Trying to buttress the failing economy, the Castro government has assigned high priority to increasing the volume of emigrant remittances. The government has also liberalized travel rules to induce more Cuban-Americans to visit the island. Both measures, together with an all-out effort to promote international tourism, are intended to generate desperately needed foreign exchange, now that Soviet subsidies are a distant memory.

Out of concern with the hardships being faced by their relatives, many families of Cuban origin have responded by sending dollars to Cuba. No one knows for sure how many dollars are being sent. Some of the figures bandied about assume that as much as $1 billion may be flowing to Cuba annually, although, in my view, the amount is far more modest. This judgement is based on a preliminary assessment of socioeconomic and income data on the Cuban-American community obtained from the 1990 U.S. census, the demographic evolution of the Cuban-American community, and on a review of international evidence regarding the determinants of remittance flows, and how these determinants apply to present-day Cuba.

How much is remitted from one country to another is determined by a complex array of factors. Some of these include the number of emigrants abroad, the wages they earn, the emigre community’s disposable income, and economic conditions in host countries. Political and economic developments in home countries are also implicated in remittance behavior, as are exchange rates, perceptions regarding the certainty remitters have that intended beneficiaries will receive the monies, and the ease with which remittances are sent.

In this paper I evaluate some of the variables likely to determine the magnitude of dollars flowing to Cuba from the emigre community abroad, particularly from the United States. I also propose a simple methodology (that could be further refined) based on U.S. census data to calculate a plausible range of remittance estimates, both while the Castro regime remains in power and after a transition occurs.

I.  Historical Determinants of Cuban-American Remittances

The remittance policy of the Cuban government is predicated on the assumption that Cubans abroad will behave in the same way as emigrants from other countries, and that they, therefore, are likely to remit a considerable portion of their disposable income. There are several factors associated with Cuban emigration to the United States, however, that suggest that the Cuban-American community is prone to have a lower propensity to remit than other Latin American emigrant communities residing in the United States. Most significant and obvious is the political underpinning of Cuban emigration, particularly during the early post-revolutionary years. Emigrants from other origins (e.g., Mexico, Dominican Republic) left their countries mainly to seek better economic opportunities abroad, or came to the United States for a mixture of political and economic reasons (e.g., El Salvador), just as many Cuban arrivals did in the past and continue to do today.

These distinctions can be used, a grosso modo, to identify two historically distinct Cuban emigration flows: those arriving in the United States before and after 1980. This year is, for several reasons, an important watershed in Cuban emigration. The 1980 immigrant cohort was the largest (about 125,000) ever to arrive in the United States in a single year; 1980 was also the only year since 1971 when the number of annual arrivals exceeded 50,000, well over the average (except for 1980) of the last two decades (Diaz-Briquets and Pérez, 1981) [ref]Between 1982 and 1992, the average annual number of Cuban nationals admitted as immigrants was 15,504 (U.S. Immigration and Nationalization Service, 1992 Statistical Yearbook, Washington, D.C., 1993, Table 3, p.31). This figure, however, does not reflect the actual number of U.S. arrivals, since it includes the adjustment to resident status of many Cubans who arrived in the United States through the Mariel sealift, especially in 1985 (20,334 immigrants admitted), 1986 (33,114), and 1987 (28,916. The immigrants admitted data also ignore those Cubans entering as non-immigrants but overstaying their visas, without inspection (crossing an international border surreptitiously), or by sea (e.g. rafters). But even when allowing for these arrivals, the annual number of Cubans entering the United States during the 1980s and early 1990s has been much lower than during the late 1960s and early 1970s. This applies today as well although the number of Cubans escaping by sea or through third countries has mushroomed since 1989.[/ref] The post-1980 arrival cohorts also differ in the aggregate in some important respects from the pre-1980 arrivals. They were less representative of the pre-revolutionary higher and middle socioeconomic classes, and had greater economic motivations to leave than earlier emigration cohorts.

