SciELO - Scientific Electronic Library Online

 
vol.27 issue1Exile Communities and Their Differential Institutional Dynamics: A Comparative Analysis of the Chilean and Uruguayan Political Diasporas¿Garantías del ingreso para garantizar el trabajo? author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Revista de ciencia política (Santiago)

On-line version ISSN 0718-090X

Rev. cienc. polít. (Santiago) vol.27 no.1 Santiago  2007

http://dx.doi.org/10.4067/S0718-090X2007000200004 

 

REVISTA DE CIENCIA POLÍTICA / VOLUMEN 27 / N° 1 / 2007 / 67 - 87

ARTÍCULOS

 

Foreign Aid and Human Rights: The Latin American Experience

 

Joel A. Capellán*1, Simonpeter Gomez1

1 Buffalo State College (E-mail Joelcapellan@gmall.com) (E-mail: gomezs@bscmall.buffalostate.edu)


Resumen

Este estudio procura responder a la siguiente pregunta: ¿Cuál es la relación entre la calidad de los derechos humanos en los países latinoamericanos y la cantidad de ayuda financiera internacional que reciben? Examinamos el papel que los derechos humanos juegan en la distribución de ayuda financiera internacional de doce países durante los años 1983 al 2002. Usando una medida de los derechos humanos más válida, encontramos que el respeto a los derechos humanos no afecta la distribución de ayuda financiera internacional, con la excepción de los Estados Unido y hasta cierto punto Dinamarca, el Reino Unido y Francia. Estos resultados contradicen el compromiso que estos países han tomado de hacer los derechos humanos una parte importante del criterio usado en distribución de ayuda financiera internacional.

Abstract

This study attempts to answer the following question: What is the relationship between the human rights record of Latin American states and the amount of foreign aid they receive? Vie examine the role human rights play on the aid allocation for 12 donors over a 21 year period (1983-2002). Using a new and, we think, more valid measure of human rights, we find, in general, that recipient respect for human rights is not a major determinant of aid, with the exception of the United States and Germany and to a lesser extent Denmark, United Kingdom and France. These findings contradict the stated commitment these states have taken to make human rights an aspect of their foreign aid allocations.

PALABRAS CLAVE • Human rights • Foreign aid • Development assistance • Latin America • Organization for economic cooperation and Development


I. INTRODUCTION

The tireless efforts by both intergovernmental and nongovernmental organizations to end human rights violations have resulted in the verbal commitment of donor states to make human rights an aspect of the criteria used for foreign aid eligibility. This study examines the extent to which human rights impact foreign aid allocation to Latin America for twelve of the major donors from the Organization for Economic Cooperation and Development (OECD), Canada, Denmark, France, Germany, Italy, Japan, the Netherlands, Norway, Sweden, Switzerland, United Kingdom, and the United States. It is a test of how closely their rhetoric matches the reality. To do so, we employ a new, more valid, human rights scale developed by Cingranelli and Richards (1999), which measures government's respect for physical integrity rights.

Following the lead of previous empirical studies, we model aid allocation as a two stage process. In the first stage recipients are selected; in the second stage aid is distributed among the recipients. Using a Heckman two-stage model on a sample of 21 Latin American states, over the time period 1983 to 2002, we find that, despite the verbal commitment, human rights are only significant for four donors at the gatekeeping stage, Denmark, Germany, the United Kingdom, and the United States. While we find that Germany, the United Kingdom and the United States are more likely to give aid to Latin American countries with better human rights records, for Denmark, human rights has the opposite effect from what one would expect as states with poorer human rights records are more likely to get aid. In concurrence with previous research (Neumayer, 2003a), the so called "like-minded" states -Canada, Denmark, the Netherlands, Norway and Sweden- despite being viewed as selfless promoters of human rights, are found to be no different in their foreign aid allocation than other donors.

At the level stage, human rights are insignificant to all donors, except France and the United States. Both tend to give more aid to those Latin American countries with poorer human rights records. The results lead us to conclude, that despite the verbal commitment, donors states, in general, do not reward good human rights records in their foreign aid allocation.

Many donors claim that human rights play an important role in their foreign aid allocation (Neumayer, 2003a). This study assesses the impact, if any, respect for human rights has on the foreign aid distribution to Latin American states. This study expands and improves on existing literature in four major ways. First, our study looks, for the first time in the literature, at foreign aid allocation from the major OECD donors to Latin America. Second, it is a cross-national study examining the aid programs of twelve donors, which is uncommon on this particular topic; with most literature placing an emphasis on US foreign aid allocation. Neumayer (2003a, 2003b), and Alesina and Dollar (2000) are the only two known exceptions. Third, we use the Cingranelli and Richards human rights measure (CIRI), which, evidence suggests is a more valid indicator of respect for physical integrity rights than the commonly used Political Terror Scale (PTS). Finally, we employ a more expansive time frame than any previous study of foreign aid, relying on data from 1983 to 2002.

II. LITERATURE REVIEW

Existing studies on the relationship between foreign aid and human rights have, for the most part, focused exclusively on the United States (Apodaca and Stohl, 1999; Cingranelli and Pasquarello, 1985; Meernik, et. al., 1998; Poe, 1992 and 1994; Schoultz, 1981; Watson and McCluskie, 1997; Webster, 1992). This is to be expected for a few reasons. First, in gross terms the United States aid program is the largest of all donors. Second, until very recently it was very difficult to get access the large amount of data that would be required to investigate the determinants of aid for more than a few donors. Overall, these studies provide a rather mixed picture, with most suggesting that human rights do play a role in foreign aid allocation.

Although some studies have taken a global approach in studying US foreign aid allocation, others have focused on the geographic area of Latin America (Cingranelli and Pasquarello, 1985; Schoultz, 1981; Turnan, et. al., 2001; Watson and McCluskie, 1997; Webster, 1992). Findings among these studies also vary; Cingranelli and Pasquarello (1985) find, for example, "an increased importance for human rights considerations in U.S. bilateral aid decision (439)". They find human rights to be significant at the level stage and insignificant at the gate-keeping stage. Cingranelli and Pasquarello's findings were later challenged by two follow-up studies (Carleton and Stohl, 1987; McCormick and Mitchell, 1988) they argued "...the results were due to deficiencies in the researchers' design" (Poe, 1994; 495). Their results were significantly affected by the exclusion of El Salvador, which was considered an outlier.

Despite some methodological concerns, the "mixed picture" still remains. Schoultz (1981), for example, indicates" ...that aid has tended to flow disproportionately to Latin American governments which torture their citizens" (155). On the other hand, Watson and McCluskie (1997) finds no relation between human rights and foreign aid, indicating "...that human rights may not have factored into such aid allocations one way or another" (252). Poe (1994), however, finds human rights to be significant at the gate-keeping stage but insignificant at the level stage. Reconciliation between these findings is unlikely.

