateral status of forces agreements. The one nation ap- pearing in Table 12 that has witnessed no exposure to re- cent United States military activity, or maintains Ameri- can bases on its territory, is Mexico. Apparently, fac- tors other than those associated with military facilities are responsible for its popularity as a retirement area. Location Choice A number of military retirees selected sites simply on the basis of their adventurous nature and desire to settle in some exotic or unfamiliar place. The few re- tirees scattered throughout Africa, South Asia, and South America could reflect this situation (Figure 10). Others have chosen a favorite foreign nation in which they pre- viously served during active military duty, a place for 77 which they retained many fond remembrances and experiences. Some have made the choice on the basis of favorable retire- ment environments, including mild climates and other amen- ities, as well as the companionship of fellow military re- tirees. The Mediterranean and Middle American areas offer these advantages. The majority of military retirees located outside of the United States, however, probably owe their presence in a foreign nation to kinship ties and job oppor- tunities, as exemplified by Western EurOpe and East Asia. In many instances, those retirees working abroad are em- ployed by the United States government at military install- ations which can provide them benefits similar to those en- joyed by active military personnel stationed there. Mili- tary retirees not employed by the United States government are not authorized full use of military facilities by most foreign governments. For example, the United Kingdom, Italy, Turkey, and the Netherlands allow retirees no mili- tary related privileges; Germany and Korea permit limited use of the military exchange and commissary facilities; Ja- pan and Okinawa authorize hospital and dependent school use, while Spain has few restrictions for retirees. In all countries where United States military troops are stationed, retirees are given base privileges that do not violate the host nation's customs and tax laws. These include the use of clubs and transient quarters, recreation facilities, laundry and dry cleaning services, military clothing sales 78 stores, and legal assistance, as well as travel by govern— ment aircraft on a space available basis (Leonberger, 1973, p. 25). The United States territories, of course, provide retirees the same services as those allowed at military in- stallations in the United States, accounting partly for their attractiveness to retirees. Other than the foreign nationals and other non-resi- dents who have served in the United States military and re- turned to their homelands after retirement (Puerto Rico, Philippines, Guam, etc.), the principal numbers of American military retirees attracted to foreign areas for kinship reasons are found in those nations where large contingents of military forces are, or have been, stationed. Chief among these places are Germany, Japan, South Korea, United Kingdom, Italy, and Spain. Many of the military personnel serving in these areas married local nationals, completed their military careers and then returned to their spouses' country for retirement. Besides employment with the United States government in defense-related jobs or other governmental agencies (U.S. Agency for International Development, U.S. Informa- tion Agency, etc.), military retirees have also been hired by large American firms operating overseas, as well as by some foreign governments. Recently, the nations offering retirees the most lucrative employment opportunities are in the Middle East, especially Iran and Saudi Arabia, who 79 are using retired military persons in various technical, training, administrative, and educational positions. These nations have no status of forces agreements in effect, which means that retirees can enjoy the added benefits of existing American military facilities, most notable of which are those located in the Teheran, Iran area (Army Times, 1975). Nations with whom America shares a similar culture, political foundation, heritage, and a common language are the choice of those retirees who wish to live in a foreign country with minimum political and social risks. The United Kingdom, where an estimated 12,000 American retirees, 12 per cent of them military, have settled (Fischer, 1973, p. 3), is the most popular place for retirees in this group, especially for retired Air Force personnel. The presence of major United States air bases is a contributing factor. Other nations in this category include Canada, Australia, and New Zealand, all of which have joint military agree- ments with the United States for the location of military personnel and facilities on their soil. Spain, a nation which ranks seventh in terms of mili- tary retirees, has attracted over 500 military retired per- sons, many of whom have settled in the Costa del Sol and other areas along the Mediterranean or in the Madrid area. For military retirement, Spain has much to offer. Several major United States military facilities which can be used 80 by retirees are located in that country; there are direct military space available flights from the United States; it is a low cost of living area; it offers pleasant climate and cultural attractions; and it provides special privi- leges to resident pensioners for the importation of house- hold goods and effects. With the exception of United States territories, few areas provide the advantages that are ap- parent in Spain. Although it still is a favorite retire- ment site for Americans, its changing political situation may curtail future retirement migration, with possible re- direction toward similar environments in Latin America. The Mediterranean areas of France, Italy, and Greece also remain popular as foreign retirement centers. Their at- tractiveness for the military, however, is waning, because of increased living costs and restrictions on use of United States military facilities. Migration Potential For migration to occur, it is necessary to have a pool of potential migrants and attracting forces which act to spatially redistribute this mass. The distribution pattern of retirees (Figure 10) illustrates the result of this movement in that the most attractive places for retirement migration are those that contain the major concentrations of retirees. Using the number of United States retirees as a potential migrant base, then, and the numbers of 81 retirees at each foreign location as modes, interaction could be determined by multiplying the two figures and di— viding by the intervening spatial distance (gravity model). This mathematical expression, if implemented, would undoubt- edly show that those foreign areas containing the largest numbers of retirees and located the shortest distance from the United States have the highest potentials for continued migration. For a more realistic portrayal, however, var- iables pertaining to physical, political, and economic fac— tors are required as constants in the formula to account for such restraints as travel over water surfaces versus roads, political restrictions for migrant entry, and high cost of living areas. Nonetheless, once a sizeable number of migrants locate at a particular place, their presence and activities are projected to other potential migrants who may decide to join them. Consequently, any foreign area in which military retirees are located (Figure 10) represents a potential for further retirement migration. The growth pole aspect of this condition at specific sites is illustrated by the formation of foreign chapters of the Retired Officers Association, an organization of some 237,000 members. Once sufficient numbers of military retirees concentrate in a popular retirement place, inter- action generally leads to social organization, which for retired officers could be represented by a local chapter of the national Retired Officers Association. By 1972 this 82 association had chapters in San Juan, Puerto Rico; St. Thomas, Virgin Islands; Seoul, Korea; and Naha, Okinawa. Chapala, Mexico was added in 1972, followed in 1974 by Guadalajara, Mexico and Teheran, Iran. The latest addi- tions were London, England and San Jose, Costa Rica in 1976. The possibility of organizing and maintaining mem- bership in local chapters of this nature must be attributed to a concentration of retirees in those localities. Con— sequently, this concentration effect enhances their migra— tion potential. Middle America Middle America, a geographic region which includes Mexico, Central America, and the West Indies or Caribbean Island complex, represents one of the three major world re- gional concentrations of retirees, along with Western Europe and the Western Pacific area (Figure 10). Since military retirees have located extensively throughout this area and many of the nations share similar locational factors for these migrants, a general survey of the region is presented prior to focusing on Central America. General Distribution Middle America, like the Mediterranean area, has at- tracted substantial numbers of retirees because of its 83 favorable climate, natural scenery, and other environmental amenities. Unlike the Mediterranean countries, however, those of Middle America have the advantage of being more accessible to the United States and, in most instances, have lower costs of living. Consequently, as the climat- ically favorable areas of the United States (Florida, Cal- ifornia, and Texas) become overcrowded and less desirable for retirement purposes, Middle American nations, which offer similar retirement-related amenities, can be easily reached for pre-retirement inspection by potential migrants. In the Caribbean Antilles, Puerto Rico dominates all other countries in numbers of military retirees (Figure 11). Most of these retirees are native Puerto Ricans with Amer- ican citizenship who have returned home, although a number of them are mainland Americans who previously served in Puerto Rico, liked it, and later retired there. Others chose the area on the merits of its overall advantages for retirement; i.e., climate, military facilities, and social amenities, to mention a few. The American Virgin Islands, ranking second in the Antilles and also a United States territory, offer retirement advantages similar to those en- joyed in Puerto Rico. The remaining islands of the Antilles have attracted few military retirees, the majority having located in the English speaking nations, especially the Bahamas off-shore from southern Florida. Cuba undoubtedly would have a greater concentration of retirees if its 84 Awbmav noeaoae omen“: ea uooaeooe sesame“: .HH nausea [OON IoOn 000 out 005 .me 060 _ .oo a&_ 45. 3:33 .0 2.255an ”eoneom — m 0.. law mu4_: 00—1 . m /e ‘ .\\.~ . AI. 2 .FI.W new KN: .Ilfil/ :ILIIII. $51 /0 J. eon coo. ac: 85 political philosophies were patterned along democratic lines. United States policy, incidentally, forbids payment of government pensions to persons living in any communist controlled nation. On the Middle American mainland, the United States- governed Canal Zone contains the largest number of retirees (404), followed by Mexico (285) and Central America (93) (Table 6). Utilizing the field survey data collected in this study, the figures for Central America, excluding Pan- ama, are increased to 197 (Table 2). These latter figures will be used in subsequent analyses for Central America. Mexico and Panama Mexico and Panama (including the Canal Zone), border- ing Central America on the north and south, have attracted sizeable numbers of retirees for differing reasons. Since these two countries are not included in the field survey, but yet are important adjuncts in the study analysis, they are given general coverage. Distribution Due to the proximity of Mexico to the United States, the inherent advantages of its laws regarding migrants, and the combination of its varied scenic beauty and agreeable highland climates, numerous American retirement communities have developed there. Included among the retired Americans in these colonies are 285 military peOple. The specific 86 locations of all 285 retirees are not known. However, based on locational data for 112 of them, it has been determined that these migrants are concentrated in the central highland region of the country (Figure 12). Furthermore, over fifty per cent of the military retiree migrants are located in the state of Jalisco and only three additional states con- tain significant percentages of the total; i.e., Guanajuato, 9.6%; Morelos, 7.0%; and the Distrito Federal, 6.1% (Table 13). The major American retirement communities of Mexico are all located within those four states. In fact, the largest urban American community outside of the United States is found in Guadalajara, in the state of Jalisco. Also in Jalisco is the Lake Chapala area, including the towns of Chapala and Ajijic, which accommodates the larg- est non—urban settlement of American expatriates. In the state of Guanajuato are located the popular retirement com- munities of San Miguel de Allende, one of Mexico's oldest colonial towns, and Guanajuato, long noted as a quaint sil- ver mining center. The well established retirement town of Cuernavaca is in Morelos and the major cultural center, Mexico City, is in the Distrito Federal. Unlike Mexico, which draws many American retirees due to its proximity to the United States, Panama owes its re- tirement migration attractiveness to the Canal Zone. 0f the 427 military retirees found in Panama, only 23 have located in the Republic. The 404 retired military persons 87 ICON 8me cosmos ca mommauom humped“: .NH shaman _ oom- eo:e_ooeo< 200:3 pesto: on... ”0030» “3.2.322. a. .2323 on .26 . o. 2..“ a case. a 32.. fl ozuou... OONI 88 Table 13. Per cent of Retirees in Mexico by States State Per cent Nayarit 1.7 Queretaro 0.8 Jalisco 64.1 Puebla 1.7 Morelos 7.0 Distrito Federal 6.1 Guanajuato 9.6 Tamaulipas 0.8 Aguascalientes 0.8 San Luis Potosi 1.7 Sonora 1.7 Baja California (N) 0.8 Guerrero 0.8 Michoacan 0.8 Yucatan 0.8 Colima 0.8 Total 100.0 Source: The percentages in the above table were calculated from data on 112 military retirees which were ob- tained in June of 1973 from the Retired Officers Association. 89 residing in the Canal Zone are there principally for em— ployment. Although other reasons may play a part in their location decision, the dominant importance of employment must be accepted, since American residency in the Canal Zone is authorized only to those working there. Conse- quently, retirees are concentrated in the Pacific and At- lantic coastal complexes near the Zone's military facili- ties, along with those of the Panama Canal government and the Panama Canal Company, the principal employers (Figure 13). Though locational data are not available, it is be- lieved that the few retirees settled outside the Canal Zone live in the principal cities adjacent to the Zone (Panama City and Colon), along the Pacific beach area southwest of the Canal Zone, or in the highlands to the west, with the scenic Boquete area as the center of attraction. Historical Perspectives In mainland Middle America, the foundations for re- tirement migration were laid in both Mexico and Panama at earlier stages than in the Central American area. Mexico, since it borders with the United States, provides easy ac- cess routes for migrants and minimized distances for the diffusion of retirement-related information. Its merits as a favorable retirement area, therefore, have been widely known throughout the United States, especially since World War II. With an influx of American veterans into Mexico after the war, followed in the late fifties and sixties by 90 Aoemev noeaoa< Haaeeoo ea unomenoom oooeoom seaweed: .mH oasufim 5.33. 2.39539 a 232:. .m 2322. ON 2.. . o 28: no.2o . K/r..\. 2.023. m2~ . q _ 0mm 3:30. 565. .o 32:2 3. .2 22. Stem “ooaeom . q q . . . com 00. o $03200 m l- 3 .22.. on w . com. o osteoao 2.2.8.. % Eon 2.....ch O . 2302 no. I I o. 3.325 0.9.3:! 2 a 3.250 :2. com 8.53... 3.322.. .\ O \ 3.3.2.3... . 820... O .:~< 26¢ a \ 0 ON loo. ozwoum as... 5.. \ 2.6 2.3.. s \ \I I II ”/ Ru Id . " I -' L 000 00$ 1 .p F _ 393.3 com I)! e. / ” 91 retirees seeking comfortable and inexpensive settlement, American colonies have developed in the climatically fav- ored regions. The buildup of the military retirement ele- ment in these places was evidenced by the formation of Re— tired Officers Association chapters at the pOpular sites around Lake Chapala in 1972 and Guadalajara in 1974. Con- sidering the fact that only seven other chapters of this organization exist outside the United States, including two in its territories (Puerto Rico and the Virgin Islands), the existence of these chapters attests to Mexico's popu- larity for military retirees. American military personnel have been exposed to Pan- ama since 1903, when the United States assisted that nation in achieving independence from Colombia. Since then, in accordance with the United States-Panama treaty of 1903, the United States has maintained military forces in the Canal Zone. During World War II, military and naval bases out- side the Zone were also negotiated and manned by American military forces; e.g., Rio Hato Air Base and Taboga Island Naval Base (USDA, 1972, pp. 225-226). The attributes of Panama for retirement, therefore, have long been dissemi- nated amOng the military population, including those of re- tired status. As a consequence, significant numbers of military retirees have taken advantage of the Opportunities Offered there, particularly in the Canal Zone. The migra- tion inertia built up over time in movements from the United States to both Mexico and Panama will contribute 92 toward continued migration, unless the major advantages are negated; e.g., Mexican border-crossing restrictions or United States loss of the Canal Zone. Incentives and Deterrents Mexico, besides the beneficial aspects of its accessi- bility to the United States and lower cost of living, of- fers retirees a variety of natural scenery and climates, combined with favorable governmental migration policies. The central section of Mexico, a highland plateau referred to as the Meseta Central, is the political, economic, and population center of the country. It also has the most favorable climate in Mexico, i.e., temperate highland, a type which attracts retirees. Although other areas of Mex- ico offer scenic and climatic contrasts, mountains and coastal lowlands, few retirees have located outside this temperate highland region. In accordance with Mexican law, retirees can live in that nation as tourists, visitors, or immigrants. The qualifications, responsibilities, rights, and privileges vary greatly in each of these migrant categories. For ex- ample, the allowable length of residency without renewing entry permits is six months for a tourist, two years for a visitor, and indefinite for an immigrant (For details refer to Ford, (1974), pp. 43-50). Although the environmental and political attributes of Mexico are appealing to many retirees, it is recognized 93 that for Americans to successfully retire in that nation, they must adjust their way of life to certain of the Latin cultural and social conditions. In these circumstances, many Americans are unwilling to accept a position as a mem- ber of a minority group, rejecting the Latin racial, relig- ions, and linguistic approaches to their cultures, and thereby becoming totally disaffected as foreign retirement migrants. This is less likely to happen in the case of mil- itary retirees, but nonetheless is something that they must face and deal with in a realistic manner. Other than the common inconveniences that all retirees encounter in the Latin countries, i.e., sanitation, shopping, driving, cus- toms and immigration laws, security, etc., those privileges most missed by military retirees in Mexico are the shopping and medical privileges provided by American military in- stallations. Most critical for older retirees, of course, is the provision of medical facilities, since United States Medicare is not applicable in foreign areas and military retirees are not eligible for CHAMPUS benefits after age 6 65. Consequently, those retirees not willing to relinquish the conveniences inherent in North American culture or who 6The Civilian Health and Medical Program of the Uni- formed Services (CHAMPUS) authorizes medical care for re- tired military members and their dependents at civilian facilities when use of military facilities is not feasible. The retiree pays 25 per cent of the costs for hospitalizaa tion and professional services, plus the first $50 each fiscal year for outpatient care (USDA, 1975). 