The emigration decision of most pre-1980 refugees particularly during the early 1960s was largely determined by political and social class motivations (e.g., the adverse socioeconomic consequences of the early revolutionary redistributive policies), although, as time went on, other Cubans began to leave as much in response to economic as to political dissatisfaction. Over time, it became more difficult to separate economic from political motivations, since both are intertwined. The political refusal of the Cuban leadership to introduce meaningful and far reaching internal economic reforms has intensified the economic crisis engulfing Cuba since 1989. While many recent refugees continue to allude to a repressive political environment as a reason for departing, others decided to leave largely for economic reasons. These distinctions are important since it can be hypothesized that the propensity to remit is influenced by familial ties and obligations, as well as by political feelings towards the Castro regime.

Political emigration, and specifically post-1959 Cuban emigration, differs in some very crucial respects from the labor migrations generally discussed in the literature. The latter, by definition, does not entail opposition, or at least disagreement with policies pursued by the home country government, or preclude the return migration option, whether or not labor migrants decide to settle permanently in other countries. Labor migrants, further, are not generally institutionally proscribed from participating in their countries’ economic life (except in former socialist countries, such as Yugoslavia, where there were limitations regarding the private ownership of assets).

Cuban emigrants, of course, have not been allowed to return, and like Cubans on the island, are not permitted (except for some well-known instances) to hold private productive assets in Cuba. While the reasons leading many Cubans to leave the country may not be conceptually different from those motivating labor migrants from other countries (e.g., the search for better economic opportunities abroad), the emigration of most Cubans has been at least partly determined by their displeasure with, and opposition to, the economic policies and repressive measures instituted by the Castro government. For more than three decades, individual emigration decisions have been shaped by political and economic structural forces beyond the migrants’ control.

This is unquestionably true regarding the early 1960s emigration flows, but also applies with varying force to more recent flows (up to the present) given that most Cubans continue to flee an ideologically rigid and personalized political system that controls an institutionalized and highly inefficient economy. In its drive to revamp the nation’s socioeconomic fabric after 1959, the Castro government imposed economic rules and structures inimical to economic efficiency that have grievously interfered with the normal conduct of economic activities and the country’s development. An unwillingness on the part of the government to profoundly alter these rules and structures, even under the current economic crisis -and despite some major and deepening economic policy changes -exacerbates an already tense social and political situation. One outcome is a rise in emigration pressures, as many Cubans find the national situation hopeless.

Thus, while many current features of Cuban emigration are comparable to those permeating international labor migrations from other countries, the politico-economic dimensions of Cuban emigration are such that it is more accurate to describe it as primarily driven by political, rather than economic factors. The genesis of the economic push lies in the totalitarian nature of the Cuban communist state.

II.   Demographic Dynamics of the Cuban-American Population

Aside from the political motivations associated with the emigration of many Cubans, there are many other factors that could influence the remittance behavior of the exile community under the more liberalized travel and remittance regime being implemented by the Castro regime since July 1993. These include, aside from the domestic steps taken by Havana, the demographic dynamics determining the growth and structure of the Cuban-origin community abroad and its demographic and socioeconomic characteristics. U.S. regulations regarding the amounts of foreign exchange that Cuban-Americans may remit and take to Cuba while visiting also play a role.

By reviewing these issues, it can be evaluated which segments of the Cuban-origin population abroad are likely to contribute dollars to the Cuban economy (via remittances and visits) and which are not, and to estimate a reasonable range of foreign exchange remittances. Of major significance are those variables indicative of strong attachment to the homeland (and to relatives left behind). Included among these are the length of time since the emigrant left Cuba, the extent to which immediate families have been reunited abroad, the percent of Cuban-origin population not born in Cuba, and foreign naturalization rates. We begin by examining the demographic dynamics that have shaped the current Cuban-origin community abroad. We then evaluate its socioeconomic and demographic characteristics.

A.  Cuba as a Country of Emigration

Since the 1959 revolutionary takeover, Cuba has become -in a major historical reversal -an emigration country. Close to ten percent of the Cuban population (including those native to Cuba and their foreign born descendants), or between 1.20 and 1.25 million Cubans, can be grossly estimated to have been living outside the country in 1990, mostly in the United States (1.053 million). The number of Cubans in other countries is unknown, but the most comprehensive statistical source providing data on persons of Cuban birth enumerated in other countries during the 1980 and earlier census rounds lists 53,962 such individuals residing in 17 Western Hemisphere countries, including Canada and 4 European countries (CELADE, 1989). Major concentrations of Cuban-born emigrants in countries other than the United States are reported for Spain, Venezuela, Jamaica, Costa Rica and Mexico. The data for some countries dates to the 1970 or earlier round of censuses. The census data, further, only provides information on the Cuban-born and not on their children born abroad.