Differences in methodology and indicators may account for the disparity among these findings. Take human rights, for instance, the Freedom house rating of political freedom, regardless of being an inadequate measure of human rights, is commonly used for such purpose. Neumayer (2003a) uses the Purdue Political Terror Scale (PTS), which is an average of content coding of the annual Amnesty International human rights report and the US department of state's country report on human practices. Cingranelli and Richards (1999) developed a new promising scale of human rights, called CIRI. This scale "...expands on PTS by providing a yearly evaluation of countries' protection of a range of different human rights including what they also term 'Physical Integrity Rights,' namely the rights to freedom from extrajudicial killings, disappearance, torture, and political imprisonment...CIRI also contains an index of five important civil liberties" (Braham, 2004; 19-20). For these reasons, CIRI will be preferred in this study.

How one measures foreign aid may also affect its relation to human rights. While a majority of studies have compiled different types of aid -military aid, loans, grants, medicines- into one "net aid" variable, other studies (Poe and Meernik, 1995) have only looked at military aid. Their results "support the conclusions of Cingranelli and Pasquarello... that once decision makers decide that a country is deserving of aid, they do not let the levels of human rights abuse interfere with other criteria" (Poe and Meernik, 1995; 406). There is also variation in methodology concerning the use of statistical assessment models. The two stage model and the Heckman's two step estimator are often used for these kinds of analysis. However, the two stage method assumes that decisions at the gate-keeping stage are taken independently from the decisions at the level stage- in other words it does not account for bias. Heckman's two step estimator fixes this problem, providing validity to the study.

Cross-national studies are not immune to these deficiencies in methodology, thus yielding the same mixed findings. There are a relatively small number of cross-national studies. Some suggest, however, that the U.S. is not unique in terms to its foreign aid allocation. Neumayer (2003a) finds that personal integrity rights, which are at the core of human rights, are significant only at the Gate keeping stage- when states decide who receives aid. Yet, personal integrity rights are insignificant at the level stage. In other words, once a state has chosen the aid recipients, donors make no attempt to give greater rewards to those with better human rights records.

Alesina and Dollar (2000) find similar results, suggesting "that the direction of foreign aid is dictated as much by political and strategic considerations, as by economic needs and policy performance of the recipients" (33). According to Alesina and Dollar's study, countries that are relatively open, relatively democratic, and having more colonial past should receive greater amounts of aid than the ones that are not. Egypt and Israel should receive very little aid according to this criterion; however, Egypt receives 400% and Israel 450% more than the countries with similar characteristics. This suggests that political motives do, in fact, play a role in aid distribution.

Trumbull and Wall (1994) find political and civil rights to be a positively significant factor in determining aid flows. However, they do not disregard the significance of other interest variables. Zanger (2000) conducts an empirical study of the relationship between recipient human rights and aid allocation for three OECD states; Germany, France, and the United Kingdom. However, she fails to account the fact that aid allocations are made in two stages, the gate-keeping stage and the level stage. Thus, the results of her study are to be considered with reservations.

The two studies mentioned above, Alesina and Dollar (2000) and Neumayer (2003a), are global studies, which look at all recipients in the world. One could argue that in quantitative research, a larger sample translates into greater general knowledge, but a loss of specifics. Our study tries to retain specific data to achieve a greater understanding of foreign aid allocation to Latin America; which is still haunted by a history of gross human rights violations. To the best of our knowledge there is not a single cross-national study that focuses on the geographic area of Latin America. We intend to use the best current human rights measure, the most recent data, in addition to the most appropriate statistical assessment model to fill this gap in the literature.

III. RESEARCH DESIGN

1. The Dependent Variable

The dependent variable is net official development assistance (ODA) that a country receives from a particular donor in a given year in millions of year 2000 constant US dollars, as reported in the Geographic Distribution of Financial Flows CD-ROM by the Organization for Economic Cooperation and Development for the years 1983 to 2002. Official development assistance is defined by the OECD as non-military aid that has a grant element of at least 25%. The use of net ODA instead of gross ODA is employed in order to make our results comparable to recent cross-national studies of donor aid (Alesina and Dollar, 2000; Neumayer, 2003a, 2003b; Zanger, 2000). Since we will be modeling aid allocation as a two-stage process, in the first stage, or gatekeeping stage, aid is measured as a dummy variable coded as 1 if the recipient got any aid from the donor in a given year, and 0 otherwise. In the second stage, the actual dollar amount, in millions of year 2000 US dollars will be used.

2. The Independent Variables

We employ a number of variables in our model of donor aid disbursements. The key independent variable is recipient respect for physical integrity rights, but we also employ a number of control variables, as well. Since much of the difference in the results of studies of the human rights-aid nexus may be due to differences in the measurement of human rights, it is imperative that we fully explain our human rights measure. In addition, the measure we employ in this study is relatively new, and may not be familiar to all researchers.

a. Measuring Government Respect for Physical Integrity Rights

Previous studies of human rights and aid have used one of two common measures of human rights. One measure is the Freedom House measure of political rights and civil liberties. This is a standards based measure that scores states on dual seven point ordinal scales based on the extent that they provide civil liberties and political rights. Researchers usually then average these two scores to create a single summary measure. The problem with using this indicator to measure the concept "human rights" is that this variable measures democracy, and not human rights. While previous research has shown that the two are correlated (e.g. Cingranelli and Hofferbert, 1997), the strength of the correlations varies among countries, and the two concepts, while related, are clearly distinct. So while most human rights researchers accept Freedom House as a good indicator or democracy, it is not widely accepted as a measure of human rights.

The other common measure of human rights is the political terror scale (PTS) compiled and published by researchers from Purdue University1. Unlike the Freedom House variable, PTS is designed to measure physical integrity rights2. This variable is a standards based measure that ordinally ranks countries on a five-point scale based on coded content analysis of the annual State Department Country Reports on Human Rights Practices and the annual Amnesty International Reports. Most researchers then average each of these scores to create a measure of physical integrity that summarizes the extent to which phenomena such as political imprisonment, torture, and extrajudicial killing occur within a state. The PTS measure is the most common way that human rights are measured in the literature (e.g. Apodaca and Stohl, 1999; Neumayer 2003a, 2003b; Poe, 1992; Poe and Meernik, 1995).

However, there are two problems with the PTS that, in our view, undermines its utility as measure of human rights. First, since the PTS is an average of two five-point ordinal codings (1, 2, 3, 4, 5), it is possible, and even likely, that a state may receive a score that is at the midpoint of the ordinal rankings, which is conceptually meaningless (Cingranelli and Richards, 1999).

Second, countries with the same overall PTS score may have very different human rights conditions, as one country, for example may routinely imprison people for their political views, but refrain from other types of physical integrity violations, while another may systematically torture, but refrain from other violations as well (Cingranelli and Richards, 1999). Therefore, using PTSto measure the concept of physical integrity rights means that we are assuming countries with similar scores will be treated similarly by donors, when they may in fact have different human rights environments, environments that the variable may not capture, but that donors may very well be able to see.