94 desire access to military and medical facilities are de- terred from Mexican retirement. Panama offers similar advantages for retired life. Although it does not contain the large diversity of topo- graphy and climate found in Mexico, its small size and pe- culiar geographic "S" shape, bisected by the Canal Zone, provides a varied and interesting landscape. In addition, it lies outside the major earthquake and hurricane zones found in Central America and Mexico. Unlike Mexico and the other Central American countries, where principal settle- ment has taken place in the more pleasant highlands, Pan- ama's population center developed and has remained around the central isthmus lowlands, with the Canal as the focal point. The more favorable climatic areas of its highland interior have remained sparsely populated, thus represent- ing virgin territory for those retirees seeking a pleasant and remote retirement environment. The Republic of Panama has tourist laws which provide for temporary residency by American citizens, whereby a retiree may remain in-country up to 90 days as a tourist. To encourage foreign retirees on a permanent basis, in 1970 Panama implemented a law which gives qualified retirees special residency benefits. The law applies to any foreign national who receives a minimum monthly net income of $400 in the form of a life-time pension from governmental or private institutions located outside Panama. Qaulified re- tirees are given tourist-pensionee visas which authorize 95 them to reside indefinitely in Panama and receive benefits pertaining to duty free importation of certain household goods and a private vehicle (Instituto Panameno de Turismo). Since the promulgation of this law, in mid 1970, through 1975, there were 313 persons, 273 of them Americans, who gained pensionee status.7 Although the numbers of mil- itary retirees in this figure are not known, it is possible that the 23 retirees_identified by the Department of Defense have taken advantage of the pensionee law. Consequently, they represent 8.4 per cent of the American pensionees. A great advantage for those military retirees living in the Republic is the availability of the vast number of military facilities in the Canal Zone which can provide American goods and services, including medical. The ad- vantages for the several hundred retirees who work and live in the Zone are obvious. Within the Zone is a com- plex of military installations which includes six major Army posts, two major Air Force bases, two Navy stations, a Marine Corps post, and numerous miscellaneous facilities serving all the uniformed services. Another benefit pro- vided to these retirees is government space-available air transportation between the Canal Zone and the United States, 7Figures derived from discussions with Enrique Diez, IChief of the Personnel Department, Panama Tourist Bureau, in May 1976. At that time it was estimated that approxi- mately 300 pensionees were settled in the Republic of Panama by mid year 1976. 96 an important consideration since Panama is less accessible by land routes than Mexico or Central America. Despite the many attributes of Panama for retirement purposes, the strained relations between the United States and Panama over the Canal Zone issue have created a tense political situation, which in the past has resulted in dem- ostrations and rioting against Americans in the Zone. This situation is potentially serious in the Panama City area, where most anti-American activities have taken place. The effects have created some dissention throughout the Repub- lic, and will surely be detrimental to increased retirement migration, thus directing some retirees, or potential ones, to similar environments in other Central American nations. Central America From the foregoing analysis, it is evident that most military retirees residing in foreign nations and United States territories selected those areas for reasons related to economics, kinship, common language and customs, or availability of military facilities. Mexico is a major exception; though lacking military facilities, its proxim- ity to the United States and its ability to provide retire- ment-related amenities have been strong pull factors for many years. Since Central America, which has similar re- tirement attributes, is less accessible than Mexico, only recently has it become known as an attractive area for 97 American retirees. Henceforth the examination of military retirees and their characteristics as international migrants will focus on that group in Central America. Moreover, it will be accomplished in accordance with the concepts of the retirement migration schema (Figure 4). Within the recip- ient component of that model, the migrant represents Amer- ican military retirees in Central America; the destination is congruous with the Central American places where these retirees have settled. In Central America, Costa Rica is by far the most pop- ular nation for retirees (Figure 13). From a military re- tiree population of 151 persons in the six countries for whom place location data Were found, 105 were distributed throughout Costa Rica, with 91 of these persons located in the Meseta Central area which centers on San Jose (Table 2).8 Of these 91 retirees, 35 were living in the central city and adjacent settlements, 16 in Escazu, 10 in Alajuela and vicinity, six in Santa Ana, and the remaining 24 scat- tered among the other villages. The other 14 retirees found in Costa Rica are located along the Pacific (Nosara, Punta Cuchilla, Jaco, Golfito) and Atlantic (Cahuita) coast areas, 8The figure of 151 was obtained from Table 2 data by adding the total numbers of retirees in the columns for ngpleted Interviews, Not Contacted, and Refused. 'The specific lOcations for persons in these three categories were known, whereas those for the Not Located category were not. The total numbers for the Individual countries were determined in the same manner. 98 in the General Valley (San Isidro), and a few other isolated areas (Miramar, Quesada, Guapiles, Turrialba). The second country of importance in terms of total retirees is Guatemala with 21, most of whom are located in the capital, Guatemala City (13), and the nearby villages of Villa Nueva (2) and San Juan Sacatepequez (1). Antigua, the old colonial capital of Guatemala and popular tourist center, has three retirees. Solola, above Lake Atitlan, and Zacapa, in the Motagua Corridor, each have one retiree living within their confines.9 Honduras, El Salvador, and Nicaragua each have eight retirees. In El Salvador they are all located in the capital of San Salvador, and in Nic- aragua all reside in Managua, also the capital, except one (Totagalpa). Both cities lie within the volcanic axis of Central America. The Honduras retirees reside in the set— tlements of Choluteca (l), Tegucigalpa (3), Agua Azul (1), La Lima (1), and San Pedro Sula (2), located generally along the north-south oriented Comayugua Depression. In Belize, a self-governing British Colony, only one military retiree, who lives in Belize City, was located. 9The field survey in Guatemala was undertaken during the month of January 1976 and ended abruptly with the major earthquake of February 4, 1976. At that time only three of the known retirees had not been contacted, two in Guate- mala City and one in Zacapa. A later search for the Zacapa retiree was unsuccessful. These figures may have changed since the February 1976 disaster. 99 This retiree locational structure in Central America supports hypothesis 4.a., i.e., the retirees locate mainly in the central cities or adjacent areas connected by the major routes of communication. These are the places where military retirees have settled. Their attributes and attractions for retiree migrants will be the subject of Chapter III. CHAPTER III FACTORS RELEVANT TO CENTRAL AMERICAN RETIREMENT MIGRATION This chapter will examine the various factors which make Central America attractive for retirement, thus pro- viding answers to the why_questions of the retirement mi— gration process. In the migration schema model (Figure 4), these factors are represented by the external environmental conditions (stimuli), as well as certain of the internal control and adjustment mechanisms. Place Utility Place utility can be considered the perceived useful- ness of geographic sites as determined by an individual, depending on that person's satisfaction or dissatisfaction with the place. Different places reflect differences in utility, and those places which provide or promise a higher level of utility are the most in demand (Wolpert, 1965, p. 62).» This concept, of course, applies to retirement sites, evaluated on the basis of their desirable attributes. 100 101 Location Choice Theory Location theory is generally used in geography for determining locations of economic activity. By measuring gross benefits against costs, net benefits at a given place can be derived. Conceptually, then, retirement locations could be determined in a similar manner, considering favor- able factors as gross benefits, unfavorable ones as costs, and the composite value of these factors as the net bene- fit. The various composite values could also represent place utility ordering. Applicability in International Retirement Migration In choosing a foreign retirement nation or locality, a potential migrant must evaluate the positive and nega- tive attributes of places under consideration to insure se- lection of a location which best meets anticipated place utility. For this purpose a retirement location choice model is introduced (Figure 14). In this model the pgpgp- tial migrant makes an assessment of the positive and “252? tive factors which could influence the choice of a retire- ment location and, guided by this evaluation, ranks the places according to that person's preferences, or envisaged place utility. Tb this ranking of preferences are applied any constraints which might be applicable, e.g., political restrictions in communist nations, recent natural disasters, 102 oesoeo cofipmooq ucoaomeumm .va oaswam WUHOEO ZOHBupmmoz one O>aaamom mo . fizmammmmm< umoamno>oo no hvfidanmum seamen“ Immooo< mosmumm Hammadz V hasmcon macadao — coduaasnom BKdflOHI Adnazubom noeoanoae 103 prohibitive costs, or poor accessibility. Eliminating these places from the ranking, the location choice can then be made from the remaining places under consideration. Criti— cal to the successful application of this concept is know- ledge of the important factors to be considered in the as- sessment. For leisure retirement purposes, major factors appear to be pleasurable climate, favorable cost of living, social amenities, special retiree privileges, and recreation opportunities. Those factors which are most relevant in attracting retirees to Central America and could be applied in the model are subjects for the remainder of this chapter. Geographic Factors Geographic location and the physical characteristics of a place are the most important factors influencing the choice of a foreign leisure-type retirement site. Central America, less Panama, is located generally between 8° and 180 north latitude and oriented northwest to southeast (Fig- ure 1). The western Guatemalan border lies on a longitud- inal line with central Louisiana, while the eastern Costa Rican border falls on a meridian passing through the west coast of the Florida peninsula. The entire region encom- passes 179,270 square miles, equalling about one-twentieth the size of the United States, or one-fourth of Mexico's total area (Table 14). El Salvador is the smallest nation, “whereas Nicaragua, almost seven times its size, is largest. 104 Table 14. Area and Population Comparisons Area Pers.per Country (Sq.Miles) Population Sq.Mi. Belize 8,870 142,000 16 Costa Rica 19,700 2,023,000 103 El Salvador 8,260 4,128,000 500 Guatemala 42,040 6,016,000 143 Honduras 43,300 2,823,000 65 Nicaragua 57,100 2,224,000 39 Central America 179,270 17,356,000 97 United States 3,745,358 215,966,000 58 Mexico 764,000 62,159,000 81 Panama 29,208 1,719,000 59 Source: National Basic Intelligence Factbook (CIA, 1976) In population, Central America contains 17,356,000 persons, almost the same as that of the state of New York, or about one-twelfth of the overall United States population. With— in the region, Guatemala is the most populous and Belize the least populated. In all the nations except Honduras and Belize, the capital cities are the major center of pop- ulation, as well as the economic, commercial, political, and transportation hub. In Honduras the pOpulation and economic activities are distributed between the capital, Tegucigalpa, and the north coast city of San Pedro Sula; 'while in Belize the population and economic concentrations remain in Belize City, its former capital and major port. 105 Some interesting comparisons can be made with population densities and retirees. Belize and El Salvador, both of similar area, have densities of 16 and 500 persons per square mile respectively, and both also have attracted few retirees. Costa Rica, on the other hand, has attracted the most retirees and its population density of 103 persons per square mile is near the overall Central America density of 97. This figure almost doubles that for the United States (58). Although pOpulation density did not account significantly for military retiree locations within the United States (Table 10), it could be an important consid- eration in foreign retirement area choice. Most signifi— cant, however, are those factors pertaining to the physi- cal enVironment and accessibility of an area to the United States. Environmental Conditions The main environmental conditions which affect deci- sions on retirement sites pertain to climate and surface configuration. For year—round retirement, favorable cli- mate allows perennial freedom of action, especially for activities related to the sun and sea and other forms of outdoor recreation such as hunting, fishing, camping, golf- ing, etc. Varied terrain provides spectacular scenery, such as rugged mountains, beautiful lakes, beaches, barrier reefs, volcanoes, waterfalls, canyons, etc. Based on 106 altitudinal change, they also affect climate and create variations in vegetation and wildlife habitats. Surface Configuration The physical character of Central America, in highly generalized terminology, is represented by rugged mountains, escarpments, and hills covering the major interiors, in- terlaced with low, undulating intermontane basins (Figure 15). Surrounding these uplands are low, narrow plains and rolling hills on the Pacific shores, and on the Caribbean, lowlands ranging from narrow coastal plains to broad, ex- tensive ones. From the Mexico-Guatemala border to middle Costa Rica an almost continuous line of volcanoes, some active, runs along the Pacific rim, forming the most sig- nificant mountain range in Central America. This diversity within the confines of a relatively small land area cre- ates a large variety of accessible landscapes. Belize, which borders the Caribbean Sea, is mostly flat and swampy along the coastal areas. However, the ter- rain rises inland, mostly as rolling limestone hills, cul- minating in the Maya Mountains of the south with eleva— tions of over 3,000 feet. The main physical attractions are the off-shore islands and adjacent Great Barrier Reef, the largest in this hemisphere. Guatemala's landforms consist of tropical, flat-to- rolling limestone lowlands in the north; the sparsely populated Peten area; and mountainous highlands composed 107 86‘ 84' 18' i 2... ~ I‘ \‘J a ‘ —. nun“. w \d 9° fl pawn, ‘ m ——‘I”:1\"—‘—fq—a° W} ,. , ‘ _ ‘6' , ,. 0., l P. ‘ _ ‘. \ nose :1. ,. ’ z . v . ' 1 I, ' \n ’ _ a ‘J . ' ° .x‘mm ¥ @;_ ,‘ J 4 v . 5 n ~;. - L .. IL. . _ ‘ _ .bA'qj-‘I PSTW' uwnn :7 my... ""\9 ' ‘7' ‘ , :5 ‘1 6%“ «.443 wr‘ I; ‘ . ' - ' 'l *9 9'- I, . A . . | v‘” s3. '3": ,' “”5 I“; fit}? ‘3‘.- 1 ‘0‘ . M ‘ 2, K nm a A?! “If-i“ Q I I ‘ o ’ .33? I'm sand; 9" - ' . " , F ', Longhy 9““ 1 0"“ h, ' F J. ' " w ’ ' . (a. l. énflférms of a , my, M... «w- ~ f*— 3:: y , ’5" my" ““me I 7_. , a L, . V . - Ifwma. 1" ENTRAL ERICA M ,, no no . , «'6. 9‘ ”A.“ M ’ 1:111 / “I (mic Wk ” - wt} A I. 6v .1; 3% ‘ . . . C O :’ (wit-«fir j ' \ ”Aunddlnrh I l (I ’ 4’ I ‘L k t « 3 I; m, :5 “ ‘ A And 1 . .. l‘q AIJMIb ”'“r‘, 8 _, J's; w- , _ L , é‘n 93“ w hill-u . > v- «mm- new: m :w 9:09“; I. 5’ s i a 3' fl ‘9 3 er a 9 From Erwin Rah». IW ' 84’ Figure 15. Physical Map of Central America 108 of two ranges in the south. These ranges (9,000-11,000 feet) are separated by the Motagua River depression, which terminates on the irregular Caribbean coast. The Pacific coastal plain, about 200 miles long, and the lowlands along the 50 mile Caribbean coast complete the configuration. The most scenic region lies on the volcanic axis of the southern highlands, where numerous volcanic peaks (10,000— 14,000 feet), lakes, and intermontane basins, including Guatemala City (5,000 feet) are located. El Salvador, which faces the Pacific Ocean, is divided physiographically into three regions. A mountainous north- ern highland (9,000 feet) is located along the border with Honduras; a central region of valleys and plateaus contains the country's volcanic axis of two rows of 22 east-west or- iented volcanoes; and a narrow lowland runs along the Pacif- ic coast on the south. A land of volcanoes interspersed with lakes, the entire country offers many varied natural attractions. Honduras is predominantly a highland nation with two major mountain ranges, which rise to elevations of 9,000 feet, bisecting the country generally from northwest to southeast. Extensive valleys and plateaus, ranging between 1,000 to 3,000 feet, occupy this mountainous interior, pro- viding favorable areas for human settlement. There are ex- tensive low coastal areas along a 400 mile Caribbean Sea frontage and a 90 mile lowland around the Gulf of Fonseca 109 on the Pacific. The off—shore Bay Islands in the Carib— bean are attractive to sea sportsmen and the highland in- terior provides a variety of mountain scenery. Nicaragua has three distinct physical regions. 0n the Pacific side are lowland plains formed by the Nicaraguan Depression, which contains the two major lakes of Central America, Lakes Nicaragua and Managua, and separating this rift valley from the ocean, a single chain of volcanoes running along the entire coast. A central highlands area, which is most extensive in the north, runs from the Hon- duran border to that of Costa Rica and contains mountain ridges reaching 3,000 to 6,500 feet in elevation. Finally, a vast Caribbean coastal plain, partially swampy, covers the eastern third of the country. The most scenic natural phenomena and landscapes are located in the volcano-edged plains along the Pacific. Costa Rica's terrain features are also categorized as central highlands, Pacific lowlands containing some coastal uplands, and wide Caribbean lowlands. The major highland region is composed of a northern volcanic range (Volcanica) with volcanic peaks projecting to 11,500 feet (Irazu); a central upland basin of 3,000 to 5,000 feet elevation (Me— seta Central), and a southern granitic mountain range (Talamanca) reaching almost 13,000 feet. Along the Pacific coast ubiquitous alluvial lowlands and scattered coastal uplands are also found. But unlike its neighbors to the 110 north, Costa Rica has two prominent rocky peninsulas (Ni- coya and Osa) and two significant gulfs (Nicoya and Dulce), which are the result of geologic faulting. The Caribbean coastal plain is extensive in the north, narrowing to the south.10 This complexity of landforms provides a wealth of natural beauty that, because of a better developed infra— structure, is more accessible to the prospective retiree than that of the other Central American countries. Climatic Conditions Central America lies entirely within the latitudinal limits of the tropics and zone of the Northeast Trade Winds. These phenomena, along with differences in the surface con- figurations of the land, account principally for the cli- matic conditions of Central America. The Caribbean lowlands have tropical rainforest or monsoon climates, while the low- lands of the Pacific, being on the leeward side of the high- lands, have tropical savanna climates. In the central high- land sections, where increasing altitude has a cooling ef- fect, most climates are classified as temperate highland, and are desirable for leisure living. According to Ullman (1954), the ideal climates of the world are those that fit the categories of the low latitude-high tableland areas of 10Some of the information for the individual country landform descriptions was obtained from Background Notes, ?ublished by the Department of State, WaShington,’D.CT 'USDS). 111 Latin America; and trade—wind island classes, such as that found in Hawaii. With the exceptions of Nicaragua and Belize, Central American populations are concentrated in the highland ba- sins and plateaus, taking advantage of the favorable cli- mate. Military retirees have also been selective in this respect. Most of those who have settled in Central Amer- ica chose the Meseta Central, a highland plateau contain— ing the capital city of San Jose, Costa Rica (3,900 feet), where temperatures remain almost constant throughout the year. Average low temperatures in San Jose range from 570 in January to 620 during June, whereas average high temper- atures range from 760 in January to 810 in April. Overall monthly averages range from a January low of 670 to a May high of 71°, the annual monthly difference being only 4° (Table 15). Other popular highland cities have similar temperatures, with some change due to variations in lati- tude and altitude; e.g., Tegucigalpa, San Salvador, and Guatemala City. In contrast with these favorable highland climates, Managua, located in the Pacific lowlands, has monthly averages in the 805. For a familiar and chilling contraSt, Detroit, Michigan is included in Table 15, as well as Miami, Florida, where winter temperatures are con- sidered ideal and summer ones oppressive. 