Given these uncertainties, and allowing for: 1) unrecorded immigration; 2) the more recent arrival in other countries of growing numbers of Cuban migrants, particularly since the collapse of the Socialist bloc and the disintegration of the Soviet Union in 1989; and 3) the descendants of Cuban-born emigrants born abroad, it can be concluded that the number of Cuban-origin people (including the Cuban-born and their children born abroad) residing in countries other than the United States in 1990 ranged between 150,000 and 200,000. Roughly speaking, 12 to 15 percent of the Cuban-origin population abroad in 1990 resided in countries other than the United States. Since the United States welcomed the bulk of the post 1959 Cuban exile population, the discussion that follows focuses exclusively on the Cuban-American population.

B.  A Relatively Old and Slow Growing Population

Significant median age differences between various Cuban-American population sub-groups shown by 1990 U.S. census data are indicative of a demographic renewal process that is partly attributable to the relatively advanced age at which many Cuban refugees arrived in the United States. At 37 years (compared to 31.8 for the total U.S. population) in 1990 (see Table 1), Cuban-Americans have a high median age. Not surprisingly, the gap between the native Cuban-American median age and that of those born in Cuba is more than thirty years, 15.8 and 46.8 respectively. Equally significant is the median age differential between those Cuban-Americans arriving before 1980 (52.7 years) and those that did so between 1980 and 1990 (38.8 years).

The growth rate of the Cuban-American population has stagnated in recent years. The number of Cuban-Americans enumerated in the 1990 census closely approximates the estimated total number of Cubans entering the United States since 1959. According to a gross calculation, about a quarter of the one million Cubans reaching U.S. shores between 1959 and 1990 have died. Further, Cuban-American fertility is about ten percent below the national U.S. average. Ever married Cuban-American women in ages 35-44 in 1990 reported 1,881 children ever born, as compared to 2,100 for the total U.S. population. In contrast, the figure for all Hispanic-origin women was 2,733, or 30 percent higher than the national average. High mortality associated with an older than average population, low fertility and sharply reduced immigration rates since the early 1980s are responsible for the slow growth of the U.S. Cuban-origin population.

This demographic stagnation and the continued aging of the Cuban-American population have important implications for remittance flows to Cuba. The same applies to the likelihood of return trips to visit relatives. Logic suggests that attachment to the homeland, and to relatives left behind, is considerably stronger among the Cuban-born than among native Americans of Cuban stock. As the relative weight of the former declines and that of the latter increases, the share of those likely to send remittances declines. In 1990, 28 percent of the Cuban-American population was native born, with this share likely to continue expanding rapidly given the high median age of the Cuban-born.

C.  Family Reunification in the Absence of Return Migration

An associated phenomenon is that Cuban emigration has not generated a return migration flow of any consequence. Most migratory flows are accompanied by a substantial return flow, with a considerable percentage of the emigrants, with as many as 30 to 50 percent, eventually resettling in the home country. In Cuba, this has not occurred for several reasons. The vast majority of those leaving Cuba did so permanently, with only an inconsequential number returning to Cuba. This emigration feature, typical of refugee outflows, has been magnified in Cuba by the pervasiveness of the socioeconomic and political changes accompanying the revolutionary process, and by more than three decades of socialist rule. Most Cubans emigres are bitterly opposed to the Castro government, have been in no mood to compromise with it, and will not consider the return option as long as the socialist regime remains in power. The Cuban government, for its part, has reinforced these attitudes by its political intransigence, travel restrictions, and refusal to grant those few Cubans contemplating a return the right to do so. Even for the most recent emigrants, the no permanent return policy remains in effect, while the desire of individual migrants to resettle in Cuba is undermined by an unchanged political environment and a severe economic crisis.

One adaptation to political polarization and the absence of a return option is that over time Cuban families have shown a strong predisposition to be reunited abroad. This is not, of course, a unique feature of Cuban emigration, since family ties are a major force driving the perpetuation of migratory flows worldwide. With the return and temporary labor migration options precluded, continued emigration has been the only option left for Cuban families to remain together.