Third, based on its coding categories3, the PTS measures human rights conditions not just the human rights practices of governments. This is an important conceptual difference. Since we are interested with the extent to which government human rights practices influence bilateral aid decisions, it is important to have a variable that measures this strictly.

The variable we employ to measure the concept of physical integrity rights is the Cingranelli and Richards (1999) indicator4. This is an ordinal scale of government respect for four components of physical integrity; freedom from torture, freedom from extrajudicial killing, disappearances, and political imprisonment. These concepts are measured through content coding of the State Department Country Reports on Human Rights Practices and the Amnesty International Reports. Codings are based on a three point scale ranging from 0 to represent frequent violations of the right, 1 to represent some violations, and 2 to represent no violations (Cingranelli and Richards, 1999: 409). For all four components of physical integrity rights, "frequent violations" is defined as 50 or more violations of the right, "some violations" is defined as more than one but less than 50 violations, and "no violations" represents no confirmed violations of the right in question (Cingranelli and Richards, 1999: 409-410).

The resulting cumulative 9-point score is derived from Mokken scale analysis, a technique that allows Cingranelli and Richards (1999) to produce a unidimensional scale of the degree of government respect for the right to physical integrity. This means;

This new scale also improves upon the PTS, because a single score provides information not only on about the level of government respect for physical integrity rights, but also about the pattern and sequence of government respect for particular physical integrity rights. Here, pattern refers to the association of different levels of government respect for several physical integrity rights with a single, overall scale score. Sequence refers to the order in which governments have a propensity to violate particular physical integrity rights. The pattern of respect for physical integrity rights we discover using Mokken Scale analysis differs from the a priori pattern asserted by the PTS. The PTS does not explicitly address the sequence of government respect for particular physical integrity rights (408).

Therefore, each of the nine points in the Cingranelli and Richards' score are representative of a unique pattern of respect for physical integrity rights. This pattern is displayed in Table 1, which is replicated from Cingranelli and Richards (1999:414). As one can see from the table, use of the Cingranelli and Richards' measure will allow for greater accumulation of knowledge since future studies will be able to build upon our findings by examining exactly which physical integrity rights are more or less likely to influence a particular donor's aid allocation decisions.


We also employ a variable that captures the change in a state's human rights practices over time. This variable is constructed by subtracting a state's physical integrity rights score for year t from its physical integrity rights score for year t+1. This will allow us to determine if donors use development assistance as a carrot or a stick.

b. Control Variables

To control for the effects of factors other than recipient human rights practices, we employ several independent variables which have been shown to impact aid disbursements by previous research. These control variables are recipient level of democracy, recipient need, recipient size, donor economic interests, and donor strategic interests.

OECD official statements have linked development assistance to the level of democracy in the recipient state. Previous research has also found an empirical connection (Neumayer, 2003a; Neumayer, 2003b; Trumbull and Wall, 1994). In keeping with the previous research we operationalize democracy as the score a state receives on the Freedom House composite rating on civil and political rights5. We would expect, given the official policy of the OECD, that a state' s level of democracy to be positively related to both the probability of getting development assistance, and the amount of assistance it receives.

Drawing upon previous studies (McKinlay and Little, 1979; Neumayer, 2003a, 2003b; Wittkopf, 1972), we operationalize recipient need in terms of GDP in purchasing power parity (PPP) in constant year 2000 US dollars. The source for these data is the World Bank World Development Indicators on CD-ROM (2005).

Recipient size is measured in terms of population in millions of people. The source for this variable is also the World Bank World Development Indicators on CD-ROM (2005). Donor economic interests are operationalized as the total volume of trade between the donor and the recipient. This variable is measured in millions of constant year 2000 US dollars. The source for these data is the Direction of Trade Statistics Yearbook on CD-ROM (2005), published by the International Monetary Fund.

Donor strategic or national interests are a much trickier concept to operationalize and measure. Previous cross-national donor studies have used a variety of measures with varying degrees of success. Some indicators researchers have used include distance from the donor to the recipients' capitol (Neumayer, 2003a, 2003b), recipient military spending (Schraeder, Hook and Taylor, 1998; Zanger, 2000), size of recipients' military (Schraeder, Hook and Taylor, 1998), number of years since colonial independence (Alesina and Dollar, 2000; Neumayer, 2003a, 2003b), if the recipient is a former colony of the donor (Gomez, forthcoming; Zanger, 2000), and military aid from the United States (Neumayer, 2003a, 2003b). Of these, only the former colony and military aid variables have shown any consistent promise. Although Neumayer (2003a, 2003b) presents little in the way of theoretical justification for why military aid from the United States captures the strategic interests of non U.S. donors. So while it may be significantly related to ODA, it is unclear what it is actually measuring.

The single donor studies that exist have also encountered difficulties in modeling donor strategic or national interests. For the United States, whether or not the recipient borders a communist regime has performed well in some studies that examined US disbursements during the Cold War (Poe, 1992; Poe and Meernik, 1995; Poe and Sirirangsi, 1994; Wittkopf, 1972). In addition, Webster (2000), in a study of Canadian development assistance, finds success with a dummy variable that measures if the recipient is a member of the British Commonwealth, and another that measures if the recipient is a member of la Francophonie.

Building upon and extending the literature, we model donor strategic interests in a number of ways. First, we employ a dummy variable that is coded 1 if the recipient is a former colony of the donor. Second, to model general interests, we also include a variable that measures the distance between the donor and recipients' capitol cities6. The idea being that states that are geographically closer to the donor will receive more aid, ceterus paribus. Fourth, we include some variables that are designed to capture particular national interests of Canada and the United States.

In order to measure U.S. strategic interests relating to the global "war on drugs", we hypothesize that the United States will give more aid to countries that produce large amounts of cocaine. Therefore, we include a dummy variable for Bolivia, Colombia, and Peru, the three states that produce most of the cocaine that is sold in the world.

Building upon Webster (2000), we also employ a dummy variable for Canadian special interests. This variable is coded as 1 if the recipient is a member of the British Commonwealth, and 0 otherwise.

3.  The Sample

We apply our model of the relationship between human rights and aid on twelve donor states, all members of the Development Assistance Committee (DAC) of the OECD. The donors were chosen due to either the size or their aid program (France, Germany, Japan, the United Kingdom, and the United States) or their stated commitment to the developing world (Canada, the Netherlands, Norway and Sweden), or their inclusion in previous cross-national studies (Denmark and Italy). Our sample of recipients includes 21 Latin American countries from the time period 1983 to 2002. The full list of recipients is included in Appendix A. Three countries that are geographically located in Latin America, but are not generally considered to be Latin American in terms of their culture and history, Belize, Guiana, and Suriname, are excluded from the sample.