112 mahmno smossm accuses moucum posse: "monsom mm as ms Hm mm mm ow 5» vs as mm mm .oH “as“: mm ow mm so as ms mm mm we em mm mm .mwm afloapmn om mm mm mm mm mm mm mm mm mm mm Hm .omH nausea: Hm mm mm mm mm mm mm mm mm mm mm am .ooo.n safio «Haemuaso ms ms ms as as ms ms ms nu ms ms ms .omm.m goua>fiam can mm H» on ms ms ms ms ms es ms as me .oom.m mgaawsosmme be mm mm as oh mm as as as mm mm um .osm.m once saw 009 >mz poo mom ws< Huh can ama AQ< um: XMWh|,sNH pawnsohnch a“ moohmon ousuaua< woman mthHmHoQEmB hanusoz owmho>< .mH OHQNB 113 Natural Hazards The attractive topographic and climatic qualities of Central America can be reduced, or even negated, in those areas that are highly susceptible to major natural hazards. This, of course, includes the favored highland settlements and coastal population concentrations located along the Pacific, from Guatemala to middle Costa Rica. A volcanic axis and a series of geologic fault lines extend along this section of Central America, subjecting it to frequent vol— canic eruptions and strong earthquakes. Volcanic activity, which is more localized and less destructive than major earthquakes in terms of human casualties, nevertheless has historically caused serious damage throughout the area. Most notable in recent years (1963-1964) was the long last- ing eruption of Irazu in the Meseta Central of Costa Rica. Major earthquakes have plagued population centers in Cen- tral America throughout recorded history, including destruc- tion of the old capitals of Antigua, Guatemala and Cartago, Costa Rica; and the present capitals of San Salvador, El Salvador and Managua, Nicaragua (West and Augelli, 1966, p. 34). The devastating Guatemala earthquakes of February 1976 severely damaged widespread areas, including Guatemala City and Antigua, and was responsible for a massive number of dead and injured.11 11The Guatemala Newsletter (AAP, 1976) reported 23,000 deaths, ovéF‘50,000ifijured, and a million homeless peOple resulting from the quakes. USAID (1976) reported 22,808 dead, 76,552 injured, and 1,160,969 homeless. 114 The tropical climates of the coastal lowlands of the Caribbean, especially Belize and Honduras, which provide a great variety of recreational activities, are free from the threat of earthquakes. They are, however, susceptible to other environmental hazards, principally the destructive force of the tropical hurricane. Torrential rainfall and high winds flood interior regions and seriously damage coastal sections. Belize has been particularly plagued by hurricanes and its major port and former capital, Belize City, has experienced heavy damage on many occasions. Con- sequently, a new capital, Belmopan, was recently located some 50 miles inland. The most recent damaging tropical storm was Hurricane Fifi, which in 1974 created widespread destruction in the northern Honduras lowlands. These natural hazards are phenomena which cannot be predicted as to exact time or place of occurrence and in most instances happen so infrequently that they usually are discounted in the retiree's location decision process. Their potential occurrence does exist, however, and for some retirees the threat is very real, therefore deterring them from settling in a natural hazardous zone. Accessibility to the United States The likelihood that retirement migration to a particu- lar place could be effected is dependent upon the attri- butes of that place, its perceived utility by the potential 115 migrant, and its accessibility. Consequently, geographic space is important, in that it can indirectly limit or re- strict movements, depending upon various degrees of access of one site to another. The direct land route access of the Central American nations to the United States, as well as the short airline distances, are strong positive factors favoring migration. The importance of accessibility of specific Central American nations will be discussed in the migration decision section of Chapter IV. Effect of Distance Geographic distance, or intervening space between places, is a critical measure in some geographic analyses. It is especially important in measuring monetary and tem- poral costs related to internal spatial flows of goods and people. The gravity model, with its inherent "distance decay"12 prOperty, has been effectively applied in many economic-motivated transportation and migration studies; e.g., transporting of commodities and labor movements (Carrothers, 1956). In long distance international retire— ment movements, application of this model as a means of ex- planation is obviously inappropriate. However, as a factor to be used in evaluating retirement areas, the amount of intervening space between the United States and the potential 12Distance decay pertains to the friction of distance in the gravity model, i.e., as distance between places in- creases, there is a rapid decline in the intensity of move- ment. 116 retirement site may strongly influence the retiree‘s de- cision in choosing one place over another. Mexico’s pOpu- larity, for example, partly is due to its proximity to the United States. Also, Central America's recent appeal to retirees must be partially attributed to its relative close— ness to the United States, in comparison with South America or other more distant amenity-related areas. Intervening Opportunities Some migration theorists have placed importance on the number of opportunities in intervening space along with the effect of distance on mobility. Stouffer (1940) believed that the number of Opportunities at a particular place was proportional to the number of migrants at that place, and inversely proportional to the number of opportunities which intervened from origin to destination. His major difficulty in testing this concept was in applying measures Of "Oppor- tunities". In theory, this application appears valid for foreign retirement migration in that the chosen place for migration represents maximum opportunities and within the intervening distance numerous Opportunities could attract and retain the migrant. For example, retirees who have se- lected Costa Rica as their future retirement home because of its many attributes (Opportunities), and move overland to the destination, must pass through Mexico, Guatemala, El Salvador, Honduras, and Nicaragua (intervening opportun— ities). In any one Of these nations they could perceive 117 greater Opportunities than expected in Costa Rica and never reach the initial Objective. Of the retirees interviewed in Costa Rica, 42 per cent travelled by road and thereby were subjected to various intervening opportunities along the way. Overall, 43 per cent of the interviewees in Cen- tral America used motor vehicles as their mode of transpor— tation from the United States to the retirement destination. These retirees apparently resisted any intervening Oppor- tunities on their journeys; others of unknown number could have opted for retirement in nations other than that of their original choice. Stage Migration The process whereby migrants gain access to a final destination through a series Of moves is referred to as stage migration. Historically, this has been the accepted method in which migration takes place from rural areas, through various sized urban areas, to a final large urban center (Ravenstein, 1885). This migration process is evi- dent today in the developing nations of the world and a number of empirical investigations have substantiated it in Latin America, particularly Thomas' (1968) study on in- ternal migration to Guatemala City. Stage migration has not significantly occurred in re- tirement migration to Central America, according to re- sponses from Central American retirees. Only 14 per cent indicated that they had spent significant periods of time 118 in other places during their intended migration to a Cen- tral American nation. Most notable were the six retirees who settled in both Mexico and Guatemala prior to reaching their present retirement location in another country. Al- though few military retirees have been involved in a true form of stage migration, their retirement migration his— tories reflect a post-retirement pattern which could be referred to as a foreign migrant retiree strategy, i.e., initial movement to a home or job site, followed by a move to the United States sun belt, and thence to one or more amenable foreign areas. Internally, the presence Of a re- verse form of stage or step-wise migration from primate city to rural areas may exist in some nations. This phe— nomenon will be given further attention in Chapter V. Political Factors The political environment of nations can either posi- tively or negatively influence potential retirement mi- grants. Nations having favorable laws for retirees; stable, democratic governments; fair immigration and customs pro- cedures; and close, cordial political relationships with the United States attract retirees; whereas those exper- iencing political instability, governmental corruption, anti-Americanism, and those lacking special retiree bene— fits, tend to repel retirement migration. 119 Government Policy The Central American countries, following the examples of Spain and Mexico, have employed active governmental pro- grams to attract foreigners as part of their national devel- Opment scheme. Primary emphasis in this respect is directed toward the seasonal visitor, since tourism as an industry enjoys high priority for national development in Central America. Certain nations, particularly Costa Rica and Guatemala, have also recognized the advantages of implement— ing special programs for retirees as an adjunct to their tourism development ventures. In most of the Central Amer— ican nations tourist and retirement programs are both Oper- ated by the national tourism institutes and, therefore, their promotional endeavors can be accomplished concurrently. Special Retiree Provisions In order to encourage alien retirement in their re- spective countries as a partial means of strengthening their economic systems, four Of the six Central American nations have enacted laws which provide special benefits and privi- leges to qualified persons (Table 16). Costa Rica was the first country to implement special legislation for retired residents (July 1971), followed in August 1973 by Guate— mala, El Salvador in November 1973, and Nicaragua in NO- vember 1974. Although Honduras and Belize now have no legislation favoring retiree migrants, they will probably 120 .QOU pom coda cosmmo Emfiudoe acmeamom tha >02 me mafia cows Hacoasaz nonapmm was «smaaaOaz msoz massscom waspflpmcH smaasoe ucmoammm mums ms< @202 cows casaeopaso emuaumm msuwm madamsaso .mnms +m spa; soanoch pcmcammm mums >oz mcoz oovw so come we» so shamans: mucoamcmm msv noea>sam Hm ovzpfiwmsH EmHHSOB sosofimcmm HhmH H55 mcoz ooms caoam apmoo pcmoamom Name «cam aemoo osoz ONflHOm ow< sofimsom hosom< cause made new asucsoo .sfls .oz.cfis wsfimwooonm mosfipom .02 BOA mcoauaoamfisaso mausofiumz cwflosom pow msmq psoamhfiuom .mH wands 121 whack psom ow sofism Ousw so was msmoz mossy Ow sofisg wmcmco msumum sow mooom annomso: so momma OOpQEOxO zmmm muse» N munch o>flm Op sofisQ Opus :0 can mama» moan» Op poflsm owcmno mDQMgm sow mooow caonmmson so mossy Omansmxo hama OHanHd>m Hos mascofiwms OOHMfistu m“ muons cofipwosom pew Hasofimmomosa sflwupoo CH moms msofiuQonm* snow Ow uoflsm owcwno msumpm pow Opus new msoow OHonomson co momma sonaamxo zwm r awe» oz osoz .mow mammsmofiz OHQmOflHQm< “oz. magnusom mmow *oz osoz mo» «Hmaowmsc «mo? 02 Mama pom o . mo» soow>amm Hm Hmmw *Oz puma son m mo» «Oam mumoo manaoaflaa< poz ouasmm muamsom mapwum hum som.xsoz A.mosv wfiafiq occupfimom mo owswao ow OOBOHH< monomn< owacfimoosm muessoo .u.p=oo .oH manna 122 mewsmosa omen» wsHHocmz mOHosmwm HmsoHpms on» Scum posHmwno who; SOHQB .mBmH pcosoaHpos OOHHO may sH posHmucoo soHumEAOHcH Sony OOHHQEOU ”monsom mo moHQoo mHOOp can uOHHmsw owuso uanH so hoHHsuulmmso: a OpsHosH ado v pmosnw Sosa Oz osoosH HH< muwmm m you H szo OEHQ ocouooo.OHw sawdndon OHndOHHQQm uoz mapsosom saosna Sosa msumz m hon H con» 02 osoosH HH< .mpmom v pmswm H mHso oEHp osonooo.hw «Hmeoumsc oaosna 809% oz msoosH HH< munch m pom H cho OEHH oconooo.on soom>Hmm Hm owosnd Sony Oz mEoosH HH< vwsmom m pom H zHco oEHp o:ouooo.v& onm mamoo OHanHHam< poz ONHHom mHnmsmNHuHo COHHQEOxM usOQEH mooow UHosomsom so mafia mo owsmno xme oaoosH oas< mosh >p59 HHOQEH mosh .wa< .Nmz mapsaoo mZOHBwMMIhm mon mamoo ONHme mhowowmo Hau09 hhvssoo musdhmH: one no soHuHmoosoo OHsnmhmoaon .om OHQaB 145 Race is another demographic factor that is highly se— lective in Central America among military retiree migrants. All of the respondents were Caucasian (Table 20). The complete absence of non—Caucasian retirees in these nations may be due to the social status values placed on skin color in Latin America. Socio—Economic The sociO-economic characteristics of retirees account for the attitudes and Objectives of many, just as the demo— graphic factors differentiate them from a normal cross-sec- tion of the total American retiree population. Sociologi- cally, the family structure and educational background of the migrants are revealing factors with regard to their pO- tential stability in, and contributions to foreign com- munities. Of the total retiree population surveyed, 81 per cent were married, eight per cent single, nine per cent divorced, and two per cent widowed (Table 21). Only in Costa Rica and Guatemala were retirees found in all these categories, whereas in Honduras and Belize, they were all married. Within the married category, 98 per cent of the spouses were present in the household. In addition, 48 per cent, or about one-half, of these households included one or more children, or other dependent relatives. This substantiates hypothesis 2.b., i.e., military retiree migrants share the characteristic of being married. 146 m m o o o o m H O>HpmHom sonpo N N o o o o N o smhoHHno v m m o o o o m o sosoHHso m wH mH H N H N m o cOQOHHsU N HN ON N H H 0 0H H OHHSU H mm Hm m o NH e no H mmsoem oHosomaom sH mcomuom N N o o H o H o UOBOUHB m OH H o H o m o oooao>Hn Hm mm m 0 NH v mm H uoHanms m m o o v o m o mesHm museum HduHua: .usoo son. .02 hmmz, .som .ouu .mwmx .m.o .MHom HmHOH anussoo muomoaao moHpmHumwoahano oHsosoomIoHoom asmHuHS .HN oHnmh 147 H H o o o o H o mmsoem s m N o o o o o HeaemHs .O.nN s m H o o o m o omsoem sH NH H H m o NH o HeeemHs mhmumdfi NN 0N H H e H NH o mmsoem hN NN o H s N NH o HeeamHs madmhfideOOdm sN mN H H N H NH o mmsoem mN mN N H v o NN o HeaHmHs OMOHHOO oaom Nm wN H e e N eH H mmsoem ON mN H m m N NH H HeaHmH: tempo eeNH s m o o o o m o omnoem n o o o H o m o HeaemHs . tease new N N o o o o N o mmsoqm H H o o o o H o echemHs muse» w some: sonmosom veto awn .oz .omk}, .eom. .oeu .m.u .m.o nwom Huuoe maussoo huomouao A.o.acoov HN mHan 148 H H o o o o H o ooo.oN-NNN N N o o H o N o coo.NNuoNN N N H o N o N o coo.0NuNHN NN NN H H a H NH 0 ooo.mH-oHN Ne NN N N N H NN H ooo.oH-NN oN NN H N N N NH o ooomeIHmmep» oaoosH asoaoquom NN NN e N N N me o HmoHHHO N N H H o o N o Humane: He Ne H N oH N oN H eeHmHHem onsuoshpm oomho N N o o o o N o emeeo N NH O H N H N o mmcHaaz ON NN H H N N NN o moeom NHN NN NN N H N o NH o N>Nz NN Ne N N N H HN H NaNN OOH>Hom coaHomHsD v e H o H o N o .eam NeHNooN NH oN o N N 0 NH o .eam oaHeuHNaN NH HN v e N N N o eoNoHNaa NHHsm HN NN H o N N NN H eeNHHoN NHHsm usoshonsfi psoo Mont .o .0 .:0m .ouc . . .me .Hom HNHOB z Hz. thssOWIM NHOMOHNU A.U.psoov HN oHnmfi 149 The educational status of both the retirees and their spouses ranged from under eight years of schooling through completion of the doctorate degree (Table 21). Most sig- nificant in respect to education are the high percentages of migrants and their spouses who hold baccalaureate and higher degrees, i.e., 49 per cent of the retirees and 31 per cent of the spouses. Additionally, 25 per cent of the retirees and 27 per cent of the spouses had completed some college, with only six and nine per cent, respectively, not having graduated from secondary school. The education level attained by this group supports hypothesis 2.c., i.e., they are highly educated. The majority (61 per cent) of the retirees migrated to Central America to fully retire, whereas 18 and 17 per cent, respectively, were employed fully or part-time and four per cent were seeking employment (Table 21). This distribution reflects some association with the younger age structure of military retirees, the size of their families, and the need for employment as an income supplement or an ego-satisfying mechanism. Costa Rica, most popular of the Central American nations, led the fully retired category with 72 per cent, while all the Honduran retirees were fully or part-time em- ployed. In the remaining countries, except Belize, fewer than half of the migrants were fully retired. As pointed out in the section on economic opportunities in Chapter III, most Of these working retirees are self-employed, 150 principally as farm or ranch owners, with a few operating small businesses such as real estate sales, bars and res- taurants, gasoline stations, and recreation vehicle parks. Others are employed with United States government organi- zations and in educational institutions. The principal parent service of the retirees was the Army, with 37 per cent of the migrants, followed by the Air Force and Navy, representing 30 and 22 per cent, re- spectively (Table 21). The Marine Corps accounted for nine per cent, and the Coast Guard and Public Health Service had one per cent each. Marine Corps retirees comprise only five per cent of all military retirees. The inflated percentage for Marine retirees in Central America may be related to the assignment of Marines as guards at all Amer- ican embassies. That is to say, a higher percentage of Marines than other services has served in most of these na- tions. In numbers of retirees in Central America, however, the Marine Corps contribution is relatively small and the percentages for the other services are not out of line with overall military retiree strengths; consequently, no single military service can be considered dominant in Central America. A strong differentiating factor among military per- sonnel is that of rank or grade in the service, since it identifies not only financial capabilities, but one's place in the social structure. To a certain degree this attitude 151 carries over in the retired ranks. The Central American retirees were somewhat evenly distributed according tO rank, with 55 per cent commissioned officers, 41 per cent enlisted men, and four per cent warrant officers (Table 21). Specific retirement grades among those personnel re- tired for length of service (20 plus years) ranged between sergeant (E5) and lieutenant general (09), which in terms of current annual retired pay could be the difference be- tween $4,080 and $29,700. Military retirement income is critical for those re— tirees who have taken advantage of the special laws for retirees in Costa Rica, Guatemala, El Salvador, and Nica— ragua, since their status is dependent upon a minimum monthly guaranteed pension. On the basis of current armed forces pay scales, a retired enlisted man in pay grade E5 (sergeant) with 20 years active service would receive a pension of $340 monthly, thus qualifying him as a pensioned resident in Costa Rica ($300 mo.) or Guatemala ($250 mo.) (Table 16). If single, or with only one dependent, he could also qualify in El Salvador; however, with two or more de- pendents he would need $400 per month in El Salvador. In Nicaragua a minimum of $400 monthly qualifies a single per- son, plus $100 additional for each dependent. Some enlisted personnel, of course, cannot meet these minimum income re- quirements, especially if retired some years ago, when lower pay scales were in effect. Nonetheless, if recently retired 152 in pay grade E6 (staff sergeant) or higher, all retirees should be financially eligible for these retirement pro— grams. The annual military retirement income for the Cen- tral American retirees ranged from under $5,000 to over $25,000, the median, representing 47 per cent of the group, falling in the $5,000 to $10,000 category (Table 21). In the nations of Central America, where annual per capita in- comes vary from a low of $330 in Honduras to a high of $810 in Costa Rica, the retiree pensions represent substantial amounts. Moreover, the capital input into these small na- tions from external sources could be significant in strength— ening their economies. For example, using the upper limits Of the retirement income ranges, and multiplying these by the numbers of retirees in Costa Rica for each range, that nation benefits from a foreign capital income of about one million dollars annually. Those retirees who are employed or have other sources of income besides their retirement pensions contribute additional resources to the economies. Total income figures of the respondents in this study were incomplete, since some retirees did not wish to divulge their outside incomes. For years of active military service completed, 52 per cent of the migrants had served 20-24 years, 26 per cent served over 25 years, and 22 per cent less than 20 years. Retirees in the latter group are those retired for disability or through a reserve program in which less than 153 20 years of active duty was required for partial retirement pay. Personal Interests Military retirees have varied fields of interest. This is to be expected, considering the fact that during their active military careers they have travelled exten- sively, changed jobs frequently, and been exposed to many social and recreational opportunities. Some areas of in— terest expressed by the Central American respondents per- tain to all types of outdoor activities, such as golf, ten- nis, hunting, fishing, boating, camping, etc.; social events of both an organizational and personal nature, e.g., club and fraternal meetings and dining and night club entertain- ment; building and decorating a dream home; archeological and nature expeditions; or pleasurable jobs and business operations. As a more definitive measure for determining how the retirees are occupied, they were asked to indicate the amount of time spent each week in certain activities (Table 22). The most revealing and unexpected finding of this inquiry was that the retirees spend the highest per cent of their time on reading and writing. A young and ac- tive group of this nature might be expected to devote most of their energy to recreation. Second in the ordering of time spent was socializing, followed by recreation. As ex- pected, the majority of those retirees engaged in work or 154 Table 22. Migrant Expenditure of Time in Average Week Activity Degree of Time Spent (Per cent) 'MEETI ’Much —_Some Little FINOne Job or Business 26 2 7 l 64 Socializing 5 25 39 20 ll Housework 8 21 26 16 29 Recreation 7 11 43 20 19 Religious & Community 1 10 29 31 29 Reading & Writing 10 39 31 13 7 TV & Radio 1 8 25 43 23 Education 5 ll 16 15 53 business devote the highest degree of their time to that activity; however, 64 per cent spend none of their time on the job or in business, attesting to‘a full retirement life for most. These retirement behavioral patterns probably conform with the typical life style of most retirees world- wide. This relationship with the total retirement popula- tion could be of interest for further research. Selectivity Selectivity in the migration process pertains to the singling out of demographic and sociO-economic characteris- tics of migrants which distinguish one group from another. This section of the study identifies some selective charac- teristics of the Central American military retirees as com- pared with the general United States military retiree popu- lation. 