Social class selectivity has magnified the strong family reunification effects of Cuban emigration. Over time, many of the early refugees drawn from the upper and middle classes managed to reunite their immediate families in the United States and other countries, many Cuban-American families not having any close relatives left in Cuba today. The polarizing nature of the Cuban revolution, and past revolutionary policies proscribing family contacts between Communist party militants and their relatives abroad, also led to deep chasms within Cuban families. It is not rare for close branches of Cuban families separated by exile not to have been in touch with each other for decades. Further, the U.S.-born children of Cuban exiles are not likely to have any sense of kinship with distant relatives they have never met.

In some instances, however, family connections remain strong, although the process of demographic renewal alluded to earlier continuously erodes these linkages. Exceptions are the most recent refugee arrivals, including the 1980 Mariel outflow and those leaving for the United States and other countries since then. As will be shown below, however, most of the post-1980 arrivals are at a relative disadvantage regarding their ability to send remittances to Cuba. In comparison to pre-1980 arrivals, they are less endowed with human capital, have lower per capita and per family incomes, and are more prone to live in poverty.

III.   Remittance Flow Implications

These social and political features of Cuban emigration, and the U.S.-imposed legal limits on the amount of remittances that can be sent and dollars spent while visiting Cuba, must be considered when analyzing the potential flow of remittances and their impact on the national economy. For many Cubans, it would be tantamount to treason to visit the island as long as Castro and his associates continue in power. Other Cubans opposed to the Castro regime regard sending dollars to Cuba as unacceptable since it would help the regime weather its most difficult economic crisis. Some Cuban exile organizations, in fact, have called for a ban on remittances by equating the sending of remittances with being unpatriotic.

For a considerable number of exiles with close relatives in Cuba, however, family solidarity is likely to take precedence over political considerations. These exiles are likely to send remittances to their close relatives, regardless of the consequences for regime survival. Meanwhile, the potential pool of emigrants likely to remit continues to shrink as the larger emigrant cohorts of the 1960s and 1970s age and die, although smaller contingents of new immigrants continue to arrive in the United States.

The absence of a return flow, together with the ongoing reunification of families abroad, is significant since studies conducted in many parts of the world suggests that migrants direct a considerable share of remittances to the home country for their own use upon their anticipated or actual return. International migrants send money home or bring it with them for a variety of purposes related to their own well-being or that of their families (for reviews see, Russell, 1986; Massey, et.al., 1987; Portes and Guarnizo, 1991). These include the purchase of land, housing, vehicles, and durable goods, and investment in small factories, shops and other small businesses. Current Cuban law proscribes most of these investments, including asset ownership by Cubans living abroad. Many of these remittances are intended for small business development purposes (for an overview and several Latin American case studies, see DiazBriquets and Weintraub, 1991), an option not available in the state-controlled Cuban economy where only self-employment is allowed.

Lacking plans for an eventual return, and with many Cuban families already reunited abroad, the potential volume of remittances is prone to be well below what could have been estimated on the basis of the exiles’ economic possibilities and the experience of other Latin American emigrant communities known to send substantial amount of dollars to the home countries. Lastly, the limits imposed by the United States on the amount of dollars that households can send to Cuba ($1,200 annually) or spend daily while visiting the country is certain to have a further dampening effect on the amount of remittances reaching Cuba. However, considerable cash dollar amounts are likely to be taken directly to Cuba by visitors, thus bypassing regular commercial channels and evading official controls established by both countries. While these dollars ultimately should end up in the coffers of the Cuban government, their more immediate impact will be to fuel black market transactions.

IV.   Demographic and Socioeconomic Predictors of Remittances

In general, and for most emigration flows, there are a number of variables that can theoretically be associated with the propensity to remit. The extent to which these variables act as effective predictors varies from situation to situation and, in many cases, can be determined empirically. In Cuba’s case, these predictors are mediated by the nature of the emigration flow, and in particular by the changing gradient between political and economic motivations discussed earlier. The following profile provides a succinct description of some of the salient characteristics of the Cuban-American community. [ref]The demographic and socioeconomic profile is based on data from the 1990 U.S. census shown in the appendix tables. For comparative purposes, the appendix tables also present data for selected U.S. subpopulations[/ref] These characteristics are useful when assessing the likely remittance behavior of the Cuban-American population.

