The current issue and full text archive of this journal is available at www.emeraldinsight.com/1446-8956.

htm

IJDI 10,1

The interactive effect of remittances and civil liberties on investment and consumption
Jeffrey A. Edwards and Jennis J. Biser
Department of Economics and Finance, North Carolina A&T State University, Greensboro, North Carolina, USA
Abstract
Purpose The purpose of this paper is to investigate the level of inuence that civil liberties has on the marginal effect of remittances on gross domestic investment and consumption separately and measures it across all levels of civil liberties. Design/methodology/approach The authors employ a two-stage system generalized method of moments procedure and the civil liberties subset of the Freedom in the World Index as a proxy for civil liberty. Findings The ndings indicate a substitution effect from investment to consumption as civil liberties deteriorate for developing south economies, though not for emerging economies. In addition, the marginal effect of remittances on investment diminishes less quickly as economies become less free than it increases for consumption indicating that the substitution is not quite one-for-one. Practical implications Economies with low levels of civil liberties could benet by improving them in ways that would encourage recipients to channel remittances into investment rather than consumption. Originality/value This paper differs from previous research in that the authors evaluate investment and consumption separately rather than embedding these component parts within growth. In addition, when interactions are employed in existing literature, the inference drawn is static with regard to the varying degrees of institutional development. Third, none of the prior studies directly explores civil liberties proper; they usually aggregate civil liberties with other aspects of political or economic freedom. Keywords Investments, Consumption, Civil and political rights, Developing countries Paper type Research paper

20

International Journal of Development Issues Vol. 10 No. 1, 2011 pp. 20-33 q Emerald Group Publishing Limited 1446-8956 DOI 10.1108/14468951111123319

1. Introduction Ofcially, remittances are becoming increasingly important as a source of external nance for developing countries with measures of remittances more than tripling from 1990 to 2005. Unofcially, a large portion of remittances by wire transfers and unregulated sources may not be counted; thus, total remittances may be 50 percent higher than ofcial estimates (The World Bank, 2006). Recent literature draws no consensus on the impact of remittance ows on the growth rates of home economies (Ozden and Schiff, 2005). Some authors nd a positive relationship between remittances and growth (Pradhan et al., 2008; Jongwanich, 2007; Faini, 2001; Taylor, 1992; Stark and Lucas, 1988), while some nd a negative effect (Chami et al., 2003; Amuedo-Dorantes and Pozo, 2006a; Hanson, 2005, 2007; Acosta et al., 2009; Fajnzylber and Lopez, 2008; Funkhouser, 1992; Bourdet and Falck, 2006); and some
JEL classication O11, O15, O17

nd that remittances have no effect on growth (IMF, 2005; Spatafora, 2005). Regardless of the direct effect remittance ows have on growth, indirectly, it certainly affects consumption and/or investment in the home country. Several empirical studies nd that remittances are primarily used for consumption purposes (Stahl and Habib, 1989; Adelman and Taylor, 1990; Taylor, 1992; Glytsos, 1993; Kapur and McHale, 2005). But other studies argue that remittances are channeled into investment (Oberai and Singh, 1980; Durand et al., 1996; Chandavarkar, 1980; Burki, 1984; Russell, 1986; Brown, 1997; Adams, 1998; Funkhouser, 1995; Woodruff and Zeneto, 2001). These remittances may then be used to nance investment projects and small businesses in home countries with limited credit markets (Ameuedo-Dorantes and Pozo, 2006b). On the investment side in particular, several scholars have explored the effect of remittances on growth through the nancial sector. Mundaca (2009) studies the impact of remittances and nancial intermediation on growth and nds that greater nancial market development increases the positive impact of remittances on long-run growth by channeling remittances toward more productive projects. Contrary to Mundacas work, Giuliano and Ruiz-Arranz (2008) nd that countries with less nancial market development experience greater growth from remittance ows because remittances provide a substitute for credit-based nancing of projects. Ziesemer (2009) argues that remittances may inuence economic growth in the home country through increases in human and physical capital by enhancing savings. Especially in poorer countries, increases in savings will: . reduce interest rates thereby increasing investment; and . increase or maintain school enrollment and literacy. The increases in investment and literacy have a positive impact on growth. In a seemingly separate context, it is well known that civil liberties, sometimes immersed within broadly dened variables as institutional structure or economic freedom, positively affects a countrys level of investment (Feng, 2001; Helliwell, 1994; Pastor and Hilt, 1993; Pastor and Sung, 1995). Kormendi and Meguire (1985) nd a positive effect of civil liberties on growth with a large impact on investment. Dawson (1998) nds that institutions impact growth indirectly through investment and directly through total factor productivity. Gwartney (2009) nds a strong positive link between economic freedom and private investment as a share of GDP. In countries with greater economic freedom, investment will have a stronger positive impact on growth (Cole, 2003). The direction of causation may run in the opposite direction with faster growth leading to greater investment and better institutions. This may be particularly true for countries experiencing periods of negative growth (Helliwell, 1994; Barro, 1997). For example, the fall of the Soviet Union forced Fidel Castro to open the Cuban economy to several such reforms as that nation struggled to recover from the loss of Soviet support programs. Given the positive effect of remittances on investment and consumption outlined earlier, and the relationship between civil liberties and investment just mentioned, we hypothesize that the effect remittances has on investment and consumption is itself a function of civil liberties. We also claim that the effect remittances has on the former, relative to the latter, is larger in countries that are more free (i.e. have greater levels of civil liberties).