155 Age, Sex, and Race The predominant age group of the military migrants is 50-59 with a median age of 55, whereas that for the general military retiree population is 45-54 with an average age of 42 (Army Times, 1973, pp. 11 and 22). The Older age struc- ture of the migrants suggests selectivity for full retire— ment in Central America. Moreover, the median age of 55 substantiates hypothesis 2.a., which states that military retiree migrants share the characteristic of a median age in the mid fifties. All of the Central American migrants are males and of the Caucasian race, thus reflecting selec- tivity for these persons. Language The ability to speak Spanish is apparently an important selective criterion for Central American settlement, since 88 per cent of the military migrants have this capability. Although figures regarding the Spanish language capability of the general military retiree population are not avail- able, they are probably below 10 per cent. Retirement Income The median annual retirement income of the migrants is $8,900, a significant guaranteed income. The average yearly retirement income for all United States military retirees in 1976 was $6,648. The yearly average for Officers was 156 $10,092 and for enlisted persons $5,184 (TROA, 1977, p. 11). Although the Central American retiree figure is a median value and the overall retiree one an average, they can be used for relative comparison. Consequently, it is evident that a larger number of retirees with higher income opt for Central American retirement, verifying hypothesis 2.d., which states that military retiree migrants share the char— acteristic of having a substantial retirement income. Military Service The selectivity of migrants with regard to military service pertains to the dominance Of one parent branch of the armed forces over others and the differentiation of re- tirees by military rank. Army and Marine Corps migrants are present in larger percentages (37 and 9) than corresponding percentages of the total military retiree population (34 and 5). On the other hand, Navy and Air Force migrant percent- ages (22 and 30) are smaller than the overall retiree popu— lations for those services (26 and 35). Army and Marine Corps migrants, therefore, are somewhat selective. Military rank is selective in Central America, with 55 per cent offi- cers and 41 per cent enlisted persons, as compared with 27.5 per cent officers and 68.5 per cent enlisted persons in the entire military retiree population. No selectivity is pres- ent for warrant officers, both categories having four per cent (TROA, 1977, p. 11). 157 Propensity to Migrate Once a person's bond to a familiar environment and population is broken by spatial movement to new surround- ings, succeeding moves are likely to occur with less an- xiety and difficulty than the initial one. Consequently, persons with extensive migration histories are prone to continued movement. Military personnel epitomize this phenomenon, becoming highly specialized migrants from the time of initial entry into the service and continuing through their active careers. For many military personnel, the migration patterns and cycles established during mili— tary service carry over into their retirement years, due to their continued search for something new and interest- ing. Having experienced foreign assignments, therefore, military retirees are probably more apt to try retirement life abroad than are their civilian counterparts. Movement During Military Service To illustrate the movement patterns of career military personnel, the Central American retiree population was asked to recall the numbers of their principal military moves and places of assignment. In the continental United States the moves ranged from one to 20, with a median of six moves per person. Principal states in which these per- sonnel served, in order of ranking, were California, Flor- ida, Texas, and Colorado. Their expressed preferences for 158 military assignment were predominantly California, followed by Florida and Texas. Foreign moves ranged from zero for two retirees to 14 for one retiree, with a median Of five moves. Principal overseas assignments by ranking were Europe, Japan, Korea, Southeast Asia, and the western Pa- cific area. Predominant in Latin America were Panama and Puerto Rico. With a median of six continental United States moves and five foreign ones during the military careers of these retirees, it is evident that they have been conditioned to frequent moves and exposure to foreign areas, thus more likely to have a high propensity for foreign migration after retirement. Consequently, this finding supports hypothesis 1. Previous Retirement Migrations The Central American retiree population has, of course, exercised its propensity to migrate to a foreign area at least once since termination of active service. After this first retirement move the propensity to migrate continues to prevail in the new retirement environment. For example, the survey found that 61 per cent Of the migrants had pre- viously lived in a country different from their present foreign residence since retirement. The number Of previous retirement migrations by this group ranged from one to six, including moves within the United States; the majority of 159 that number (59 per cent) moved once, 19 per cent twice, 12 per cent three times, eight per cent four times, and one per cent each for fifth and sixth time movers. The most popular retirement residence prior to residency in Central America was the United States, where California and Florida dominated all other states. Outside the United States, the top rank order Of previous retirement residency was Mexico, Panama, and Puerto Rico (Table 23). This pattern of repet- itive migrations to favorable climatic areas, including many foreign nations, lends further support to the proposi- tion that military retirees' propensity for international migration is strong. Table 23. Previous Foreign Retirement Residences Nation No. Nation No. Nation NO. Mexico 10 Nicaragua 1 Philippines 1 Panama 6 Honduras 1 Thailand 1 Puerto Rico 5 Colombia 1 Okinawa 1 Vietnam 4 Bahamas 1 Germany 1 Ebuador 3 Spain 1 Ireland 1 Guatemala 2 Australia 1 Somalia 1 Brazil 2 New Zealand 1 The areas of concentration of these previous retire- ment residences reflect a pattern Of amenity-Oriented re- gions, both foreign and domestic. Specifically, Mexico, Panama, and Puerto Rico accounted for 48 per cent of the 160 previous foreign retirement sites; whereas in the United States, California, Florida, and Texas accounted for 59 per cent of the domestic ones. All Of these places are asso- ciated with warm climates and other retirement-related amen- ities. Consequently, these statistics support hypothesis 2.e., that is, military retiree migrants share the charac- teristic Of a similarity of previous residence. This sub- ject will be further addressed in Chapter V under direc- tional bias of the migration streams. Migration Decision In the retirement migration schema (Figure 4), the most important element activating the migration is the migration decision. Obviously, unless the potential mi- grant has made the decision to move, the situation depicted in the model remains static. In the decision making pro- cess, the potential migrant evaluates the positive and neg- ative factors at both the origin and destination in terms of his perceived Objectives, and if the destination is favorable, he decides to migrate. For the military retiree looking for a desirable re- tirement location, it could be expected that the principal factors influencing the migration decision would be those associated with pleasant climate, low living costs, and other amenities which contribute toward a leisure form of life. To determine the significance of these attractive 161 factors at the destination, as well as those which might repel migrants from the origin, the Central American retir- ees were asked to give their reasons for the migration and to evaluate certain factors according to their importance to the migration decision. Positive Reasons for the Migration Individual responses from the migrants reflected a wide range of reasons for migrating to Central America; collectively, however, the majority of answers were related to favorable climate and cost of living factors. Descrip- tive replies included: Liked the mild climate Attractive cost Of living Healthy climate Good opportunity for outdoor sports and recreation Owned property and desired to build home Wanted a total change of atmosphere Liked Latin American people Desire to live in an area with friendly people Simply want to retire Slower pace of life Favorable government Educational Opportunities Native homeland Relatives' home and property ownership Economic Opportunities and challenges Employment In order to provide a more definitive and measurable unit pertaining to the reasons for migrating, data were tabulated for the more influencial factors, according to degrees Of importance, for Costa Rica and collectively for the other Central American nations. 162 Costa Rica Since Costa Rica is believed to exemplify the nations of Central America for amenity seeking migrants, it has been analyzed separately for comparative purposes with the res maining nations (Table 24). Considering both most and great degrees of importance, it is evident that the majority of Costa Rican migrants attribute their migration decision pri— marily to agreeable climate, secondly to the low cost of living, and thirdly to the special privileges granted to re- tirees. Moreover, the fourth and fifth ranking categories in the first two degrees of importance are favorable health and political conditions. The combination of these attri- butes account for Costa Rica's attractiveness among the Cen- tral American nations as a complete retirement environment. Furthermore, it supports hypothesis 3.a., which states that in the migration decision, site selection is directly re- lated to the amenities Offered by the host nations. Other Central American NationS' Since the other Central American nations have attracted fewer military migrants than Costa Rica and also have higher percentages of employed retirees, it was deemed necessary to collectively analyze these nations to determine the dif- ferentiating reasons for the migration (Table 25). Again, as was done with the Costa Rica data, most and great degrees of importance were given main consideration. In this case, however, the ranking differs between the categories of most 163 Table 24. Reasons for Migration to Costa Rica Reason Degree of Importance(Percent) Most Great Some Little Climate 37 19 5 0 Low cost of living 16 20 9 7 Special retiree privileges 13 14 10 8 Social & cultural activities 5 14 22 Recreational activities 7 13 23 Near relatives & friends 3 5 6 Health 12 6 6 7 Job 7 l 0 0 Property ownership 0 5 5 8 Accessibility to U.S. 0 5 16 12 Political conditions 8 15 16 6 Table 25. Reasons for Migration to Other C.A. Nations Reason Degree of Importance(Percent) MoSt *Great Some Little Climate 23 20 8 0 Low cost of living 12 17 12 0 Special retiree privileges 0 14 10 5 Social & cultural activities 3 9 20 23 Recreational activities 0 8 10 31 Near relatives & friends 20 8 8 2 Health 3 4 5 18 Job opportunities 28 3 8 5 Property ownership 5 8 5 0 Accessibility to U.S. 3 5 12 8 Political conditions 3 4 2 8 164 and great degrees. Job Opportunities, climate, and proxim- ity to friends and relatives are the rank order for the most important category; whereas combining the categories of 2222 and great importance, climate is again first, fol- lowed by job opportunities, and cost of living. Although the climate and cost of living factors do not appear as strong in these nations as for Costa Rica, nevertheless, they rank high in the migration decision process and rank first and second in Central America as a whole. Conse- quently, these data substantiate hypotheses 3.b. and 3.0., i.e., in the migration decision climatic conditions are of prime importance, and cost of living is of secondary im- portance. By individual countries, the nation which most closely corresponds with Costa Rica in terms of attractiveness for full retirement is Guatemala, where climate, low cost of living, and special retiree privileges have the same rank- ing. Retirees in the remaining nations, especially Hon- duras, have located there primarily for employment or to be near relatives. Negative Factors Influencing the Migration The Central American retirees who had a previous re— tirement location were asked to specify why they had de- parted that place. Those retirees, of course, who migrated directly to Central America upon retirement from active 165 military service presumably considered only the attractive factors at the destination, not having experienced repel- ling factors at a previous retirement site. The analysis, therefore, considers only that percentage of the retiree population with a former retirement migration history (Ta— ble 26). The primary reasons for leaving the place of ori- gin were job dissatisfaction (25 per cent), unfavorable political conditions (22 per cent), high cost of living (17 per cent), and disagreeable climate (14 per cent). The latter two factors could be alleviated by a move to any of the Central American nations. Job dissatisfaction could be remedied by selecting a total retirement environment and the political situation corrected by choosing a country with a stable, democratic government. It was interesting Table 26. Reasons for Leaving Previous Retirement Location Reason for Departure Percent Job dissatisfaction 25 Political conditions 22 High cost of living 17 Disagreeable climate l4 Inadequate educational facilities Social Ills Health Poor relations with relatives & friends Huhuhtbtb Lack of community services 166 to note that none of the respondents mentioned the lack of social, cultural, or recreational activities; poor tax structures; or high risk natural hazard areas as negative factors considered in the migration decision. Role of the Amenities Economic factors, normally the most important motiva- ting forces in migration theory, have less application as determinants in migration for retirement purposes. The foregoing findings strongly support the dominant role that climate and other amenities play in retirement migrations, thus giving weight to the conjecture that this type of mi- gration is amenity-induced. Nonetheless, even in retire- ment, material improvement is desirable, and can be realized on fixed pension incomes in lower cost of living environ- ments. That is why many military retirees have migrated from the agreeable climatic areas of the southern United States to Central America; i.e., to live a more gracious life on the same income. In some areas, the gap between living costs in Central America and the United States is narrowing, thus reducing the economic advantages. Conse- quently, although amenities remain the dominant motivating factors for migration into Central America, changing eco- nomic conditions in some nations could create spatial change. 167 Importance of Place Perception As stated earlier, in deciding whether or not to mi- grate, migrants make a comparison of the perceived advan- tages and disadvantages for retirement living at the origins and destinations, and if the perceived differences are pos- itive at the destination, the migration is likely to occur. In this comparative analysis, knowledge of the potential re- tirement places is critical for making a sound decision and to better insure consequent success Of the migration. Perception of a place can be formulated through de- scriptive readings, from pictures and films, information provided by friends and relatives, and by various other sources. Fifty-nine per cent of the Central American re- tirees had relatives or friends already living in the host nations, who could have provided information to reinforce their perceptions of the potential migration sites. How- ever, if confidence in the accuracy of perception gained by these means is lacking, the alternative is a personal visit to the area. In the case of the Central American retirees, a large number had positive perceptions of the places to which they migrated, since 79 per cent had visited the selected retirement area before deciding to move. More important than the mere overview that a cursory visit to an area would offer is previous long-term resi- dency which has provided in-depth knowledge of both the country and people. In this respect, it was found that 168 seven per cent of the retirees had completed their last ac- tive military assignments in Central America, while six per cent had finished their active careers in Panama and two per cent in Puerto Rico. This represents 15 per cent of the respondents, a significant figure considering the lim- ited number of active military personnel assigned to these nations. Many other retirees had completed duty assignments in Central America and other parts of Latin America at some time during their active military service (Table 27). Other than Panama (Canal Zone), Puerto Rico, and Cuba (Guantana- mo), where conventional military forces are stationed, as- signments in the remaining Latin American nations are lim- ited to embassy staffs or military assistance groups. Con- sequently, Panama heads the listing in Table 27; however, the third place ranking of Guatemala was most surprising. These findings illustrate the importance of familiarity with an area prior to migration, particularly in strength- ening individual information about the place. Moreover, they substantiate hypothesis 3.d., which states that pre- vious exposure to, or residency in, Latin America is an im- portant consideration in the migration decision. It is interesting to note that 51 per cent of the re- tirees had considered other foreign retirement areas during the decision making process. The nations most often men- tioned were, by ranking: Mexico, Spain, Costa Rica, Panama, Guatemala, Puerto Rico, Honduras, Portugal, and Australia. 