The unit of analysis for the study is the country-year; which means that there is a unique observation of the dependent variable for each donor-recipient combination for each of the 20 years examined. The resulting data are pooled cross-sections, with a total sample size of 420 for each donor. The actual number of observations will vary from donor to donor based on available data, and the results of the so-called gatekeeping stage, discussed below.

Each of the independent variables is lagged two years on the dependent variable in order to mirror the information available to decisionmakers when determining aid allocations (Cingranell and Pasquarello, 1985; Poe, 1992; Poe and Meernik, 1995; Neumayer 2003a, 2003b). This means we are assuming that the most up-to-date that a government official has when making aid decisions in, say, 1990, is information on Latin American countries from 1988.

4.  Methodology

Most of the previous research has modeled the aid programs of DAC members as a two-stage process (Cingranelli and Pasquarello, 1985; Gomez, 2004; Meernik, et. al., 1998; Neumayer, 2003a, 2003b; Poe and Sirirangsi, 1994; and Webster, 2000). The first stage, called the gatekeeping stage, is when donors determine which countries will be eligible for aid. The second stage, often called the level stage, is where donors determine how much aid the recipients who have survived the first stage should receive. However, some cross-national donor studies have modeled it as a single stage (Alesina and Dollar, 2000; Schraeder, et. al., 1998; and Zanger, 2000). We model aid as a two-stage process since the findings of recent non-US donor studies have shown it to be an accurate model (Neumayer, 2003a, 2003b; Webster, 2000), and so that we may compare our findings to theirs.

There are two ways to model the two-stage aid allocation process. The first, called a two-part model, employs a probit or logit regression to estimate the gatekeeping stage. Then, an OLS regression is utilized to estimate the level stage. However, the problem with the two-part model is that it ignores that the sample of recipients in the level stage are dependent upon the results of the gatekeeping stage. The second stage is therefore a biased sample. Failure to take this into account can lead to misestimation of the effect of human rights in the second stage.

The second way of modeling the two-stage aid process, one that takes into account the dependent sample is with a two-stage Heckman procedure (1976). This method employs a probit analysis in the first stage, then calculates a hazard rate, which is essentially a term to correct for bias in the level stage sample, and then uses this term as an independent variable in the level stage, which is an OLS regression.

IV. RESULTS

The Heckman two-stage model was used to estimate the impact of recipient human rights on aid decisions, and aid levels from twelve donors; Canada, Denmark, France, Germany, Italy, Japan, the Netherlands, Norway, Sweden, Switzerland, the United Kingdom and the United States. Table 2 displays the results of the gatekeeping stage, which is a probit analysis. The results of the level stage, an OLS regression analysis with an error term that corrects for bias in the sample, are presented in Table 3. We will now discuss the results of each analysis in turn.



1. Gatekeeping Stage

The results of the gatekeeping stage are presented in Table 2. The reported coefficients are probit coefficients. In this analysis, the dependent variable is dichotomous, equaling 1 if the country received any aid and 0 if it did not. Human rights are significant for four of the twelve donors, Denmark, Germany, the United Kingdom and the United States. For Denmark the probit results seem to indicate that countries with poorer human rights records are more likely to receive aid. In contrast, countries with better records at protecting physical integrity rights are more likely to receive aid from Germany, the United Kingdom and the United States.

If we convert the probit coefficients to the change in the probability of the dependent variable occurring with a change in the value of the independent variable, while holding the other variables constant at their mean values, then for Denmark a one unit increase in a state's physical integrity rights score, results in a reduction in the probability of getting aid of 4.5%. So, if a Latin American state were to increase its physical integrity rights score from zero to eight, this would result in a 36% reduction in the probability of receiving aid. Since there are no previous studies on the allocation of aid from non-US donors to Latin America, there are no other findings to serve as a basis for direct comparison. However, our results do contradict what Neumayer (2003a) finds for Denmark in a global sample of recipients. In his analysis, countries with better human rights records are more likely to get aid from Denmark.

So it appears that our decision to focus on the Latin American region was worthwhile, as it shows that Denmark treats this region differently than it does others. Although, it must be pointed out that Neumayer (2003a) uses PTS as his indicator of physical integrity rights, and only covers the years from 1985 to 1997, either of which, or both factors possibly could have an impact on the difference in the results.

In the case of the U.S. aid program, Latin American states that have better records of protecting physical integrity rights are more likely to receive any aid. Specifically, a one-unit increase in a country's human rights score increases its probability of getting aid by 5.5%. These results mirror those of Neumayer's (2003a) global study. While our analysis covers the time period from 1983 to 2002, they also support Cingranelli and Pasquarello's 1985 findings of the US aid program to Latin America. Our results are also in keeping with Meernik, Krueger, and Poe's (1998) findings for the United States in a global analysis from 1977 to 1990. But it contradicts their finding for the U.S. for the 1991 to 1994 time period, where they find that human rights is negatively related to the probability of getting US aid. However, the difference between our results and previous studies may be real, or they may be due, in whole or in part, to differences in the time span and/or the manner in which human rights is measured.

The United Kingdom is also more likely to give aid to Latin American states with better human rights records. If a state increases its respect for physical integrity rights by one unit, then its chance of receiving aid from the United Kingdom increases by 1%. In comparison, Neumayer (2003a) found that human rights were not significantly related to the probability of getting aid from the UK.

In the German case, while the coefficient is significant its impact on the probability of getting any is infinitesimal (7.04 to e35).

Why is it that human rights are not a significant factor for most donors at the gatekeeping stage? We can posit two explanations. First, since "human rights" is a multidimensional concept, no single measure may work adequately for every donor. Although repeating the analysis with the political terror scale instead of physical integrity rights did not improve upon the results. Perhaps, it may be necessary to conduct a closer examination of what exactly decision makers in each donor state mean by the phrase "human rights", through interviews, much like what Cingranelli and Pasquarello did in their 1985 study of U.S. aid.

The other explanation is much more straightforward: the major OECD donorsjust do not back up their words with their deeds. If this is indeed the case, and more study should be done before anyone can reach this conclusion definitively, then it is incumbent upon all of us to see to it that the donors live up to their own pledges.

The change in a country's respect for physical integrity rights is only significant for Germany and the United States, which means that we cannot conclude that the relationship between this variable aid the probability of receiving aid is real and did not occur by chance alone for the other ten donors. This indicates that most of our OECD donors do not take into account trends in human rights when deciding whether or not give aid. Germany and the United States are the lone exceptions. They are both more likely to give aid to those states that exhibit a decline in human rights from one year to the next. They are not rewarding states who improve their human rights, but doing the opposite, rewarding states that show a reduction. This result is strange given that the human rights variable is significant and positive for both donors, as discussed above. In other words, Germany and the U.S. are more likely to give aid to states with better human rights records, but are also more likely to give aid to states whose human rights records decline over time.