A.  Place of Birth, Period of Entry and Naturalization Status

The 1990 U.S. census enumerated a Cuban-American population of 1.053 million, of which 754,716 (71.7 percent) were foreign born and 298,481 were native born (28.3 percent). Although many of the Cuban stock native Americans may still retain a nostalgic attachment to Cuba passed to them by their parents, most are only likely to have a tangential interest in Cuba and their relatives there. For most purposes, it can be assumed that the probability that second and third generation Cuban-Americans may send remittances to relatives in Cuba is low, probably approximating zero. Thus, U.S. born Cuban-Americans are excluded from the ensuing review of the characteristics of the Cuban-American population.

Close to three-quarters (74.1 percent) of foreign-born Cuban-Americans entered the United States before 1980, and the balance (25.9 percent) between 1980 and 1990 (see Appendix Table 1). The most recent arrivals, as noted earlier, have a lower median age and are assumed to have a higher propensity to remit than earlier arrivals, since they presumably have more immediate relatives left in Cuba.

Naturalization status is an important predictor of remittance behavior. Acquiring U.S. citizenship is indicative of a higher degree of integration into American society and a greater distancing from the homeland. Thus, U.S. citizenship is assumed to be symptomatic of a lesser propensity to send money to relatives in Cuba. Also inversely associated with remittance behavior is length of stay in the United States, a strong naturalization predictor. About half of foreign-born Cuban-Americans enumerated in the 1990 U.S. census have become U.S. citizens. Of the 379,864 naturalized Cuban-Americans, 92.5 percent entered the United States before 1980. Only 28,425 Cubans coming to the United States since 1980, or 14.6 percent of the total post-1980 arrivals, were naturalized by 1990. Among those arriving before 1980, the corresponding figure was 63 percent.

B.  Age Composition Differentials

The differentials in age composition indicated by the median ages by period of arrival reviewed above have important implications for remittance behavior. The two most critical, percentage of the population in the prime working ages (20 to 64) and percentage in the prime family formation years (20 to 34), suggest that those entering the United States before 1980 are likely to be in a better position to remit (by having a greater presence in the labor force or a smaller number of dependent children, respectively) than those arriving between 1980 and 1990. Of the 559,450 Cuban-Americans entering before 1980 in 1990, 74.7 percent were in the prime working ages, as compared to 72.2 percent of the 195,266 arriving between 1980 and 1990. With respect to population share in the prime childbearing ages, the higher figure corresponds to the 1980 to 1990 arrivals (25 percent), as compared to 21.7 percent for the pre1980 arrivals. Higher labor force participation rates and smaller family size suggests greater access to disposable income, a determinant of the ability to remit.

C.  Family Composition and Household Size

Differences in family composition and household size between the two sub-populations of foreign-born Cuban-Americans by period of entry are to be anticipated given the large gaps observed in median age and age structure. In this case, a lower median age (and the observed age distribution differences) is indicative of a higher share of families in the prime family formation years. Among families entering the United States between 1980 and 1990, the percentage of families with children under 18 and under 6 years of age at home in 1990 (Appendix Table 1) were considerably higher than among those entering before 1980. This differential is also observed with household size. Whereas 59 percent of the households entering between 1980 and 1990 had three members or more, for those entering before 1980 the corresponding figure was 49 percent. The amount of disposable income in larger families with children is prone to be less, ceteris paribus, than in smaller families with no children at home.

D.  Educational Attainment and Ability to Speak English

Forty-six percent of the Cuban-American population 25 years of age and over in 1990 had not completed a high school education (Appendix Table 2). About 20 percent were high school graduates, another 20 percent had attended some college, and 15 percent had completed a college education or higher. Those arriving in the United States before 1980 were considerably better educated than those arriving later.