169 Table 27. Previous Latin America Military Assignments r - -‘ _ ~47 - Country NO. Country NO. Country No. Panama 29 Costa Rica 2 Haiti 1 Puerto Rico 12 El Salvador 2 Guiana 1 Guatemala 9 Honduras 2 Surinam 1 Ecuador 5 Dominican Rep. 2 Bolivia 1 Cuba 5 Venezuela 2 Chile 1 Nicaragua 3 Mexico 1 Brazil 1 Colombia 3 Trinidad l Degree of Permanency It has been shown that military retirees have a his- tory of frequent foreign moves, during both active service and retirement, having become conditioned to continued movement as a way of life. After the migration has been effected, then, it is of interest to investigate its per- manency and to consider some factors which may be influen- cial in sustaining the foreign residence. Analysis of Area Satisfaction To discover how the retiree migrants felt about retire— ment life in Central America, they were queried regarding acceptance of the host nation as a permanent home; their satisfaction with the retirement location; and if dissatis- fied, their reasoning and movement plans. Sixty-four per cent of the respondents considered the host country their permanent home, with six per cent undecided. Seventy-five 170 per cent indicated that they were satisfied with their pres- ent retirement location. About half of the 25 per cent who were not satisfied stated a desire to return to the United States, while others of this group wished to move to either Chile, Argentina, Mexico, Costa Rica, Guatemala, Spain, Italy, or Southern EurOpe. Although not representing large absolute numbers, the fact that almost half of those desir- ing to move wished to return to the United States in itself is significant. This finding supports hypothesis 2.f., which states that military migrants share the characteris- tic Of a propensity to return to the United States. Rising cost of living is the reason most given by those retirees who are dissatisfied and would like to leave (Table 28). Other significant responses in order of impor- tance were lack Of employment, lack of community and com- mercial services, and social ills. Even among those retir- ees who had expressed satisfaction with their retirement. location, much recent concern was expressed over increasing costs throughout Central America and, in the major urban areas, over rising crime rates, especially theft and burg- lary. Due to inflation and the consequent lower purchasing power Of American dollars in Central America, some retirees with small incomes, especially those with families, now feel the need of a job to supplement their retirement pay. If unable to work under existing retirement laws, the alterna- tives are migration to a more amenable nation (lower cost 171 Table 28. Reasons for Desiring to Leave the Host Nations Degree Of Importance (Percent) Reason MOSt Gfeat Disagreeable climate 0 5 Rising cost of living 32 25 Lack of employment 28 11 Lack of community & commercial services 8 14 Lack of social & cultural activities 4 8 Lack of recreational activities 4 8 Inadequate educational facilities 4 0 Poor relations with foreign neighbors 0 8 Poor relations with relatives and friends 4 0 Social ills (crime, drugs, etc.) 8 14 Political conditions (unstable government) 8 5 High risk natural hazard area (earthquakes, etc.) 0 0 of living), or to one that does not restrict work by alien retirees, or preferably both. The retirees living outside the main metropolitan areas were those mostly concerned about inconveniences resulting from a lack of community and commercial services. Although a small number considered "poor relations with foreign neighbors" an important cons tributor to their discontent, it was apparent in some com- munities, especially where more Americans were concentrated, that the local populace resented the special privileges 172 given to pensioned foreign residents, thus creating some animosity. Most interesting was the lack of concern about living in an area that is prone to natural hazards.18 Despite the fact that 25 per cent of the migrants were dissatisfied with their retirement location, only 11 per cent of the retiree population stated that they had defin- ite plans to move. However, another 31 per cent of that population did not answer this question, thereby inferring the possibility of migration. Commitments under the special retirement laws, of course, retain some disaffected retir- ees. The majority Of the retirees, however, indicated sat- isfaction with the area as their retirement home. Cyclic Nature of the Migration Retirement migration to Central America is a recent phenomenon. Consequently, the permanence or cyclic aspects of the act at this early stage cannot be precisely deter- mined. On the basis of information provided by the respon- dents on years of residency in the host nations, however, some measures of the flexibility, or temporal length of the migration, can be derived. In Central America, the mi- grants' length Of residency ranged from less than one year 18The personal interview survey was conducted in Guatemala prior to the major earthquake of February 4, 1976. It would have been interesting to interview the Guatemala respondents again subsequent to the earthquake, to determine if their Opinions had changed. 173 for 10 per cent of the retirees to 30 years for a disability retiree with a history of business ventures in the area. The median residency period was 2.5 years and the highest percentage of retirees in any single year (28 per cent) was represented by those in residence between one and two years. After four years of residency, the numbers of retirees fell dramatically, 78 per cent having been in-country less than five years. This short-term residence history for most of the migrants can be partly accounted for by the recent pOp- ularity of the area for retirement. Moreover, it appears that the penalty provisions of the special retirement laws, under which many are living, retains some retirees through the mandatory periods (3-5 years) of those laws. Based on the proven migratory nature Of the military retiree popula- tion, however, it is more likely that the turn-over is due simply to the inherent desire to move on to new and differ- ent places. Evidence of Seasonality Since the majority of the retirees have migrated to Central America for full retirement (61 per cent), they ob- viously have leisure time for travel and visitation. Con- sequently, those of pensioned resident status can exercise their rights to remain outside the host nation for periods totalling six months or less on a voluntary basis. Retirees residing in Central America in a tourist status, however, 174 must leave the country every six months. These situations are, of course, conducive to migrants spending significant periods Of time outside the.host nations, which many do, planning their absences for the most part during the more disagreeable climatic seasons. When temporarily leaving the area most retirees (65 per cent) return to the United States to visit relatives and friends, or simply for travel and vacation. These sea- sonal visits regularly expose the retirees to the United States, thus providing them the Opportunity to periodically re-evaluate the advantages Of retirement in both places. Moreover, these trips to the United States lend credence to the hypothesis that military retirees share the charac- teristic of propensity to return to the United States. The remaining 35 per cent of the retirees who seasonally depart the host nation indicated that they vacationed, or visited friends and relatives, in other foreign areas. A most pop- ular and frequent trip in this category for many retirees, especially those residing in Costa Rica, is a visit to the Panama Canal Zone, where military facilities are available for shopping and medical services. Another indicator of seasonality of the migration was reflected by the presence of travel trailers as residences for six per cent of the retirees. Although the occupants of these trailers considered them temporary residences, while decisions were being made on the type and location 175 of their permanent homes, the trailers' very existence in trailer parks suggests possible transience. Effects of Kinship and Friendship Ties Satisfaction with a foreign area is more likely to oc- cur if relatives and friends live there, through whom ads justment to the new environment can be facilitated. This adjustment can be effected through the process of accommo- dation or assimilation in the host nations. Friends and relatives residing in the area and of the same nationality as the migrant are apt to facilitate accommodation, whereas those relatives and friends native to the host country could be the means for the migrants' assimilation into the new culture system. In Central America, 59 per cent of the retiree migrants had relatives and/or friends in the area prior to migration. Assistance rendered to them by these sources during their initial introduction to the area undoubtedly alleviated an- xiety and softened the "culture shock". Moreover, continued association with and assistance by these relatives and friends during the adjustment period contributed toward satisfaction and their likelihood of remaining in the area. Strong bonds Of friendship in both Costa Rica and Guatemala were evident among members of the American Legion posts, and in Costa Rica through membership in the Retired Officers Association. Forty-three per cent of the retirees held 176 membership in the Retired Officers Association, whereas 17 per cent belonged to the American Legion. Both of these organizations provide support and assistance of a nature that contributes to migrant accommodation. Further discus- sion of the importance of these institutions to the retired Officer living overseas will be found in the next chapter. Those migrants, on the other hand, who have close ties with natives of the area, or especially those whose wives are natives of the host nation, stand a high chance of becom— ing assimilated into the local society. For the Central American migrants this assimilation possibility is strong, since 34 per cent Of them are married to natives of their host countries. This chapter has described the character of the migrant and examined his propensity to migrate, the migration de; cision, and the permanency of the migration. Chapter V will examine the migration channels and the spatial conse- quences of the migration. CHAPTER V SPATIAL SIGNIFICANCE OF THE MIGRATION This chapter is chiefly concerned with the migration channels and information flows, and the consequent forma- tion of networks and settlement patterns in the host nations. Migration Channels and Information Flows Once the decision to migrate has been made, movement commences between a place Of origin and a destination, fol- lowing specific paths and establishing patterns. Con— comitant with the migrant movement, a transfer Of informa- tion, either directly by the migrant, or indirectly through the transmittal of correspondence, is likely to occur. Consequently, migration entails the exchange of both peOple and information. How this occurs through space is of cen- tral importance to the pOpulation geographer. Migrant Circulation System To understand the process of international retirement migration from the standpoint of how migrants settle in- country, a Migrant Circulation System model is introduced 177 178 for testing in real world situations (Figure 16). In accordance with the model, most military retirees who are considering retirement in a foreign country leave the United States (origin) for the host country (destination) on an exploratory trip. Once in the host country they investi- gate most cultural and physical aspects of life, returning to the United States to evaluate their findings and make a decision on whether or not to make the permanent move. Those retirees not satisfied after their return from the exploration trip, and who decide not to migrate, may act as carriers of negative information regarding the host country. However, the individual who, after evaluating all available information, decides to make the permanent move, quite likely will become a carrier of positive information. The satisfied retirees make the physical move to the host nation, in many instances passing through intervening areas, to enter temporary living quarters (gateway resi- dence) in the host nation. These temporary living quarters provide the migrants sufficient time to thoroughly assess the area before selecting a permanent residence. At this stage in the migration process dissatisfaction can again prevail, and if sufficiently strong, the migrant may return to the United States with minimum loss of time and money. Their negative reactions, of course, quite likely are car- ried back to the United States (origin), or to an interven— ing area. Those migrants who decide to take up I DESTINATION Positive Reaction Dissatisfied Exvloratory Person Trip Disaffected Gateway Migrant Residence t Termination Negative Reaction INTERVENING AREAS I If _L ORIGIN | l l l I | l I l I I l | e I L“! Disaffected 1* Permanent I ‘ Migrant Residence _ l l I l l I i | | I Figure 16. Migrant Circulation System 180 permanent residence in the host country usually will either remain until death (termination) or become disenchanted with the host country and return to the United States. Case Study To determine the authenticity of this concept in Cen- tral America, Costa Rica is used as a test case because of its larger retiree base and accepted status as a popular retirement area. Based on discussions with the respondents in Costa Rica, it was found that most of them followed the migration procedure shown in the model (Figure 16). For example, 78 per cent of the migrants had visited the area before deciding to permanently move. During this explora- tory trip, most retirees came to the host nation directly by air. However, some travelled by auto via the Inter- American Highway, passing through Mexico, Guatemala, El Sal- vador, Honduras, and Nicaragua; while others utilized "space-available" military air transport from United States bases to the Panama Canal Zone, and then on to Costa Rica via a "space-available" military-consular flight. In the latter two instances these inspection trips proceeded through intervening areas where the retirees were exposed to other Middle American countries. Upon arrival in the host nation, the retirees usually had contacts with friends, relatives, or military asso- ciates, who gave personal assistance and offered advice during their exploratory visit to the area. During this 181 time they surveyed the host country to determine if it con- formed to their expectations. Those who liked what they saw returned to the United States, or other places of ori- gin, and made preparations for the permanent migration. Those who were not satisfied returned home, some as carriers of negative information. This movement in advance of the actual retirement move is probably the most important single act in the overall miggation process. It insures a greater chance of the migration succeeding, with the dissidents filtered out of the system at this early stage. Of those retirees who performed this exploratory trip to Costa Rica, only one of ten later made the actual move.18 Those retirees who made the decision to migrate, upon entry into Costa Rica, located initially in San Jose in temporary quarters (gateway residence) while looking for, or building, permanent homes. At the same time they pro— cessed their applications for resident pensioner status. These gateway residences, which are usually furnished apart- ments or apartment-hotels, have become well known and much in demand by retirees. Of course, some opt to live tempor- arily in hotels, while still others who have traveled with house trailers reside in trailer parks. Discussions with 18This ratio was derived from estimates Of the numbers. of military retirees who had visited the area during 1973— 1975 and the numbers that actually became resident pension- ers. The estimates were based on information supplied by members of the Costa Rica Retiree Club, who had assisted these people. 182 retirees, their families, and other information sources revealed that at this stage in the migration process, some retirees become disenchanted and return to the United States, substantiating information obtained from the Costa Rican Tourism Institute. Of the military retirees identi- fied as resident pensioners, 37 per cent were not located at addresses listed and their whereabouts in most cases were not known by neighbors or other retirees (Tables 2 and 17). It can be assumed, therefore, that the majority of these individuals left the country as disaffected migrants without notifying Costa Rican Tourism Institute officials, as required by law.19 The majority of the newly-arrived migrants residing in a gateway residence completes all necessary government paperwork, obtains resident pensioner status, and moves into a permanent residence. Sixty-five per cent of the respondents indicated that they considered Costa Rica their 19In accordance with provisions of the resident pen- sioner law, premature departure requires exportation of all goods and articles imported tax-free, or payment of the appropriate taxes if the goods are sold or remain in- country. Early departure, in some cases, occurred as a result of retiree disenchantment with the area because of bureaucratic harassment, misunderstandings concerning the limitations set by customs laws on importation of personal property, difficulties encountered in processing retiree applications, and other inconveniences. It is believed that some retirees failed to advise the Costa Rican Tourism Institute of their changes of address or departure from the country before expiration of their contractual agreements in order to avoid payment of taxes and customs duties on goods brought into the country tax-free, or other fees and gratuities that facilitate exit. 183 permanent home and, based on the number of years residency, this appears to be true. One retiree has lived there 30 years and several others more than 10 years. The majority, however, 70 per cent, have been in-country for less than four years, implying that the "disenchantmentgsyndrome" may be a continual obstacle facigg military retirees residinggin Costa Rica. This disenchantment after several years resi- dence is viewed as a real problem by the retiree population, since it could establish a reverse trend in the migration flow to Costa Rica, thereby reducing social interaction among the group. The final stage in the migration circula- tion system is termination by death in Costa Rica or disaf- fection with life overseas and return to the United States. There are, Of course, individual deviations from the norm where the retiree purchased land and constructed hous- ing without first visiting the host country. Others moved directly into households of relatives; while still others were provided housing by business corporations, private institutions, or a United States government agency. In the other Central American nations some retirees conform to this model. Most moved there directly, either for employment, because their wives were native to the host country, or they completed terminal military duty assign- ments in these nations. Although of future potential impor- tance as retirement areas, these countries presently do not attract significant numbers because only recently have they implemented legislation to attract the military retiree. 184 Directional Bias and Chain Migration The retirees that pass through the Migrant Circulation System are not generated randomly at the origin (United States), but come mainly from selected areas in the southern United States retirement belt, California, Florida, and Texas (Table 29). This group includes 58 per cent of the migrants, thus establishing a directional bias in the flow. California is the major source area, supplying 27 per cent of the migrants, and is closely followed by Florida, with 22 per cent. The Virginia-District of Columbia complex, accounting for 10 per cent, is also important and, when combined with the northeastern states of New York, New Jer- sey, Massachusetts, New Hampshire, and Maryland, can be Table 29. Migrant Origins by States :======= Per cent Per cent State Of Total State of Total Miggants Migrants California 27 Alaska 1 Florida 22 Arizona 1 Texas 9 Georgia 1 D.C. 5 Illinois 1 New York 5 Indiana 1 Virginia 5 Kansas 1 Colorado 4 Maryland 1 New Jersey 4 .Michigan 1 Hawaii 3 Missouri 1 Massachusetts 3 North Carolina 1 New Hampshire 3 185 considered significant as a generating region. Twenty-six per cent of the migrants are from that area, the only major source area other than the sun-belt. Only a few retirees originate in the midwestern and mountain states, while no retirees were found who came from the Northwest. These principal flow patterns lend support to the concept that military retirees usually work for a period directly after retirement, retire to an amenable area in the southern United States, then migrate to a foreign area. True chain migration20 was not evident in Central Amer- ican retirement migration. Although directional bias was strong for the states of California and Florida, there were no particular sites identified in those states from which several retirees had migrated, based on kinship or close friend ties. However, in view of the cohesive nature of the entire military retirement community, it is conceivable that this process could be described as chain migration. For example, many of the United States-based military re- tirees were informed of retirement life in Costa Rica by the migrants living there, and those United States retirees who chose to migrate were assisted by the in-country retir- ees simply because they were military. These new migrants were subsequently socialized into the new environment and 20According to MacDonald (1964) chain migration is de- fined as "that movement in which prospective migrants learn of Opportunities, are provided with transportation, and have initial accommodation and employment arranged by means of primary social relationships with previous migrants." 