These seemingly contradictory results may be an indicator that human rights play a very complicated role in U.S. relationships with Latin America. There are at least three possible explanations for this pair of findings. First, the U.S. is using aid as an inducement to get states to improve their human rights records. Second, the two results may be evidence of institutional conflict between Congress and the President over human rights. Historically, it has been Congress that has taken the lead in human rights conditionality (Cingranelli and Pasquarello, 1985). This accounts for the positive relationship between the human rights level and whether or not a state gets aid. But in particular cases, such as with El Salvador in the 1908s and Colombia in the 1990s, the President is able to override this Congressional push and has been able grant aid in the face of short-term changes in human rights practices. A third possibility, and one that deserves future attention, is that it may be the magnitude of change that drives the U.S. decision to grant aid. That is, a one or two point change in a country's respect for physical integrity rights may not be large enough to catch the attention of lawmakers in Washington, but larger changes may alert the lawmakers to use aid as an inducement to improve a state's practices.

In the German case, it is possible that the pair of findings is capturing the influence of political party politics more generally. Previous research has shown that party platforms of even those parties that are not in power can influence German government spending, it is possible that these two findings represent the contradictory human rights of competing parties (Hofferbert and Klingemann, 1990; Klingemann, Hofferbert and Budge, 1994). Future research should look within the donor states to examine this possible explanation.

Turning to the control variables, our results seem to indicate that the OECD donors do not generally follow their own rhetoric regarding democracy promotion. For six donors there is no relationship between a state's level of democracy and the probability that it receives aid. For three donors, Denmark, Norway and Sweden, states with lower levels of democracy are more likely to receive aid. This result may be capturing the "like-minded" donors long-standing tendency to give aid to socialist and communist states, states that score poorly on the democracy measure.

Only Germany, Japan and the United States are more likely to give aid to states that are more democratic. This indicates that at the gatekeeping stage, the Japanese government stays true to its own pledges to support democracy in aid decisions (Ministry of Foreign Affairs, 1991 and 1993). The same can be said for the United States at the gatekeeping stage. When it comes to democracy, the United States practices what is preaches. The German government has made a strong commitment to human rights and good governance since its re-establishment in 1949, so for them this result is not surprising (Baehr, 1994).

Recipient need performs as expected. Poorer Latin American states are more likely to get aid from seven donors, Denmark, Italy, Japan, Norway, Sweden, the United Kingdom and the United States. This result supports a liberal interpretation of ODA, that donors do have altruistic motives. The variable is not significantly related to the probability of receiving aid for the other donors. In terms of recipient size, measured in terms of population, larger Latin American countries have a higher probability of getting aid from Germany, Italy, Japan, Norway, Sweden, and the United Kingdom. For France and the United States, on the other hand, larger states are less likely to receive development assistance. For the United States this variable may be capturing the U.S. tendency, especially during the eighties, to favor Central American states.

Surprisingly, only two donors for whom trade is significant are more likely to give aid to Latin American countries they trade more with, France and the United States. It appears for these two donors that development assistance may serve some economic self-interest. Of the other two donors with a significant trade variable -Germany, and Sweden- Latin American states with whom they trade less are more likely to be eligible for aid. Thus, our results seem to indicate that economic self-interest is not a motivation at the gatekeeping stage for these donors. Volume of trade is not significantly related to the probability of getting aid for the other donors in the study.

Turning to the political self-interest variables, distance is positively related to the chances of getting aid for Denmark, Germany, Italy, Japan, Norway, Sweden and Switzerland. In other words, Latin American states that are further away from each of these donors are more likely to receive development assistance. This means that even after we control for population size, South American countries are more likely to make it through the gatekeeping stage of these donors than Central American or Caribbean states. Distance from the donor is not significant for the other OECD members in the study. While the distance between capitols may be relatively similar since we are focused only on the Latin American region, it is still a continuous measure that captures real differences, and also serves to capture any geographic differences between Central and South America more finely than a dummy variable for region.

The two variables we employ to measure US political interests, whether or not the recipient is a major cocaine producer (operationalized as Bolivia, Colombia and Peru) and whether or the state shares a border with a communist country, both perform as expected. When converted to a change in probabilities, a major cocaine producer is 22% more likely to receive aid from the United States than other countries. In addition, all of the states that border a communist country received aid from the United States, in every year of the study.

France gave aid to its former colonies located in Latin America in every year under study. Based on our analysis, the United Kingdom does not appear to give more aid to its former colonies, given that the variable in not significant. Similarly, being a member of the British Commonwealth is not significantly related to the probability of getting aid from Canada. This result is similar to what Webster (2000) found at the gatekeeping stage in a global of study of Canadian assistance in 1997.

In general a few patterns emerge from the analysis. First, human rights are not a major factor for most donors at the gatekeeping stage. Second, poorer countries are more likely to be included in the pool of aid recipients across most donors. Third, South American states are more likely to get aid from most donors than other states in the region. The so-called "like-minded" countries, Canada, Denmark, Norway and Sweden, do not display any consistency in terms of the determinants of aid eligibility. None of them are more likely to give aid to states with good human rights records. Although three of the four are more likely to give aid to poorer states, as one would expect. Interestingly, Norway and Sweden seem to have nearly identical aid programs at the gatekeeping stage. For both, GDP, population, and distance all significantly impact the probability of receiving development assistance in the same manner. The trade variable also behaves in the same way for both donors, although for Norway it just misses the conventional 0.05 cut off for statistical significance.

2. The Level Stage

The results of the level stage are presented in Table 3. At this stage, the reported coefficients are the results of an ordinary least squares (OLS) regression. This is a technique that isolates the impact of each independent variable on the dependent variable, while holding the other independent variables constant. At this stage, the results were obtained using the Heckman two-step procedure which is more appropriate than the Heckman selection technique because it is better to use with large datasets, as we have here, and when some of the assumptions of the Heckman selection technique do not hold, which is the case here since the Heckman selection model would not converge for many of the donors.

Recipient respect for physical integrity rights is significant but negat/Ve for both the United States and France. It seems that over the 1983-2002 time period, the United States tended to give more aid to those Latin American states with poorer human rights records. Based on our analysis, apparently the United States rewards states with better human rights records at the gatekeeping stage, but does the opposite at the level stage, rewarding states that give less respect to physical integrity rights. This finding supports Neumayer's (2003a) finding for the United States at the level stage, although as mentioned above, he employed a global sample of recipients.

France, it appears, also gave more aid to rights abusive states at the level stage. This contradicts Neumayer's (2003a) result as he found recipient human rights to be positively related to aid from France. Human rights are not significantly related to amounts of aid allocated for any of the other donors in the study. This contradicts Neumayer's (2003a) findings. Neumayer finds that human rights are significantly related to aid levels for Canada, Denmark, Japan, the Netherlands, and Norway.