Whereas 42 percent of the pre-1980 arrivals 25 years of age and over were not high school graduates, 63 percent of those arriving between 1980 and 1990 had not completed high school either. Although differences in percentage graduating from high school were not marked between the two groups, at higher educational levels the pre-1980 arrivals enjoyed considerably advantages. The percentage of pre1980 arrivals with some college education or higher was nearly double that for the cohort arriving between 1980 and 1990. Both groups compared unfavorably, however, with the educational attainment levels of the native Cuban-American population 25 years of age and over. Among the latter, the percentage not having completed high school was less than 20 percent, and close to 60 percent had attended some college or higher.

Equally significant are the differences between the two arrival cohorts in terms of English-speaking ability (Appendix Table 2). According to the 1990 census, 55.5 percent of foreign-born Cuban-Americans in ages 18 to 64 could not speak English “very well”. The percentage was much higher among those arriving between 1980 and 1990, with 80 percent indicating a poor command of the English language. The corresponding figure for earlier arrivals was 47 percent. Language competency, of course, is mediated by education and age of arrival in the United States. Among foreign-born Cuban-Americans 5 to 17 years of age arriving before 1980, for example, only 6 percent indicated that they could not speak English very well. Educational level and language competence is related to income level, and thus with the ability to remit. There is some international evidence suggesting, however, that education is inversely related with willingness to remit.

E.  Employment Status, Occupation and Industry

Approximately equal percentages of pre-1980 and 1980 to 1990 arrivals are in the labor force (Appendix Table 3). One important difference shown by the 1990 census data is that naturalized Cuban-Americans are far more likely to be in the labor force than those that have not become citizens, regardless of period of arrival. In 1989, however, unemployment rates were more than double for the 1980 to 1990 arrivals (11.3 percent) than for the before 1980 arrivals (5.4 percent). These differences remain regardless of naturalization status, although unemployment rates in general tend to be higher among non-citizens independent of entry period.

The occupational and industrial distributions of the foreign-born Cuban-American population 16 years of age and over reveal that the pre-1980 arrivals are considerably better-off than those arriving between 1980 and 1990. Percentage wise, three times as many pre-1980 arrivals are in managerial and professional specialty occupations as 1980 to 1990 arrivals (Appendix Table 3). This occupational advantage extends to technical, sales, and administrative support occupations, with 34 percent of the pre1980 arrivals holding these occupations, as opposed to 26 percent for the 1980 to 1990 arrivals. At the low end of the occupational hierarchy, the opposite can be found. A considerably higher percentage of the Cuban-Americans arriving between 1980 and 1990 are employed in manual occupations, as compared to the before 1980 entries.

A similar pattern can be found when the industrial composition of the foreign-born Cuban-American labor force is examined (Appendix Table 4): the most recent arrivals tend to be concentrated in the industrial sectors where average skill levels tend to be lower and more poorly compensated. Pre-1980 arrivals can be heavily found in communications and other public utilities, finance, insurance and real estate, professional and related services, and public administration. Construction, manufacturing, retail trade, business and repair services, and personal entertainment are the industries where the bulk of the 1980 to 1990 arrivals are employed. The share of government employment is also much higher among pre-1980 arrivals.

F.  Income Level, Type of Income, and Poverty Prevalence

The demographic and socioeconomic characteristics discussed above lead, not surprisingly, to substantial family and per capita income level differentials between the two period of entry cohorts (Appendix Table 5). Median family income for families headed by a foreign-born Cuban-American in 1989 was $31,870, as compared to $35,225 for all American families. Median income, however, was considerably lower for families headed by a 1980 to 1990 arrival ($21,622) in contrast to a pre-1980 arrival ($32,182). Whereas more than 40 percent of families headed by a pre-1980 arrival had annual incomes exceeding $35,000, only about 20 percent of families headed by a 1980 to 1990 arrival did.

Even better indicators of income disparity by arrival cohort are per capita income figures. Whereas per capita income for all foreign-born Cuban-Americans was $16,261 (in contrast to $14,420 for the total

U.S. population), for 1980 to 1990 arrivals it was less than half ($8,789) the amount for pre-1980 arrivals ($18,868). These per capita income differentials are determined not only by educational and occupational characteristics, but also by variations in age structure, and family composition and size. For instance, families headed by a 1980 to 1990 arrival tend to be younger and are more likely to have children living at home.