186 served in the same capacity as information providers and assistants to the group, thus establishing a broad chain migration effect. Information Exchange System Equally important as the method by which migrants move move to Central America is how they gain information about the area. The interjection of information flows into the migration schema is Of utmost importance in international migration. Prospective migrants are dependent upon infor- mation to differentiate potential Central American retire- ment areas, unless, of course, they have previously re- sided in or visited the area. Information can be trans- ferred by means of mass media or through interpersonal com- munication. Both of these processes are effective means of communication; however, personal contact between migrants and potential migrants has a greater impact than other forms of communication. Information feedback therefore is criti- cal, since it can either encourage or discourage the poten- tial migrant (Figure 4). The means whereby information is exchanged between the destination and the place of origin is introduced as an in- formation exchange system (Figure 17). Those exchanges of information through agencies and mgil_channels represent media sources, whereas trips and personal contacts are pri- mary means Of communication through personal interaction. Since the most influencial form of information is that 187 transmitted personally, it is apparent that the larger the migrant group in the recipient areas, the greater the flow of information through all channels to the potential migrant group in the United States. Most important, then, are the personal contacts generated when the new migrants return to the United States during social, medical, or shopping trips and interact with friends and relatives who represent potential migrants. Next to personal contact, the media plays an important part in attracting migrants. Although media can reach wide audiences and sometimes be as influencial as interpersonal communication, it is not always certain that the group to- ward which the information is directed receives it. Conse- quently, the most effective media must be that initiated by the in—country migrants in the form of personal letters and publications by organizations catering to this profes- sional group. Information transmitted through government and private agencies (embassy, military groups, Chambers of Commerce, and tourist institutes)(Figure 17) is generally aimed toward the specific Operation and mission of those agencies and therefore must be considered more biased than that exchanged through personal means. Without the existence of information exchange mechan- isms, retirement migration tO Central America would be highly disordered or random. Since information is readily exchanged through United States-Central American channels, 188 _msoHHsoHHn=N_ H mumpuoq — HHHpNHom a mosOHHh _ mBUoHNaHs= CON . m 060 s Hw 60H . NOW CON NVW OOH . Chfl QPHdO UO>OHQEH .M NNN.N NNN.H N can Hm>see NNN oos NNN.H ooN ooo.H NNN steam NNN.N NNN.N Nos.s cos.N NNN.NH Noe.H Heeos HmoHHSV mmsaanm H H H o H o .HN ooo.NHuN sneeze N N NH N N H .ee coo.Nue sneeze n v b m2 NH e mosmusm psosmshom NHN NNN NNN NNH NeH NN NHNNNN HNv NNN NNN NvH oNH NN Haeos Hsmnsszc NNHNHNNHN NNN.H oms NNN Nz NNN NHN mHnamHsas Hence HmOHHSV thBHowNS oNN sNN NNN msN see esoz moHH: Haeos mNNBHHNm mswmpmomz mapsosmm NHNEoussu Ammm>HNm Hui «Ohm ammoo ONHHom muowovmo Nnassoo I III! II «Hun-”lull II"! I [I'll meomHHanoo scammm eoHprHoomsNHB HmsHovsH .Hm OHQNB 201 the San Juan River as an outlet to the Caribbean Sea. Other significant navigable waterways are located in the Lake Iza- bal area of Guatemala and the Ulua River system in Honduras. Airways Air transport is an important means of travel in Cen- tral America, since some remote places are accessible only by air. It was found that several of the retirees, former Air Force pilots, had their own aircraft at their places of residence, for use on business, recreation, or social trips. Others, on occasion, hired private aircraft, or flew on scheduled commercial lines. The importance of in- ternal air travel is evident by the large numbers of small airfields located throughout the Central American nations, particularly in the vast inaccessible areas of eastern Nic- aragua, northeastern Honduras, and northern Guatemala. Comparatively, Nicaragua has the largest number of usable airfields (413), Costa Rica the most permanent surfaced runways (19), and Guatemala the most airfields with runways of 4,000-8,000 feet long (16). Also Costa Rica, Guatemala, Honduras, and Nicaragua have airports with runways of 8,000- 12,000 foot lengths. Highways The principal means of transportation by which the re- tiree population has located and interacts in Central Amer- ica is the automobile. The Inter-American Highway serves as a trunk line through the individual nations, with 202 primary, secondary, and tertiary linkages forming the in- dividual highway networks. Exceptions to this pattern are found in Honduras, where the Inter-American Highway passes through only a short southern section of the nation and in Belize, which has no direct connection. The Honduran road system is oriented along its main northHsouth highway con— necting the Caribbean port of Puerto Cortes, San Pedro Sula, Tegucigalpa and the Inter-American Highway. The retirees in Honduras have located along this axis. The Belize sys- tem connects with Mexico in the southeastern section of the Yucatan Peninsula, passes through Belize City and Belmo- pan, and terminates at the northeastern Guatemalan border. In Guatemala, El Salvador, and Nicaragua the retirees have located in the capital cities along the Inter-American Highway or short distances from that route, with direct ac- cess to it; e.g., Antigua and Solola in Guatemala and Toto- galpa in Nicaragua. Only in Costa Rica are the retirees more dispersed throughout the country, with some located at considerable distance from major paved highways; e.g., Gua- piles, Cahuita, Nosara, and Jaco. Most Costa Rican mili- tary retirees, however, are concentrated in settlements along the Inter-American Highway, principally in the Meseta Central (San Jose complex) and the General Valley (San Isi- dro). Others are served by secondary road connections; i.e., Miramar, Quesada, Turrialba, and Golfito. This re- tiree distribution pattern substantiates hypothesis 4.a., which states that the retirees locate mainly in the central 203 cities or adjacent areas connected by the major routes of communication. In total road mileage, Costa Rica overwhelmingly leads the other Central American nations (14,300 miles), followed by Nicaragua with 8,050 miles and Guatemala with 7,700 miles. In terms of paved road mileage, however, Guatemala ranks first with 1,600 miles; Costa Rica and Nicaragua fol- low with l,000 and 850 miles, respectively. Gravel-sur- faced and improved earth roads are often not well maintained and during the rainy season many are impassable. Since the majority of roads in Central America are of these latter classifications, it is apparent that the improvement and strengthening of these systems are necessary, if retirees are to be attracted to areas outside the main urban centers which lie along the Inter-American Highway axis. Social Networks Transportation systems form networks which are visable (highways and railways), or at least easily perceived (air- ways and waterways), in a spatial context. Other networks of equal importance in migration, but those difficult to envisage, are the social networks which result from human activities of a social, economic, or political nature. Like the transportation networks, these have two systems in Op- eration that are differentiated according to retiree inter- actions; i.e., those which take place between the host na- tion and the United States (external contacts), and those 204 that occur within the host nations (internal contacts). In these contact systems, retirees can be carriers, trans- mitters, or receivers of information and ideas. External Contacts Social networks linking host nations with other coun- tries are formed by external contacts of retirees with friends, relatives, and associates while visiting, vaca— tioning, participating in inter-regional organizational functions, shopping, etc. Social Activities The most significant external linkages reflecting the presence of military retirees in Central America are those based on the social activities inherent in military-related organizations. In the Central American nations 64 per cent Of the retirees hold membership in military affiliated asso- ciations and societies. Among this member group the most popular organization is the Retired Officers Association, which has attracted 67 per cent of those retirees who have joined a social organization, followed by the American Le- gion with 26 per cent membership, the Veterans of Foreign Wars and the Air Force Association with eight per cent each, and the Disabled American Veterans and Non-Commissioned Officers Association with five per cent each. Several other organizations hold smaller memberships. National conven- tions and meetings of these organizations are conducted in the United States periodically and attract members of the 205 Central American retiree group, thus linking the host nation with the United States in this social relationship. Furthermore, Costa Rica and Guatemala have local Amer- ican Legion posts, which participate in annual regional conventions held in the Panama Canal Zone. These posts are thus linked with similar ones in other Latin American na- tions, enabling the members to interact and exchange in- formation. Also, Costa Rica is unique in Central America, in that it has the only local Retired Officers Association chapter in the area. This chapter, therefore, is a node that is linked to other Latin American chapters of the Re- tired Officers Association (Mexico, Puerto Rico, and Vir- gin Islands), as well as with the United States national headquarters and other United States chapters. Economic Activities For military retirees in Central America, distinctive external economic linkages result from shopping and medical trips to both the United States and Panama. With available free military air transport between the Central American capitals and the Canal Zone, many retirees periodically use these facilities to make medical and shopping visits at the numerous military installations in the Canal Zone, where they take advantage of commissary, post exchange, and other military facilities. Some also use this means to travel from the Canal Zone to the United States for the same pur- poses. Private vehicles are used for similar trips to both 206 the Canal Zone and the United States. Other economic link- ages with the United States and neighboring Central American nations are those derived from business trips and other com- mercial ventures. Political Activities Obtaining documentation and authorization for interna- tional travel requires retirees to initiate connections with both United States and foreign governmental agencies, thus creating diplomatic-oriented linkages. An important media connectivity which links all retirees with their for- mer parent military services are newsletters written espec- ially for retirees and published by the retired activities Offices of the Army, Navy, Air Force, Marine Corps, and Coast Guard. These newsletters are transmitted bi-monthly by the Army, Navy, and Air Force and quarterly by the Marine Corps and Coast Guard to each retiree of their respective service, thus forming linkages between these United States- based organizations and individual retirees. This media form enables the retirees to maintain contact with one another. Internal Contacts In any new environment, individuals go through a space- searching stage in which they develop formal and informal contacts. For migrants settling in a foreign area, this search stage is particularly critical, since successful attainment of early contacts may sustain the migration, 207 whereas an absence Of contacts could result in disappoint- ment and failure. Migrants are first exposed to the local surroundings where they come in contact with many individ- uals and establishments. Subsequently, they learn of spe- cific institutions that interest them, and eventually, through these connections, form personal friendships. These combined linkages of a social, economic, and political na- ture, therefore, collectively form an internal contact net— work (Figure 20). Most individuals engage in the activi- ties depicted in this model, although by no means are these activities all-inclusive. The model does, however, provide a framework for discussion of certain important linkages in the internal contact network. Social Activities The social organization of space in Central America is a consequence of retiree participation in a variety of E2? cial activities. Most numerous are social linkages result- ing from personal visits with relatives and friends. An especially interesting relationship observed in San Jose, Costa Rica, was that of informal meetings by small groups of retirees in different coffee shops, some meeting daily and others once or twice a week. This activity appeared to be a favorite means of information exchange, since telecom— munications are limited in Central America and few retirees have private telephones. 8 0 2 xaosuoz womueoo HmenousH .ON mhsmHh — mo0H>Hom — amusmusmuwom_ — .m.u.<.H _ — eOHHNxNB — — mmosHmsm — _ meHxsmm _ _ NNNNHHHS _ _ _ _ _ _ _ _ _ .395 : 2.395 _ 72232:: :38: ~ _ _ _ — mmmsnsm fl _COHHNHMHEEHH —ouspmm HNOm_ — wcHooonm /\ mmHBH>HBU< HHBO< OHEOZOOM Hz /\ mWHBH>HBO< H «Hooaom oosHNmmsc mHNHoasz Ho soHuNoHHHpsooH wN bN NH DH 0H vH HH ON NH NH DH mH mH vH NH HH omeHsoo madam eOHom om oHsous< cam mN>NA .mOHm mosh mums mdeow NHOsHNH< Nd< Nueam chums cam sudomm omom sum r-INCOV‘ID CDNQOSO I-I 222 NN mNsNHN H azure 10°00. mmNKCuc no cumin: 0 O 0 ® ® @ e a. at. r .00... r W H o. n mcwhwlonzx J O 223 nearest retiree neighbor, seven per cent each from 6-10 miles and 11-15 miles, three per cent from 16-25 miles, and 11 per cent with retiree neighbors more distant than 25 miles. These findings attest to the integration of mili- tary retirees within their respective local communities. Consequently, hypothesis 4.c., which states that retiree clustering is directly related to housing and recreational deveIOpments cannot be supported. Despite the fact that military retiree clustering has not occurred within settle- ments in these nations, new housing projects and the devel- opment of recreational facilities in certain areas have evidently been influential in attracting retirees to those places. For example, in San Jose, Costa Rica the eastern suburban settlement of San Pedro, where the University of Costa Rica is located, is a popular living area for retir- ees because of the availability of new and modern houses, apartments, and apartment hotels. The western suburbs, also popular for retirees, Offer many new housing and apartment units, as well as recreation facilities (Sabana Athletic Field, Costa Rica Tennis Club, San Jose Indoor Club). Es- cazu, a suburban community four miles west of San Jose, is attractive to retirees because of its proximity to the Costa Rica Country Club. Several subdivisions have been developed in that area, including the prestigious "Bello Horizonte" section where the Costa Rican President resides. These areas, along with the Cariari Country Club and its surround— ing housing development located about six miles northwest 224 of San Jose, have the potential of attracting more retirees, if the retiree pOpulation in Costa Rica continues to grow. Effect of Distance in Costa Rica In migration theory, two generally accepted concepts are that migrants travelling long distances go directly to one Of the large urban centers, and that the numbers of migrants located in a major population center will decrease as distance from the pOpulation center increases (Raven- stein, 1885). For military retirement migration to Central America, the first of these concepts has been found to apply; i.e., the majority Of the retirees proceeded directly from the United States to the primate cities. Within Cen- tral America, it was believed that the number of retirees located outside the major cities would be inversely related to their distances from those cities. TO test this rela- tionship, Costa Rica was selected as a case study, since most of the retiree population of Central America is found there and its major urban center, San Jose, has the great— est concentration of retirees. To measure the effect of distance relative to the num- ber of retirees located outside of San Jose, two correla- tion-regression analyses were computed, using numbers of military retirees as the dependent variable and distances from San Jose as the independent variable. The first of these tests regressed the number of retirees located throughout Costa Rica against the most direct road distances 225 between San Jose and each location. This amounted to 29 distributions (retiree locations) in the Meseta Central area (Figure 22) plus 10 lying outside that region (Figure 13). Nine of the distributions contiguous with San Jose were combined into a San Jose complex, thus giving a total of 20 Observations in that area. Combined with the 10 others, there was a total of 30 Observations in the test. The results of the regression indicated little association (r = -0.l9) between these variables and only four per cent explained variation (r2 = .04). Moreover, a "t" test indi— cated no significance at either the .05 or .10 level, thus reflecting a weak linear rerationship. Since the main concentration of retirees is in the Meseta Central area (Figure 22) and the longest road dis- tance from San Jose to an outlying retiree location in that area is only 45 kilometers, it was believed that a separate test Of this area would reflect a valid relationship between the two variables. In this test, the 20 observations of retirees in the Meseta Central were again regressed against shortest road distances from San Jose to the retiree loca- tions. Although a higher association (r - -.42) was found in this case, still only 18 per cent (r2 = .18) of the re- tiree distribution was explained by the distance factor, and results Of the "t" test showed no significance at the .05 and .10 levels. Consequently, hypothesis 4.b., which states that the number of retiree migrants varies 226 inversely with the distance from the capital or major cities, cannot be substantiated. Since the distance factor accounts for little of the variation in retiree distributions, it is obvious that other locational factors play a more dominant explanatory role. As has been mentioned, the presence of modern housing, which conforms to American standards, and avail- able recreation facilities are important, as well as the availability of community and commercial services, and other socio-cultural amenities. CHAPTER VI SUMMARY AND CONCLUSIONS Central America has recently become recognized as a favorable location for retirement by United States military retirees because of its pleasant climate, less expensive living costs, and other amenities. In this study the nature of this specialized migration and the characteristics of its relevant migrant group are investigated, using a retire- ment migration schema as a framework. Summation The pertinent findings of this study are summarized in terms of military retirement concentrations, locational factors, migrant characteristics, and the spatial signif- icance of the migration. Military Retiree Concentrations In 1976 about 23,000 of approximately 1,050,000 mili- tary retirees resided in 115 foreign nations and terri- tories. These retirees are concentrated principally in the Western Pacific region, Middle America, and Western EurOpe. Within the Middle American area, Central America is a recent entry as an attractive region for retirees. 227 228 Among the six nations Of Central America, only Costa Rica, with 70 per cent of the retiree population, is pres- ently important as a retirement area. Sixty per cent of the retirees are located in the Meseta Central region Of that country; consequently, it is expected to remain most popular as a retirement site, since the presence of these retirees attracts others. In the remaining nations, most military retirees reside in the primate city. Second to San Jose in the numbers of retirees attracted is Guatemala City. That urban center, therefore, is expected to compete most strongly with San Jose in the settlement Of new retiree migrants, unless the February 1976 earthquake created an adverse effect. Factors Relevant to Central American Retirement Migration Physical, economic, social, and political factors play an important role in location choice of retirees by provid- ing a means of differentiating the individual nations. To evaluate the attributes of potential retirement sites, a retirement location choice model was structured for use in the assessment process. This model provides a means by which a potential retiree considers a number of positive and negative factors which could influence the selection of particular retirement sites, ranks selected sites, applies constraints, and then makes a choice. The relevant factors in attracting retirees to Central America are pleasant climate, accessibility to the United 229 States, special retiree privileges, political stability, social amenities, and low cost of living. The most favor— able climate Of Central America is found in the highland basins and plateaus. Consequently, the majority of mili- tary retirees have selected these areas for retirement, e.g., the Meseta Central of Costa Rica. Only in Nicaragua and Belize have the retirees chosen the less favorable low- land tropical areas, i.e., Managua and Belize City, respec- tively. Accessibility to the United States presents few prob- lems, since there are adequate road systems for travel by private automobile, direct air routes from several United States coastal cities, and "space-available" military air transportation from the United States to Panama, with con- nections to the other Central American nations. The cur- rent appeal of Central America to retirees, therefore, is partially attributed to its relative proximity to the United States, especially in comparison with accessibility to South America or other more distant potential retirement areas. The Central American nations that attract the major- ity of retirees have beneficial retirement laws; stable, democratic governments; fair immigration and customs pro- cedures; and close, cordial political relationships with the United States. Costa Rica, Guatemala, El Salvador, and Nicaragua have enacted laws which provide special benefits and privileges to qualified retirees. The major qualifying criterion of these laws is a guaranteed lifetime income of 230 a minimum monthly amount, ranging from a low of $250 in Guatemala to a high Of $400 in Nicaragua. Most military retirees with lifetime—assured government pensions can easily qualify and the majority of those in Central Amer— ica have taken advantage of these special programs. In all of Central America, 72 per cent of the military retir- ees had attained retired resident status under these laws, whereas 81 per cent had done so in Costa Rica, thereby at- testing to its pOpularity for full retirement. Moreover, Costa Rica's history of political stability and democratic freedom was found to be a strong factor in favor of its successful foreign retirement program. In the remaining nations, except Belize, which is a self-governing British colony, histories of frequent revolutions, government by dictator, and insurgency have tended to discourage retiree settlement. Amenities such as social, cultural, and recreational facilities; medical care; educational institutions; and commercial and transportation facilities exist in all the Central American nations, especially in the primary urban centers. A greater variety of amenities that satisfy the needs Of North American retirees, however, was evident in Costa Rica, followed by Guatemala, Nicaragua, El Salvador, Honduras and Belize. This ranking corresponds with that of the total number of military retirees located in these nations. 