The change in human rights has no significant impact on aid from the OECD donors at the level stage. It appears, generally, that donors do not consider yearly changes in recipient human rights as a factor in aid decisions. Perhaps a variable that measures yearly changes is measuring change too fi nely. It is possible that donors consider longer term changes, over a multiple year time period. Future investigations may try to model this possibility.

Recipient level of democracy is significant for only two donors, Germany and Switzerland. In both instances they give more aid to less democratic states, the opposite of what one would expect to find given the OECD's emphasis on promoting democracy. This is more evidence that the OECD members do not back up their own words with deeds.

Poorer states, measured in terms of GDP in purchasing power parity, get more aid from five donors; Germany, Japan, the Netherlands, Switzerland, and the United States. This result is somewhat surprising for Japan, since the literature tends to conclude that Japan's aid is based almost exclusively on selfish concerns and not recipient need (Hook, 1995; Schraeder, et. al., 1998). Perhaps Japan is moving toward a more altruistic stance in aid giving in recent years, something that earlier studies would have been unable to identify. Recipient need is not significantly related to aid for the other donors.

The volume of trade between the donor and the recipient is significantly related to aid allocations for only Italy, Switzerland and the United Kingdom. In all three cases, it appears that they give more aid to Latin American states with which they have more trade. Trade is not significant for any other donors. It is somewhat surprising to us that trade is not significant and negative for Japan, since it has traditionally been seen as using aid to promote bilateral trade. Based upon our study, this conclusion must be called into question. Again, we think what we are capturing is progress on behalf of the Japanese government to move away from selfish motivations toward more selfless ones in its aid policy. Future studies should try to identify if such a trend continues.

Recipient population conclusively affects the volume of aid for France, Germany, Japan and the United States. For the first three donors, more aid seems to go to the more populous countries of Latin America. The United States, it appears, provides less aid to the more populous states. To us, this is one indication that the United States tended to allocate more aid to Central American countries, particularly El Salvador in the 1980s, and Nicaragua in the 1990s, after the electoral defeat of the Sandinistas. France, Germany, the Netherlands and Switzerland all tend to allocate more aid to those states that are further away. Given that we are focusing on Latin America, this means, in real terms, that South American countries get more aid from these three donors at the level stage.

The special political interest variables all perform at the level stage as expected. The United States gives significantly more aid to the Latin American states that are major cocaine producers. This is as expected, since the United States is engaged in the so-called "War on Drugs", and part of that war is to provide economic assistance to promote alternate means of generating income in cocaine producing areas. The cocaine producers get about 53 million more dollars in aid than the other aid recipients. In addition, the United States gives significantly more aid to Latin American countries that border a communist state. According to our equation, those that border a communist state got about 109 million more dollars in economic aid from the United States than other countries in the region, during the years that the variable is included in the analysis.

The French political interest variable is also significant and in a positive direction. Former French colonies, it seems, receive about 18 million more dollars in aid than non-colonies. The United Kingdom gives significantly more aid to its former colonies of Latin America. Former British colonies get about 3 million dollars more in aid than other states. Canada allocates more aid to those Latin American countries that are fellow members of the British Commonwealth.

At the level stage a few patterns can be discerned. First, respect for physical integrity rights is not rewarded. Second, the one characteristic that is most consistently related to aid levels across the donors is recipient need, with poorer states getting more aid from five of the donors. Third, the indicators we have constructed to measure donor political interests are significant, for the most part. Overall, our results do reveal important aspects of the role of human rights in the aid allocations of OECD for an important region of the world, and for a more expansive time period than any previous study7.

It should be noted that we conducted a separate test to estimate the impact of the Cold War on the relationship between donor aid and human rights at both the gatekeeping and level stage. In both stages the results were not significantly different than those for the whole period, or for just the post-Cold War years (1991-2002).

V. CONCLUSIONS

Few studies examine the relationship between human rights and foreign aid allocations to Latin America from OECD donors. Our work is an attempt to fill this gap in the literature. In fact, to our knowledge this is the first study to comprehensively analyze the relationship between aid and human rights to Latin America. Our study also contributes to the field by employing, for the first time in such an analysis, a more valid and empirically rich measure of recipient respect for human rights the Cingranelli and Richards' measure. This variable captures government respect for physical integrity rights and each value in its range indicates a specific and unique pattern of government respect for four components of physical integrity, disappearances, extrajudicial killings, political imprisonment, and torture.

In addition, we have a very expansive dataset that covers the twenty-one year period from 1983 to 2002, a much longer time span than any previous study.

We find, in general, that recipient respect for physical integrity rights is not a major determinant of aid at either the gatekeeping or level stage. At the gatekeeping stage, only the United Kingdom and the United States reward states that exhibit more respect for physical integrity rights. One other donor, Denmark, is more likely to give aid to rights abusive Latin American states. For the other donors included in our analysis, recipient respect for human rights is not significantly related to the probability of receiving aid.

At the level stage, where donors decide how much aid to give to Latin American states that make it into the pool of recipients, respect for physical integrity rights is significant only for two donors, France and the United States. For both, it appears that more aid is given to states with poorer human rights records.

While it is difficult to place our findings in the proper context, since no previous studies have examined aid allocations to Latin America for non-US donors, we think it is enlightening to compare them to recent studies of US aid, and studies using a global sample for the other donors.

At the gatekeeping stage, our results contradict Neumayer's (2003a) findings for France, Denmark, Sweden and the United Kingdom. Denmark we find that recipient human rights is negatively related to the probability of getting aid, in direct contradiction to Neumayer's results for this donor. We find that recipient human rights is not significantly related to the probability of receiving aid for Sweden, while Neumayer finds that human rights is significant and negative. In addition, our results for Canada at the gatekeeping stage conflict with Webster's (2000) findings. Webster finds that recipient human rights are positively related to the probability of getting aid from Canada, while we find that the relationship is insignificant. These differences may be due to the increased time span that we employ, or to the way we specified the human rights variable.

In the case of the United States, our results, that better human rights records increases the probability of receiving aid, are consistent with Neumayer's (2003a) global study. They are also consistent with Meernik, Krueger and Poe's (1998) study of US aid during the Cold War period, and with Cingranelli and Pasquarello (1985) and Poe (1994).

At the level stage, we find that human rights are significant and negative for France and the United States. Neumayer (2003a) finds the opposite is the case for France. Although Neumayer (2003a) also finds that recipient human rights are significant for Canada, Denmark, Italy, Japan, the Netherlands and Norway. While we find no evidence that these donors take human rights into account at the level stage. Webster (2000) also finds that human rights are positively related to aid levels from Canada, again using a global sample.

There are two possible explanations as to why these OECD donors do not generally take into account a state's human rights record in aid decisions. First, the donors may be using human rights (and democracy promotion) as an excuse to reduce the overall size of their aid programs, an allegation made by many in the global south. The lack of an empirical connection, as we find here, may give greater weight to the argument.