Four out of every five Cuban-American households received income earnings. More households headed by 1980 to 1990 arrivals relied on earnings (84.1 percent) than pre-1980 entry headed households (80.4 percent), largely because the latter were older and had gained the right to pension income, including social security benefits. Three times as many pre-1980 entry headed households (29 percent) received social security as 1980 to 1990 arrival households (10 percent). About 16 percent of all foreign-born Cuban-American households received public assistance in 1989, 19 percent of 1980 to 1990 entry households and 16 percent of pre-1980 entry households. The high percentage of pre-1980 arrival households receiving public assistance responds to the advanced age of many of their members.

In 1989, 11.5 percent of all families headed by a foreign-born Cuban-American were living in poverty (see Appendix Table 6). The family poverty rate was three times as high (23.2 percent) among families headed by a 1980 to 1990 arrival, as among families headed by a pre-1980 arrival (8.6 percent).

G.   Housing Ownership and Rental Indicators

Additional indicators of the ability to remit relate to housing ownership and rental cost data. With respect to housing ownership, the pre-1980 entries show a considerable advantage. Of the 274,564 pre-1980 entry households, 57.2 percent were owner-occupied, as opposed to 28.4 percent for the 1980 to 1990 entries. A higher percentage of 1980 to 1990 arrivals had mortgage payments in excess of $700 per month (63.5 percent) than among the before 1980 arrivals (60.5 percent). Among renters, median rent differences ($458 for 1980 to 1990 entries, and $461 for before 1980 entries) were not significantly different. Whereas 15.4 percent of the owner-occupied houses for before 1980 arrivals were mortgage free, the corresponding figure for households entering between 1980 and 1990 was less than one (.8) percent. Pre-1980 arrivals have higher rates of housing ownership and mortgage free housing, and pay slightly less in mortgage payments than those entering after 1980. These data also suggests a lower ability to remit for those foreign-born Cuban-Americans entering the United States after 1980.

H.   Ability to Remit by Arrival Cohort: Summary

Many features of the popular image of Cuban-Americans as a relatively prosperous community are validated by the 1990 census data. But the census data unmistakably reveals as well that a wide socioeconomic gulf separates the pre-1980 arrival cohort from the cohort reaching the United States between 1980 and 1990. These cohort differences are partly a function of the shorter period of assimilation for the most recent arrivals, but also respond to the fact that many of the more recent arrivals (e.g., during the 1980 Mariel salified) were drawn from lower Cuban socioeconomic strata than earlier arrivals.

On the basis of the above analysis, it seems reasonable to conclude that remittances to Cuba from the Cuban-American community are likely to be well below what could have been anticipated on the basis of its socioeconomic characteristics. The ties with Cuba of many, if not most, of the Cuban-Americans with the greatest human capital stock and higher incomes have weakened over the years. With few immediate relatives left behind, and with their lives largely centered in the United States, the pre-1980 migrants have neither the interest nor the relatives to whom to send remittances. The exact opposite applies to the 1980 to 1990 entries. Although they would want to remit to their relatives in Cuba, their ability to do so is limited by lack of disposable income arising from lower earnings and heavier family obligations.

V.   Propensity to Remit: A Census Data-Based Estimating Approach

By combining census data and a number of assumptions, it is possible to derive an empirically-based range of annual remittance estimates. For these estimates, and for reasons noted above, it is assumed that Cuban-American households headed by pre-1980 arrivals are far less likely to remit than those arriving after 1980, and that the propensity to remit for households headed by a native born Cuban-American is zero. The next step in the estimation procedure is to assume what is the proportion of households by period of arrival likely to remit, and what proportion of their mean income these households are likely to remit. By changing and combining these parameters, an empirically-based range of estimates can be derived. The estimating equation is as follows:

R =Hp*Ip,

or, Ri =[(Hb*pw)(Ib*px)] + [(Ha*py)(Ia*pz)]

where, R = Estimated annual volume of remittances (in dollars)

H =Number of all foreign-born Cuban-American households in 1990 (346,510 households)

Hb =Number of foreign-born Cuban-American households arriving before 1980 (276,258 households)

Ha =Number of foreign-born Cuban-American households arriving between 1980 and 1990 (70,252 households)

I =Annual mean household income for all foreign-born Cuban-American households ($36,570)

Ib =Annual mean income for foreign-born Cuban-American households arriving before 1980 ($39,869)

Ia =Annual mean income for foreign-born Cuban-American households arriving between 1980 and 1990 ($23,593)

p =proportion, and

i =various remittance estimates according to assumptions regarding the proportion of households remitting and the proportion of annual mean income remitted, as given by w, x, y and z.