231 Living costs were found to vary somewhat from one nation to another and the high inflation rates Of the past few years are beginning to narrow the gap between costs of living in the United States and Central America. For the present, however, that area is still less expensive for retirement than the United States, especially in terms of housing, utilities, fresh foods, domestic services, public transportation, and building costs. Despite the fact that living costs in Costa Rica are higher than the other nations, it still attracts larger numbers of retirees because it provides a wider range Of acceptable commercial and domes- tic services and other amenities. Several important factors may deter a possible move. Since employment Opportunities are limited in Central Amer- ica, military retirees with low incomes and younger retir- ees with families, who by economic necessity are required to work, are more apt to remain in the United States or move to a foreign area which Offers employment opportuni- ties. Consequently, the Central American nations remain most suitable for migrants who desire full retirement. Natural hazards, which occur frequently throughout Central America in the form Of volcanic eruptions, earthquakes, and trOpical hurricanes, are very real threats to some retirees, thus deterring their settlement in these areas. Certain aspects of the Latin way of life and their attitudes discourage some retirees. For example, differ- ences in language, customs, traditions, and other social 232 institutions cannot be accepted by many North Americans and the shock Of this culture differentiation eliminates Central America as a desirable retirement area. The mafiana attitude Of the Latin irritates North Americans, as does the practice of extra-legal payments in business and polit- ical affairs. Moreover, goods and services available in the United States are often lacking in Central America, causing inconveniences that deter some potential migrants; e.g., potable water is rarely provided by local water sys— tems and faulty plumbing is common. Migrant Characteristics Military retirees have a history of frequent foreign moves due to their previous active military service, in which foreign assignments are numerous. The Central Amer- ican military retiree population has had a median number of five foreign assignments during their active military careers. Due to this conditioning to movement, it could be expected that former military personnel will continue to move periodically, including retirement migrations to foreign areas. This is the case with the retirees in Cen- tral America. It was found that 61 per cent of the migrants interviewed had at least one previous retirement residence before migrating to their present location. Of that number, 19 per cent had moved twice, 12 per cent three times, eight per cent four times, and two per cent were in their fifth 233 fifth and sixth retirement residences. Consequently, it appears that frequent movement by this group will continue. Seventy five per cent of the migrants indicated satis— faction with their present retirement site and 64 per cent considered it their permanent home. Nevertheless, some Of these "satisfied" retirees appeared disenchanted with the area because of the rising cost of living, lack of employ- ment, lack of community and commercial services, and emerg- ing social ills. The short length of residency in the host country also indirectly reflects the nomadic nature Of military retirees. The median residency period was 2.5 years, with 78 per cent of the retirees having lived in their retirement area less than five years. The short term residence history is ac- countable to some degree to the recent popularity of the area for retirement. It is more likely, however, that a turn-over of migrants has occurred, due simply to the in- herent desire of military retirees to move on to new and different surroundings on a cyclic basis. The imposition of penalties on those retirees covered by the special re- tiree laws who depart prior to the mandatory periods will force many to remain in-country for three to five years, still relatively short periods of residence. A similar study Of military retirees in the Lake Chapala region of Mexico revealed an average stay Of five years in that area (Allman, 1975). 234 Adjustment by the migrants to the Central American environment was facilitated in those cases where relatives and friends resided in-country. Fifty-nine per cent of the migrants had relatives and/or friends in the area prior to migration. "Culture shock" and anxiety were lessened by these contacts during initial introduction to the area, and appear to increase the likelihood that the migrants will remain. ‘ Retired members of the military are younger than the average civilian retiree, since the military retirement system offers the Option of retirement after 20 years of active service. The median age group at retirement for the migrant retirees in Central America was 40-49, whereas their median age at the time of the survey was 55. This time gap reflects either long term retirement residency in Central America, or periods Of retirement in places other than Central America, the latter representing the majority of cases. All of the retirees were of the male sex. Al- though this occurrence could be attributed to the least chance selection process, since there are fewer numbers of female military retirees, the Latin attitudes toward fe- males and the dominant role played by males in that society is more likely to have deterred female retirees. Moreover, the absence of non-Caucasian retirees in these nations may also be explained by the lower social status values placed on darker skin color in Latin America. 235 Eighty-one per cent of the retiree population was married and 98 per cent of the spouses of this group were present in the migrant household. In addition, about one- half (48 per cent) of these households included dependent children Or other relatives. This higher than expected number of families with dependent children is perhaps a reflection of the retirees married to native wives (34 per cent) who frequently are at child-bearing age. Retirees in Central America are an educated group; three—fourths have had some college, with 49 per cent hav- ing received a baccalaureate or higher degree. The educa- tion level of their spouses is somewhat lower, 58 per cent having attended college. Also of selectivity significance is the fact that 88 per cent of the retirees are able to speak Spanish with varying degrees of fluency. The high level of education may be reflected by a favored pastime of the retirees. According to the survey, the greatest proportion of their leisure time is spent in reading and writing, followed by socializing and recreation. This ranking of activities was an unexpected finding, since it is usually believed that a young and active retiree group might devote most of its time and energy to recreation. The majority (61 per cent) Of the retirees migrated to Central America for full retirement. This is especially true in Costa Rica where 72 per cent of the migrants are fully retired. In the other nations (except Belize), how- ever, most of them are either fully or partially employed. 236 In fact, employment is perhaps the most important factor influencing migration to those countries. The retirees who have migrated to Central America to enjoy full retire- ment are able to do so, since they receive a relatively high guaranteed income, the median annual pension being $8,900. In a relative comparison with the yearly average retirement income for all United States military retirees ($6,648), that of the Central American group is signifi- cant. Moreover, it indicates that retirees with higher incomes Opt for Central American retirement, thus providing capital needed by these small developing nations. There is no selectivity for the parent military ser- vice of the retirees, since all are represented in near proportion to their sizes; i.e., Army, 37 per cent; Air Force, 30 per cent; Navy, 22 per cent; and Marine Corps, nine per cent. Neither is military rank selective, al- though more commissioned Officers (55 per cent) are found among the migrants than enlisted men (41 per cent) and warrant officers (four per cent). Retirement grades among these personnel ranged between sergeant (E5) and lieutenant general (09), which means an annual retired pay range of $4,080 to $29,700. Consequently, lower ranking enlisted personnel with small annual incomes apparently are less likely to Opt for foreign retirement, especially in areas that offer few employment Opportunities. The principal reasons given by the migrant retirees for their decisions to migrate to Central America were 237 favorable climate and low cost of living. In Costa Rica, the most attractive of the Central American nations, the majority of the migrants attributed their migration decision primarily to agreeable climate, secondly to low cost of living, and thirdly to the special privileges granted to retirees. Among the other nations, that country which most closely corresponds with Costa Rica in terms of attractive- ness for full retirement is Guatemala. It, too, was chosen according to the same ranked criteria as Costa Rica. Re- tirees in the remaining nations, however, especially Hon- duras, have located in those countries primarily for em- ployment or to be near relatives. Spatial Significance of the Migration Migration entails an exchange of people and informa- tion. To determine how this occurs through space, a Migrant Circulation System model was designed which shows that mil- itary retiree migration to Central America is an ordered phenomenon. In this model a migrant is carried through a series Of steps in which he evaluates and moves to a retire- ment destination. To test the concept of the model, Costa Rica was used as a case study. A component in the system, the exploratory trip, is the single most important element in the migration process for the retiree group, since it insures a greater chance of the migration succeeding, with dissidents filtered out of the system at this early stage. Of those retirees who made this exploratory trip to 238 Costa Rica, only one of ten later made the actual move. Moreover, 78 per cent of the Costa Rica respondents decided to migrate on the basis of positive information obtained during the exploratory trip. At this point, early in the migration process, personal observation and evaluation can indicate the potential success or failure of the proposed migration. If overriding negative information is exchanged, a decision not to migrate can be made before large sums of money are expended on the move. After the exploratory trip and the decision to migrate is made, the migrant arrives in-country and settles in a gateway residence while processing resident pensioner ap- plications and making arrangements for permanent housing. These gateway residences, located in San Jose, provide temporary quarters, usually consisting of furnished apart- ments or apartment—hotels, which have become well known and much in demand by retirees. Some, Of course, live tempor- arily in San Jose hotels, and those who entered by travel trailer reside in one of the several nearby trailer parks. The majority of these newly-arrived migrants obtain resi- dent pensioner status and move into a permanent residence. Though most of the retirees considered Costa Rica their permanent home, 70 per cent had lived there less than four years and their migration histories suggested many Of them would eventually move on, either back to the United States or to another retirement location. Those remaining in Costa Rica until death complete the final stage of the 239 migration circulation system. The other nations of Central America do not presently have sufficient numbers Of mili- tary retirees to apply the model; however, they are of potential importance as viable retirement areas and the model's application may be of future use. At all times during the migration process the migrant is a carrier Of information which can be passed on to other potential migrants. This is especially true during visits to the United States or Panama, which the retirees make periodically. Here they normally interact with the mili— tary community and may influence others in making a retire- ment location choice. A majority of the potential migrants had contacts in Costa Rica, either friends, relatives, or other military retirees who assisted them in most part because they were military. It is the latter case that suggests a modified chain migration effect, e.g., these new migrants who had been assisted by the in-country retirees were subsequently "socialized" and they then became providers of information and assistance for new migrants. Directional bias was present in the migration streams. Most retirees entering the migrant circulation system came mainly from the popular retirement states of CalifOrnia, Florida, and Texas. These three areas supplied 58 per cent of the migrants, with California alone providing 27 per cent. The only other important generating area was the Virginia-District of Columbia complex, which accounted for 240 10 per cent of the Central American retirees. About 15 per cent of the migrants were from the major population states of the northeast and the few others from midwestern and western states. It is believed that most military retirees in Central America worked for a while directly after retire- ment from the military, retired to an amenable area in the southern United States, and then migrated to Central America. Two networks are in Operation to sustain the migration stream and information flows. These consist of transporta- tion and social systems which operate externally to connect the United States and the host nations, as well as intern- ally, where they represent the means for in-country move- ment and personal interaction. The interrelationships and degree of connectivity among the components of these net- works account for their complexity. The more complex the networks, the greater the probability of migrant and infor- mation flows. Within the individual nations of Central America, the retirees have located mainly in the major urban areas or their surrounding suburbs and villages (83 per cent), espec- ially those situated along the main highways. Of those not living in urban areas, 10 per cent resided on farms, four per cent at beach areas, and three per cent on ranches. The primate cities attracted most, apparently because of the available services and activities that are important to retirees, and few of which exist at other places. 241 There was no evidence of military retiree colonization or clustering within the major settlements and 96 per cent of the retirees indicated that they resided in integrated communities. Correlation-regression analyses were made to measure the effect of distance relative to the number of retirees located outside Costa Rica's primate city, San Jose. Results of these tests indicated that the distance factor accounts for little Of the variation in retiree dis- tributions. Migration Theory There are no precedents in the literature relative to international retirement migration per se. However, the general principles involving internation movements, or any form Of migration for that matter, are similar. Conse- quently, identification of deviations from the generally accepted migration theories, and the application of new ideas which incorporate these differences, are necessary for the advancement of knowledge in this field. Application to Existing Theory The process Of migration is normally considered in migration theory by means of a conceptualized spatial model which includes an origin, a destination, and an intervening space which contains a number Of possible Obstacles. Within this spatial framework, potential migrants react to positive or negative factors at the origin and destination, and if 242 the decision to move is made, during the move they are sub- jected to various Obstacles that exist between these places. Obviously, any form of migration can be considered within this conceptual framework. Furthermore, the basic laws Of migration, as developed by Ravenstein (1885) and modified by Lee (1966), can be examined for their applicability at various stages of this migration process. Development of a Retirement Migration Theory In this study, the existing concepts and theories of migration, along with the concepts of Mabogunje (1970), were used to formulate a framework for examining the Cen- tral American case. This system, the retirement migration schema (Figure 4), provides a guide that insures complete investigation of all aspects of the retirement migration process. The schema illustrates the functions of migration elements and the processes involved in migrant movements, as well as the relationships Of the component parts to each other and the outside environment. Within this schema, donor and recipient components, along with their respective sub-components, represent specific populations located within a spatial context. (For a detailed explanation of the model refer to Figure 4 and pages 32-34.) Although the schema illustrates the basis for spatial interaction by mil- itary retirees, certain components within the model require additional explanation; i.e., migration channels and infor- mation feedback. Consequently, to more definitively explain 243 migrant circulation patterns and information flows, supple- mentary models were designed to illustrate and explain these phenomena (Figures 16 and 17). It is believed, therefore, that these models constitute the foundation for an under- standing of international retirement migration and will prove useful in developing theory applicable to that end. This study represents a beginning toward that task, since it has introduced and identified some deviations from tra- ditional migration theory; e.g., the inapplicability of time/cost distance measures, stage migration, and economic motivation; while demonstrating the importance Of inter- action and information exchange, and the significance of amenities, as criteria for site selection. Consequently, to advance research and develop theory in the increasingly important field of retirement migration, it is suggested that other empirical cases be examined using the framework of this study. Application This study has provided information on military retiree migrants in Central America, including their locations, con- centrations, method of transport to Central America, and how they interact within their respective communities. Other than its inherent contribution to the international migra- tion literature, therefore, the study should prove useful to potential migrants by providing an overview of the area, and to host nation governments for planning purposes. 244 Socio-Economic Aspects In the Central American countries there is little evi- dence of an "Americanization" effect upon the host country's cultural landscape. NO retirement colonies have been devel— oped in the area where retirees might cluster, although a few real estate schemes for sub-divisions and recreation areas which might cater to foreign retirees are found in a developing stage; e.g., the Nosara Beaches of Costa Rica and the San Jose suburban area of Escazu. Consequently, most retirees have successfully integrated into their re- spective host countries and participate in many of the so- cial and cultural activities of the local community. Retirees living in the host nations under the provi- sions of special retiree laws represent a source of imported capital, since their income is derived from outside sources. However, at the present, in all countries except Costa Rica, the amounts are insignificant, due to the few military re- tirees in these nations. But, in Costa Rica this income represents an annual input to the local economy of over $1,000,000, a significant amount for a nation of this size. In addition to the dollar inflow to the economy, the pres- ence of retirees creates employment Opportunities for na- tionals, especially in the sectors of domestic help, con- struction labor, and commercial services. If the host countries wish to retain their military retirees, it is important that they provide amenities and 245 commercial services, such as supermarkets, drug stores, restaurants, and recreational facilities, some of which are found in most of these nations. These types of facilities are especially prevalent in the Escazu suburb of San Jose, Costa Rica, where military retirees are concentrated. With- out the provision of amenities and commercial services, the attractiveness of Central America for its climate, scenery, and lower cost of living, could be out-weighed by inconven- iences. Furthermore, these problems are compounded in Cen- tral America by government bureaucracy. The possibility of a "disenchantment syndrome" in some nations could set in motion a reverse movement that would reduce the military retiree population. This situation is now apparent in Costa Rica, with larger numbers of retirees becoming discontented over escalating problems with the government, the same so- cial ills found in many large cities in the United States, and inflation boosting the cost of living. Consequently, some retirees have already departed and others have made plans to leave over the next year or two, especially if living costs become equal with those of the United States (Harding, 1976).