Another explanation is that while the OECD has, in recent years, made human rights and democracy a central goal of development assistance, perhaps this goal has yet to be fully institutionalized in the member states. Our findings indicate that the OECD as an organization must make greater efforts to get members to follow their collectively agreed upon goals. Equally important, the citizens within the donor states must put pressure on these governments to abide by their own commitments. NGOs, should also seize upon these results to shame the donors to follow their rhetoric with action.

Sadly, in the face of the war on terror, which has allowed, or forced, many of these same donors to abandon fundamental human rights, we do not expect any positive change to be forthcoming. In fact we would expect any change that occurs to be in the direction of even less concern for human rights and democratic principles.

Since we examine the aid programs of twelve donors, it is not possible to closely examine the idiosyncratic factors that may impact aid allocations in any one donor state. These domestic factors, while discussed in depth for Germany and the United States, are important and merit future study. We suggest that future research focus on single donor studies for the OECD members other than the United States and, much like Cingranelli and Pasquarello (1985) did for the United States, explore the myriad of factors that may influence the aid allocation process in a given donor.

It appears that our findings for a sample of Latin American states are starkly different that previous studies. These differences underscore the importance of our study. Donors do treat Latin American states differently when it comes to the role that human rights play in aid allocations. This is an important contribution to the literature on the relationship between human rights and development assistance.


 

NOTES

The authors would like to thank The McNair Scholars Program for the financial support in acquiring the data; in particular we would like to thank Sandra Washington and Jennifer Johnson. We also thank Sheldon Tetewsky for his help with the data analysis, and Keith Henderson for his help with earlier versions of this paper.

1 The Political Terror Scale article as well as the data for the years 1980-2004 can be found at the following website: http://www.unca.edu/politicalscience/images/Colloquium/faculty-staff/gibney.html

2 Physical integrity rights are also commonly referred to in the literature as personal integrity rights (e.g. Neumayer, 2003a and 2003b) or life integrity rights (Fein, 1995).

3 The coding categories are "level 1: countries are under a secure rule of law, people are not imprisoned for their views, and torture is rare or exceptional...political murders are extraordinarily rare. Level 2: There is a limited amount of political imprisonment for nonviolent political activity. However, few are affected, torture and beatings are exceptional...Political murder is rare. Level 3: There is extensive political imprisonment, or a recent history of such imprisonment. Execution or other political murders and brutality may be common. Unlimited detention, with orwithout trial, for political views is accepted... Level 4: The practices of Level 3 are expanded to larger numbers. Murders, disappearances, and torture are common parts of life...In spite of its generality, on this level violence affects primarily those who interest themselves in politics or ideas. Level 5: The violence of Level 4 has been extended to the whole population...the leaders of these societies place no limits on the means or thoroughness with which they pursue personal or ideological goals (McCann and Gibney, 1996:19).

4 Cingranelli and Richards (CIRI) human rights dataset can be obtained in the following website: http://ciri.binghamton.edu/myciri/my_ciri_login.asp

5 http://www.freedomhouse.org/uploads/FIWrank7305.xls

6 These data can be found at http://www.wcrl.ars.usda.gov/cec/java/capitals.htm

7 We conducted a separate analysis that included a dummy variable that measured whether or not the recipient state was engaged in armed conflict. This variable did not result in significant results for any donor for either stage, therefore we discarded it from the final results.

 

REFERENCES

Alesina, Alberto and David Dollar. 2000. "Who Gives Foreign Aid to whom and Why?" Journal of Economic Growth 5 (1): 33-63.         [ Links ]

Apodaca.Clair and Michael Stohl. 1999. "United States Human Rights Policy and Foreign Assistance". International Studies Quarterly 43 (1):185-198.         [ Links ]

Blanton, Shannon. 2005. "Foreign Policy In Transition? Human Rights, Democracy, and U.S. Arms Exports". International Studies Quarterly 49 (4): 647-667.         [ Links ]

Braham, Eric. 2005. "Getting to Know the Bottom of the Truth: Examining Truth Commission Success and Impact". Paper presented at the International Studies Association Conference.         [ Links ]

Breuning, Marljke. 1995. "Words and Deeds: Foreign Assistance Rhetoric and Policy Behavior In the Netherlands, Belgium, and the United Kingdom". International Studies Quarterly 39 (2): 235-254.         [ Links ]

Brigham, Jeremy. 1999. "Human Rights and International Political Economy In Third World Nations: Multinational Corporations, Foreign Aid, and Repression". Social Science Quarterly 80 (3): 621-622.         [ Links ]

Carleton, David and Michael Stohl. 1985. "The Foreign Policy of Human Rights: rhetoric and Reality from Jimmy Carter to Ronald Reagan". Human Rights Quarterly 7 (1): 205-209.         [ Links ]

Cingranelli, David and Simon Peter Gomez. 1996 "Progressive Aid to Latin America?" In Foreign Aid Toward the Millenium, edited by Steven W. Hook. Boulder, CO: Lynne Relnner, 209-226.         [ Links ]

Cingranelli, David and Richard I. Hofferbert. 1997. "Democratic Institutions and Respect for Human Rights". In Human Rights and Developing Countries, edited by David Louis Cingranelli. Greenwich/London: JAI Press.         [ Links ]

Cingranelli, David and Thomas Pasquerello. 1985. "Human Rights and Practices and the Distribution of U.S. Foreign Aid to Latin American Countries". American Journal of Political Science 29: 539-563.         [ Links ]

Cingranelli David, and David Richards. 2005. The Cingranelli-Richards (CIRI) Human Rights Dataset.         [ Links ]

Cingranelli, David and David Richards. 1999. "Measuring the Level, Pattern, and Sequence of Government Respect for Physical Integrity Rights". International Studies Quarterly 43 (2): 407-417.         [ Links ]

Fein Helen. 1995. "More Murder In the Middle: Life- Integrity Violations and Democracy In the World, 1987". Human Rights Quarterly 17 (1): 170-191.         [ Links ]

Fraser, Donald. 1979. "Human Rights and U.S. Foreign Policy: Some Basic Questions Regarding Principle and Practice". International Studies Quarterly 23 (1): 174-185.         [ Links ]

Furuoka, Fumitaka. 2005. "Human Rights Conditionality and Aid Allocation: Case Study of Japanese Foreign Aid Policy". Perspectives on Global Development and Technology 4 (2): 125-146.         [ Links ]

Gomez, Simon Peter. forthcoming. "Human Rights and the Allocation of Foreign Aid: A Cross-National Analysis of the Cold War Years". The Social Science Journal.         [ Links ]

Heckman, James J. 1976. "The Common Structure of Statistical Models of Truncation, Sample Selection and Limited Dependent Variables and a Sample Estimator for Such Models". Annals of Economic and Social and Social Measurement 5: 475-92.         [ Links ]

Heckman, James J. 1979. "Sample Selection Bias as a Specification Error". Econometrica 47 (1) 153-162.         [ Links ]