As shown in Table 1, R(a) is the maximum amount of remittances that Cuban-American households could possibly remit to their homeland. For this estimate to be valid, each and every one of the 346,510 Cuban-American households would have to send ten percent of their 1989 mean annual income, two rather unlikely assumptions. The remaining estimates are based on less extreme, increasingly realistic assumptions. R(c) assumes, for instance, that every Cuban-American household sends to relatives in Cuba the maximum allowable number of dollars under current U.S. law. R(g) relies on highly plausible differentiated assumptions by period of arrival, yet it only yields a remittance estimate of $154 million, far less than it is generally assumed. Although crude, these estimates suggest that the dollar windfall expected by Castro may not be as significant as has been assumed. In my opinion, remittances currently do not exceed $300 to $400 million annually.

The estimating procedure presented here could be further refined by considering the effects of other socioeconomic and demographic variables (shown in the appendix tables) on the volume of remittances. The estimates could be adjusted, for example, by considering the number of Cuban-American households living below the poverty line, or by assessing the effects of educational attainment on remittance behavior. The estimates can also be improved by taking into account the expenditures of Cuban-American visitors to the island, although these expenditures would have to be estimated net of monetary remittances. Of course, some of the other determinants of remittance behavior discussed in this paper would have to be considered to obtain dependable sets of estimates.

Table 1. Range of Remittance Estimates R(a) = H(1)*I(.1) = $1.267 billion

R(b) = H(1)*I(.05) = $ 634 million R(c) = H(1)*($1,200) = $ 416 million

R(d) = [Hb(.4)*Ib(.05)] + [Ha(.6)*Ia(.1)] = $ 320 million

R(e) = [Hb(.1)*Ib(.1)] + [Ha(1)*Ia(.1)] = $ 276 million

R(f) = [Hb(.4)*Ib(.05) + [Ha(.6)*Ia(.05) = $ 270 million

R(g) = [Hb(.1)*Ib(.05)] + [Ha(.6)*Ia(.1)] = $ 154 million

R(h) = [Hb(.05)*Ib(.05) + [Ha(.5)*Ia(.05)] = $ 69 million

With a change of government in Cuba, the amount of remittances flowing to Cuba is likely to increase. The resettlement possibility, a tolerant political climate, and a more congenial investment and private property environment would induce many Cuban-Americans to send money and to consider Cuba as a potential location where to invest their money. Even the continued aging of the Cuban-American community may contribute to an increase flow of foreign remittances under this scenario, since many elderly Cubans may find that their retirement and social security dollars will go a lot farther in a Cuba that is likely to be more affordable than even the most inexpensive retirement community in the United States.

Bibliography

Bureau of the Census (1993), U.S. Department of Commerce, Persons of Hispanic Origin in the United States, 1990 CP-3-3, Washington, D.C.

CELADE (1989), Boletín Demográfico, Centro Latinoamericano de Demografía, Santiago, Chile, 22:43.

Sergio Díaz-Briquets and Lisandro Pérez (1981), “Cuba: The Demography of Revolution,” Population Bulletin 36:1, April.

Sergio Díaz-Briquets and Sidney Weintraub, editors, (1991), Migration, Remittances and Small Business Development, Westview Press, Boulder.

Douglas Massey, Rafael Alarcón, Jorge Durand, and Humberto González (1987), Return to Aztlan, University of California Press, Berkeley.

Alejandro Portes and Luis Guarnizo (1991), “Tropical Capitalists: U.S.-Bound Immigration and Small Enterprise Development in the Dominican Republic”, in Sergio Díaz-Briquets and Sidney Weintraub, editors, Migration, Remittances and Small Business Development, Westview Press, Boulder, pp. 101131.

Sharon Stanton Russell (1986), “Remittances from International Migration: A Review in Perspective”, World Development 14:6, pp. 677-696.