‘ Planning and Policy Formulation Officials and planners in the host natiOns can use this study to ascertain where military retirees are located in their countries and their possible impact in areas of concentration. Those nations desiring more dispersed 246 distribution of retirees and the development of recreation- type settlements such as Nosara Beaches Of Costa Rica, must improve their communications infrastructure, thus providing more avenues for settlement. Nations now having few retirees and wishing to attract more for national development purposes (capital import) could follow the example and experiences of Costa Rica, which has a history of successful governmental programs oriented toward settlement of foreign resident pensioners and promoting its advantages for retired living. It appears that a concentrated promotion and advertising campaign is required which describes the attributes of the nation for retirement and its advantageous laws for retirees. Toward this end, the study provides information regarding those areas in the United States where retirees are concentrated, enabling promotion activities to be directed to a more re- ceptive group. A final conclusion pertains to the field survey method used in Obtaining the research information. It is apparent that an interpretative study of this type can only be ac- complished through personal exposure to the geographic area where direct observations and experiences reveal general trends as well as specific information on the investigated phenomenon. This field procedure, combined with the per- sonal interview technique, is therefore the recommended method for data collection for similar research studies. APPENDIX A MILITARY RETIREE MIGRATION HISTORY QUESTIONNAIRE UIIbOONI-I APPENDIX A MILITARY RETIREE MIGRATION HISTORY QUESTIONNAIRE (Country) Part I -- Personal Data . Name (Optional) . Address (Town) (State) Sex: Male _ Female Race: Caucasian Negro Other Present age: Under 40 40-44 45-49 50-54 55-59 60-64 65—69 70 plus . Marital Status: Single Married Divorced Widowed . Persons in present household: Spouse # Children Other (who and #) Education (highest year completed): Self____ Spouse____ Employment Status: Fully Retired___ Fully Employed___ Part-time employed___ Seeking employment___ If employed, working in: For self Education Government Business Other Part II -- Financial Data . Present military retirement income: Under $5,000 $5-10,000 $10-15,000 $15—20,000 $20-25,000 $25-30,000 Over $30,000 Present annual income from all sources, including joint: Under $5,000___ $5'10’000___. $10-15,000___ $15-20,000___ $20-25,000___ $25-30,000___ Over $30,000___ 247 248 Part III -- Military Service Data . Parent Service: Army___ Navy___ Air Force___ Marines____ Other Retirement grade or rank (El-E9, WOl-WO4, 01-010) (Specify) Years military service completed: 19 or less___ 20-24___ 25—30___ Over 30____ . Special skills and interests during military duty Age at retirement: Under 40__ 40-49__ 50-59__ 60-69_ Last duty station: (Countryl’ (state) (Eity/EOSt) . Place of residence prior to entering the military Age at entry into military: Part IV -- Military Service Miggations Number of continental U.S. moves during mil. career: . Principal duty stations or states to which you were as- signed in the U.S.: What states or areas did you like the most during this period? Number of foreign moves during military career: Principal countries assigned: Which of the above countries were most appealing to you during these assignments? . Why? Part V -- Retirement Migrations . Previous Retirement Residency (The following moves per- tain to those accomplished prior to your current foreign area residency. If your current foreign area residency was initiated directly after retirement, proceed to paragraph B.) 1. FIRST MOVE: a. Place (countryll (fifty)? (State) b. Year arrived Year departed 249 c. Reasons for choosing this location (See factors below): (1) Most important (2) Second most important (3) Third most important (Factors to consider in answering above) Close to military facilities Climate (warm, temperate) favorable Low cost of living area Social and cultural activities Near last duty station (job, friends, school) Low risk natural hazards area Job Opportunities (government/private) Hometown (relatives and friends) Property ownership (land/home) Recreation activities Favorable taxes Health considerations (medical facilities) Political conditions favorable Other (please describe) d. Reasons for leaving this location (See factors be- low): (1) Most important (2) Second most importint‘ (3) Third most important (Factors to consider in answering above) Inadequate educational facilities Poor relations with relatives, friends Lack of community and commercial services Lack of social, cultural activities Lack Of recreational activities Political conditions (unstable government, etc.) Disagreeable climate High cost of living Poor tax structure Job dissatisfaction Social ills (crime, drugs, etc.) High risk natural hazard area (earthquakes, etc.) Other (please describe) 2. SECOND MOVE: a. Place (country) (city) (stateTF b. Year arrived Year departed c. Reasons for choosing this location (See factors listed above) (1) Most important (2) Second most important (3) Third most important d. 250 Reasons for leaving this location (See factors listed above : (1) Most important (2) Second most important (3) Third most important 3. THIRD MOVE: a. b. Place (country7' ‘(Oity)’ (state) Year arrived Year departed c. Reasons for choosing this location (See factors listed above): (1) Most important (2) Second most importint (3) Third most important d. Reasons for leaving this location (See factors listed above : (1) Most important (2) Second most important (3) Third most important B. Present Retirement Residency 1. Pre-Migration Information: a. b. What was the most important source of information about the host nation which influenced your re- tirement migration decision? (Specify) How much influence did each Of the following in- formation sources have in providing you with know- ledge Of the area before moving there: Most Great Some Little (1) Personal visits (2) Relatives or friends (3) Radio or TV '___— (4) Newspapers (5) Magazines & books (6) Host gov't publ. (7) Mailed advertisements (8) U.S. gov't info. (9) Other Did you visit the host country before deciding to move there? Yes NO Before moving to the host nation did you have rel- atives or friends living there? Yes No Did they influence your movement decision? Yes No . Did you own property in the host country before moving there? Yes No . If yes, how long before? Months 251 2. Migration Decision Data: a. When did you actually arrive in the host country for retirement purposes? Date Years of residency b. What were your main reasons for this migration? c. Indicate the degree of importance you attach to following reasons for migration? the (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Climate Low cost of living Special retiree priv. Social & cult. act. Recreational activities Near relatives, friends Health considerations Job considerations Property ownership Accessibility to U.S. Political conditions Other Most Great Some Little * _ —— d. During your migration decision making process, did you consider other foreign areas for retire- ment? Yes No . If yes, where? e. Why did you choose the host country instead? 3. Place Location Data: a. Where did you first locate in the host nation? (geographic place) b. Type of residence: Apt. Mobile home in Condominium House c. Is the residence located: On a farm suburbs village central city beach area ranch d. If urban oriented, is the residence in a so-called "retirement community" neighborhood ? or in a regular mixed e. What is the distance between your present resi- dence and that of your nearest military retiree neighbor? Miles 4. Permanency and Area Satisfaction Data: a. Do you consider this to be your permanent retire- ment home? Yes No b. If No, where is your permanent home? C. p. C swarm- . Do you plan to move soon? Yes No 252 If you consider your residency in this nation to be of a seasonal or temporary nature, how fre- quently and for what period of time do you leave the area? Frequency Period of time . When you leave the area, where do you normally go? (1) Vacation in other foreign areas (2) Visit relatives and friends in Otfier foreign areas (3) Vacation Ifi—the U.S. (4) Visit relatives and TFTends in the U.S. (5) Return to permanent home____ During the visits or vacations specified above, do you encourage other retirees or potential retirees to live in the host nation? Yes No If Yes, reasons why: If No, reasons why: In your retirement environment, how much of your time during the average week is spent on the fol- lowing activities? Most Much Some Little (1) Job or business (2) Socializing (3) Housework (cleaning repairs, gardening) (4) Recreation (sports) (5) Rel. & community act. (6) Reading & writing (7) TV & radio (8) Education (9) Other Are you satisfied with your present retirement location? Yes No If No, where would you like to move? Why there? If you plan to permanently leave the host nation, to what degree Of importance do you attribute the following reasons for leaving? Most Great Some Little (l) Disagreeable climate (2) High or rising cost of living (3) Lack of employment (4) Lack of community & commercial services 253 Most Great Some Little (5) Lack Of social & cultural activities (6) Lack of recreat. act. (7) Inadequate educational facilities (8) Poor relations with foreign neighbors (9) Poor relations with relatives & friends (10) Social ills (crime, drugs, etc.) (11) Political conditions (unstable gov't) (12) High risk natural hazard area (13) Others 5. Miscellaneous Data: a. What mode of transportation was used in your move to the host country? Auto Ship Air Other . Place of departure Place of arrival Was your move direct to the host nation from place of departure or was it accomplished in stages or steps? Direct Stages If in stages, were there any significant periods of time spent in other nations? (Specify) As a retiree, do you receive any special rights or privileges granted by the host nation? Yes NO If applicable, what government law or decree per- tains to these privileges? Do you speak the language of the host nation? Yes NO DO you hold memberships in any military or mili- tary-related organizations? Yes No If Yes, specify Do you subscribe to or receive any military-related publications? Yes NO If Yes, specify Was or is your spouse a native of the host nation in which you now reside? Yes No APPENDIX B U.S. MILITARY STRENGTH OUTSIDE THE UNITED STATES APPENDIX B U. S. MILITARY STRENGTH OUTSIDE THE UNITED STATES March 31, 1976 Total Outside the United States . . . . . 481 000 U.S. Territories and Possessions (includes Afloat) . . . . . . . . . . 31,000 Foreign Countries . . . . . . . . . . . 450,000 Western Europe and Selected Areas Related Areas . . 306 000 Southeast Asia . . 4,000 _Belgium . . . . 2,000 T5ailand' . . . 47000 Germany . . . . 217,000 Greece . . . . . 4,000 Western Pacific . 129,000 Iceland . . . 3,000 Japan (incl. Italy . . . . 12,000 Okinawa . . . 48,000 Morocco . . . 1,000 Philippines . . 16,000 Netherlands . . 2,000 South Korea . . 41,000 Portugal/Azores 1,000 Taiwan . . . . . 2,000 Spain . . . . . 10,000 Afloat . . . . . 22,000 Turkey . . . . . 7,000 United Kingdom . 21,000 Other Areas . . . 41 000 Other . . . . . 1,000 AuEtralIa . . . 1,000 Afloat . . . . . 27,000 Bermuda . . . . 1,000 » Canada . . . . . 1,000 Cuba . . . . . . 3,000 Less than 250 Guam . . . . . . 9,000 Bahamas Iran . . . . . . 1,000 Bahrain Midway Is. . . . 1,000 Barbados Panama C.Z. . . 9,000 Ethiopia (incl.Eritrea) Puerto Rico . . 5,000 Johnston Is. Other . . . . . 5,000 Leeward Is. (Antigua) Afloat . . . . . 5,000 Norway Brazil Less than 1,000 Greenland Saudia Arabia All other countries: Less than 100 U.S. military personnel Totals may not add due to rounding. Source: Department Of Defense (DOD, 1976) 254 BIBLIOGRAPHY BIBLIOGRAPHY AAP (1976). Asociacion de Amigos del Pais, Guatemala News- letter (Guatemala City) April 22. Allman, Phillip H. (1974). "Case History of a Career Mili- tary Retiree," an unpublished paper prepared at Michi- gan State University, February, 1974. (1975). "A Comparative Study Of American Military Retiree Migrants in Mexico: The Case of San Miguel de Allende and the Lake Chapala Region," an unpublished field problem report prepared for the Department of Geography, Michigan State University. American Legion (1976). "News of the American Legion," The American Legion Magazine, 100/6: 28-32. Army Times (1970). "Retirement Supplement," Arm Times (Washington, D.C.: Army Times Publishing Co.) OEtober 28, 1970, p. 13R. (1972-1976). "Retirement Supplement," Army Times (Washington, D.C.: Army Times Publishing Co.). (1973). "Military Facts and Figures," Army Times (Washington, D.C.: Army Times Publishing CO.)_Septem- ber 26. (1975). "Retirement Supplement," Arm Times (Wash- ington, D.C.: Army Times Publishing 0. ep ember 17, pp. 1-28 0 Beaujeu-Garnier, J. (1966). Geography_of Population (Lon- don: Longmans, Green and CO. Ltd.). Buckley, Joseph C. (1967). The Retirement Handbook (New York: Harper and Row). Bunge, Mario (1971). "Four Models of Human Migration," General Systems, 16: 87-92. Carrothers, Gerald A. P. (1956). "An Historical Review of the Gravity and Potential Concepts of Human Interac- tion," Journal of the American Institute of Planners, 22/2: 944102. 255 256 CIA (1976). Central Intelligence Agency, Office of Geo- graphic and Cartographic Research. National Basic Intelli ence Factbook (Washington, D.C.: Government Piinting Office) JuIy. Costa Rica (1971). Regulation of Law No. 4812 and Proce- dures (San Jose: InstithtO‘COstarricense de Tirismo). Cox, Kevin R. (1972). Many Location, and Behavior: An éntroduction to HumanGeography (NewYOrkT_Wiley and ons)L Demko, George J., Harold M. Rose and George A. Schnell (1970). Population Geggraphy: A Reader (New YOrk: McGraw—Hill). DOD (1976). Department of Defense, Office of the Assistant Secretary of Defense (Comptroller), Directorate for Information Operations and Control. Selected Manpower Statistics (Washington, D.C.: Government Printing ce une. Donovan, James A. (1970). Militarism, U.S.A. (New YOrk: Charles Scribner's Sons). El Salvador (1973). "Ley Especial Para Residentes Ren- tistas," Diario Oficial, Decree 476 (San Salvador: Ministry of the InterIOr). Fischer, Edward (1973). Why Americans Retire Abroad (New YOrk: Sheed and Ward: Inc.). Ford, Norman D. (1974). Fabulous Mexico (Greenlawn, New York: Harian Publicitions). Foster, Donald C. (1976). "Membership Survey," The Retired Officer, 32/10: 18-21. ‘ ’ Frisbee, John L. (1973). "Retiring in Costa Rica," The Retired Officer, 29/5: 22-25. Gates, Ed (1974). "Where to Retire," The Retired Officer, 30/5: 22-26. Golant, Stephen N. (1971). "Adjustment Process in a Sys- tem: A Behavioral Model of Human Movement," Geograph- ical Analysis, 3/3: 203-220. Guatemala (1973). Congress of the Republic, Decree 58-73, Law of Retired and Annuitant Residents (Guatemala City: Guatemalan Tourism Institute). 257 Haenszel, William (1967). "Concept, Measurement, and Data in Migration Analysis," Demography, 4: 253-261. Hagerstrand, Torsen (1969). "On the Definition of Migra- tion," Yearbook of Population Research in Finland, 11: 63-72. Harding, Wade P. (1973a). "Dear Former Friend, or Retire- ment in Costa Rica Revisited," The Retired Officer, 29/10: 30-33. (1973b). Personal letters (San Jose, Costa Rica) May 1973 and February 1974. (1976). Personal letter (San Jose, Costa Rica) August. Hollingsworth, T.H. (1970). Migration: A Study Based on Scottish Experience Between I939 and I964 (Edinfiurgh: Oliver and BOyd). Hotchkiss, P. (1975). "The United States Army," The Re- tired Officer, 31/6: 18-22. Instituto Panameno de Turismo. Retire to Panama and Live, an undated brochure (Panama City). Jones, F.E. (1956). "Sociological Perspective on Immigrant Adjustment," Social Forces, 35/10: 39-47. Kahn, Theodore C. (1974). "Military Careerists: A New Scapegoat Minority," The Retired Officer, 30/1: 20-23. Kariel, H.G. (1963). "Selected Factors Areally Associated with Population Growth Due to Net Migration," Annals of the Association of American Geographers, 53: ZIO- 223. Kirk, Dudley (1958). "Major Migrations Since World War II," Milbank Memorial Fund Quarterly, 36: 11-28. Lee, F.S. (1966). "A Theory of Migration," Demography, Leonberger, Loren B. (1973). "A Marvelous Mirage," The Retired Officer, 29/6: 22-25. Lien, Maurice L. (1974). "Notes and Comments," The Retired Officer, 30/4: 12-19. (1976a). "Notes and Comments," The Retired Officer, 32/3: 16-17. 258 (1976b). "News, Notes and Comments," The Retired Officer, 32/10: 12-15. (1976c). "Notes and Comments," The Retired Officer, 32/12: 12-19. Mabogunje, Akin L. (1970). "Systems Approach to a Theory of Rural-Urban Migration," Geographical Analysis, MAcDonald, John S. and Leatrice D. MacDonald (1964). "Chain Migration, Ethnic Neighborhood Formation, and Social Networks," Milbank Memorial Fund Quarterly, 42: 82-97. Maselli, G. (1971). "World Population Movements," Interna- tional Migration, 9/3-4: 117-125. Neiva, Artur H. (1965). "International Migrations Affecting Latin America," Milbank Memorial Fund Quarterly, 43/4: Nicaragua (1974). National Constitutional Assembly, Decree 628 (Tegucigalpa: Ministry of Economy, Industry and Commerce, National Tourism Office). Peterson, William (1955). Planned Migration: The Social Determinants of the Dutch-Canadiin Movement—(Berkeley: UniverSity of California’Press). (1958). "A General Typology of Migration," American Sociological Review, 23: 264-273. RAB (1976). Retired Army Bulletin. "Women in the Military Services," Retired Army Bulletin, 6/11-12: 5. Ravenstein, E.G. (1885). "The Laws of Migration," Journal of the Royal Statistical Society, 48: 167-227. Sawatsky, Harry L. (1971). They Sought a Countiy: Men- nonite Colonization in Mexico (Berkeley: University of CaIiiornia Press). Sleed, Joel (1974). "Retired Gringos Find Green Fields in Old Guadalajara," Detroit Free Press, January 13. Stouffer, Samuel A. (1940). "Intervening Opportunities: A Theory Relating Mobility and Distance," American Sociological Review, 5/6: 845-857. Tarver, James D. (1961). "Predicting Migration," Social Forces, 39: 207-213. 259 Ter Heide, H. (1963). "Migration Models and Their Signif- icance for Population Forecasts," Milbank Memorial Fund Quarterly, 41: 56-76. Thomas, Robert N. (1968). Internal Migration to Guatemala Cityy C.A., a Ph.D. dissertatiOn, Départment of Geo- graphy, Pennsylvania State University. Times of the Americas (1973). "Mexico Considers Limits on FOreign Residents," Times of the Americas (Washington, D.C.) October 17, p. 12. TROA (1975). The Retired Officers Association. "Some Facts on Veterans' Numbers," The Retired Officer, 31/12: 41. (1977). The Retired Officers Association. "Military Retirees: Who and Where," The Retired Officer, 33/3: 11. Turner, Kernan (1973). "Costa Rica Promoted as Retiree Home," Inquirer and News (Battle Creek, Michigan) February 182 Ullman, E.L. (1954). "Amenities as a Factor in Regional Growth," Geographical Review, 44: 119-132. USBC (1970). United States Bureau of the Census. Census of Population (Washington, D.C.: Government Printing ‘Oiii667. USDA (1972). United States Department of the Army. Area Handbook for Panama (Washington, D.C.: Government Printing Office). (1975). United States Department of the Army. Hand- book on Retirement Services (Washington, D.C.: Govern- ment Printing Office). USAID (1976). United States Agency for International De- velopment. "The Relief Effort: A Matter of Loving," War on Hunger, 10/4: 10. USDS United States Department of State. Background Notes, Fact sheets on foreign nations (Washington, D.C.: Government Printing Office). Wagley, Charles (1973). "An Introduction to Latin American Culture," Handbook of Required Readings - International Relations and Government Department TFOrt Benjamin Hirrison, Indiana: DETEnse‘lnformation School), pp. 73-89. 260 Ware, George W. (1968). The New Guide to Happijetirement (New YOrk: Crown Publisher'). West, Robert C. and John P. Augelli (1966). Middle America: Its Lands and Peoples (Englewood Cliffs, N.JT: Pren- ticeaHall)} Wolpert, Julian (1965). "Behavioral Aspects of the Deci- sion to Migrate," Papers and Proceedings of the Re- gional Science AssociatiOn, 15: 159-169} (1966). "Migration as an Adjustment of Environmental Areas," Journal of Social Issues, 22: 92-102. Zipf, G.K. (1946). "Some Determinants of the Circulation of Information," American Journal Of Psychology, 59: 401-421. "IIIIIM'LIIIIIIIIIIIITIIB