Hofferbert, Richard I. and Hans-Dieter Klingemann. 1990. "The Policy Impact of Party Programs and Government Declaration In the Federal Republic of Germany". European Journal of Political Research 18: 277-304.         [ Links ]

Hook, Steven C. 1995. "Self-Interest and Foreign Economic Policy: A Cross-National Perspective". International Studies Notes 19(1): 26-36.         [ Links ]

International Monetary Fund. 2005. Direction of Trade Statistics CD-ROM. Wahslngton DC: International Monetary Fund.         [ Links ]

Klingemann, Hans-Dieter, Richard I. Hofferbert and Ian Budge. 1994. Parties, Policies and Democracy. Boulder, CO: Westvlew Press.         [ Links ]

Landman, Todd. 2004. "Measuring Human Rights: Principle, Practice, and Policy". Human Rights Quarterly 26 (4): 906-931.         [ Links ]

McCann, James A. and Mark Glbney. 1997. "An Overview of Political Terror In the Developing World, 1980-1991". In Human Rights and Developing Countries edited by David Louis Cingranelli. Greenwich/London: JAI Press.         [ Links ]

McCormick, James M. and Nell Mitchell. 1988. "Is U.S. Aid Really Linked to Human Rights In Latin America?" American Journal of Political Science 32 (1): 231-239.         [ Links ]

McKinlay, R. and R.D. Little. 1979. "The U.S. Aid Relationship: A Test of the Recipient Need and Donor Interest Models". Political Studies 27 (3): 569-587.         [ Links ]

Meernik James, Eric Krueger, and Steven Poe. 1998. "Testing Models of U.S. Foreign Policy: Foreign Aid during and after the Cold War". The Journal of Politics 60 (1): 63-85.         [ Links ]

Meyer, William H. 1999. "Confirming, Inflrmlng, and 'Falsifying' Theories of Human Rights: Reflections on Smith, Bolyard, and Ippollto Through the lens of Lakatos". Human Rights Quarterly 21 (1): 220-228.        [ Links ]

Neumayer, Eric. 2003a. "Do Human Rights Matter In Bilateral Aid Allocation? A Quantitative Analysis of 21 Donor Countries". Social Science Quarterly 84(3): 650-666.         [ Links ]

Neumayer, Eric. 2003b. "Is Respect for Human Rights Rewarded? An Analysis of Total Bilateral and Multilateral Aid Flows". Human Rights Quarterly 25 (2): 510-527.         [ Links ]

Nossal, Kim. 1988. "Mixed Motives Revisited: Canada's Interest In Development Assistance". Canadian Journal of Political Science/Revue Canadienne de Science Politique 21 (1): 35-56.         [ Links ]

Organization for Economic Cooperation and Development. 1992. Annual Report. Washington DC: Organization for Economic Cooperation and Development.         [ Links ]

Organization for Economic Cooperation and Development. 2005. Geographic Distribution of Financial Flows on CD-ROM. Washington DC: Organization for Economic Cooperation and Development.        [ Links ]

Poe, Steven C. 1991. "U.S. Economic Aid Allocation: the Quest for Cumulation". International Interactions 16 (4): 295-316.         [ Links ]

Poe, Steven C. and James Meernik. 1995. "U.S. Military Aid In the 1980s: A global Analysis". Journal of Peace Research 32 (4): 399-411.         [ Links ]

Poe, Steven C, Suzanne Pilatsovsky, Brian Miller Ayo Ogundele. 1994. "Human Rights and US Foreign Aid Revisited: The Latin American Region". Human Rights Quarterly 16 (3): 539-558.         [ Links ]

Poe, Steven C. and Rangsima Sirirangsi. 1994. "Human Rights and U.S. Economic Aid during the Reagan Years". Social Science Quarterly 75 (3): 494-509.         [ Links ]

Regan, Patrick. 1995. "U.S.Aid and Political Repression: And Empirical Evaluation of U.S. Foreign Policy". Political Research Quarterly 48:613-628.         [ Links ]

Schouitz, Lars. 1981. "U.S. Foreign Policy and Human Rights Violations In Latin America: A Comparative Analysis of Foreign Aid Distribution". Comparative Politics 13 (1): 149-170.         [ Links ]

Schraeder, Peter, Steven W. Hook and Bruce Taylor. 1998. "Clarifying The Foreign Puzzle; A Comparison of American, Japanese, French, and Swedish Aid Flows". World Politics 50 (2): 294-323.         [ Links ]

Svensson, Jakob. 1999. "Aid, Growth, and Democracy". Economics and Politics 11 (3): 250-297.         [ Links ]

Trumbull, William and Howard Wall. 1994. "Estimating Aid-Allocation Criteria with Panel Data". The Economic Journal 104 (425): 876-882.         [ Links ]

Tuman, John, and Craig Emmert. 2004. "The Political Economy of U.S. Foreign Direct Investment: A Reappraisal". Latin American Research Review 39 (3): 8-28.         [ Links ]

Watson, Robert and Sean McCluskie. 1997. "Human Rights Considerations and U.S. foreign policy: The Latin American Experience". Social Science Journal 34 (2): 249-258.         [ Links ]

Webster, Craig. 2000. "Canada's Human Rights Policy and Its Impact on Foreign Assistance Allocation". Peace Research 32 (4): 85-97.         [ Links ]

Webster, Gerald. 1992. "A Time-Series Analysis of Political Support, Strategic Location and the Geography of U.S. Foreign Aid to Latin America and the Caribbean, 1966-1987". Human Geography 74 (2): 125-132.         [ Links ]

Wittkopf, Eugene R. 1972. "Western Bilateral Allocation: A Comparative Study of Recipient State Attributes and Aid Received. Thousands Oaks, CA: Sage.         [ Links ]

World Bank. 2005. World Development Indicators on CD-ROM. Washington, DC: World Bank.         [ Links ]

Zanger A. Sabine. 2000. "Good Governance and European Aid: The Impact of Political Conditionality". European Union Politics 1(3): 293-317.        [ Links ]

 

Joel A. Capellán Is a senior majoring In Political Science and Criminal Justice at Buffalo State College, NY. He Is a McNalr Scholar. He has presented on Human Rights and Latin America at both regional and national Political Science Conferences. He plans to go to Graduate school to obtain a Ph. D. Originally from the Dominican Republic, he Is Interested In Latin American politics, democratization and human rights.

SimonPeter Gomez Is a lecturer In political science at Buffalo State College, NY. He received his Ph.D In 1999 from Blnghamton University In New York and his undergraduate degree In Spanish, History and Political Science from Brldgewater College In Virginia In 1989. He has also studied at la Universidad de Barcelona, Spain. He has published widely on development assistance policy, the quantitative study of human rights, the political attitudes of college students and professors In American colleges and universities, and the Impact of 9/11 and the so-called "war on terror" on teaching about human rights at the college level.

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License