Remittances and civil liberties

21

IJDI 10,1

22

In other words, if an individual that receives remittances from abroad feel that their civil liberties are respected by the home government, then that individual is more likely to invest the remittance payments than they are to use it for consumption purposes. Otherwise, as an individuals civil liberties become increasingly violated, and state actors become more rent seeking in their behavior, returns to investment decline and so does the incentive to invest. To this end, agents in the home country receiving the remittance payments will likely resort to purchasing consumables whereby the state has little control after the purchase is made. For example, assume that a spouse in a home country knows that the state will not infringe on her/his ability to operate a business, he/she may be well inclined to use the remittance payments from their partner working abroad to start a shoe repair business. However, if this same person believes that once the business is operating, state actors will negatively affect its ability to prot, this same person may then use the payments for purchasing shoes rather than the equipment to repair them. Until now, this interaction may be clouded by the fact that no one actually investigates the marginal effect of remittances on investment or consumption as a function of civil liberties. Further complicating the picture are studies that indirectly address the institutional effects of remittances on growth but not on growths component parts (Abdih et al., 2008; Le, 2008; Pradhan et al., 2008; Catrinescu et al., 2009). We use annual data that covers over 30 years and 130 developing countries, and the civil liberties subset of the Freedom of the World Index (FWI) as a proxy for civil liberties. We nd that, in general, the marginal effect of remittances on investment is positive and statistically signicant, but decreasing in civil liberties up to level 6 out of 7 (a civil liberty level of 1 indicates a country that is free and 7 indicates not free). For consumption, this effect increases and is signicant after a civil liberty level of 2. Together, both dynamics indicate that there may indeed be a substitution effect from investment to consumption as civil liberties deteriorate. However, this effect only holds for developing south economies, not for emerging economies. We also nd that the marginal effect on investment diminishes less quickly as economies become less free than it increases for consumption indicating that the substitution is not quite one-for-one; exactly why this is the case is an area for future research. This study is organized as follows. Section 2 outlines the data used for the study and the econometric method employed. Specically, this section details not only the standard variables investigated but also a lengthy explanation of the civil liberty variable we employ in this paper, while the econometric subsection gives a detailed outline of the multi-stage dynamic panel model employed. Section 3 reveals the rst- and second-stage results from our model(s), as well as outlining the inference gathered from each, and Section 4 concludes the study. 2. Model and data As argued above, existing literature implies that the effect remittances has on investment and/or consumption is a function of civil liberties. But these studies typically embed investment and consumption within growth rather than separately evaluating these component parts. Furthermore, when interactions are employed, the inference drawn is static with regard to the varying degrees of institutional development. And third, none of the studies directly explore civil liberties proper; they usually aggregate civil liberties with other aspects of political or economic freedom. We intend to not only measure the level of inuence civil liberties itself has on the marginal effect of

remittances on investment and consumption separately, but also measure it across all levels of civil liberties. To this end, we employ a simple dynamic panel model that captures this effect; hence, equation (1) takes the form: yit a i 0 a1 yit21 a2 yit22 a3 r it a4 cl it a5 r it cl it eit : 1

Remittances and civil liberties

Equation (1) states that investment or consumption as percentages of GDP, yit , for country i at time t, is a function of a linear representation of two lags of itself, yit21 and yit22 , to purge any business cycle effects from the effect(s) of interest, remittances as a percentage of GDP, rit, civil liberties, clit, and the interaction of the two. (It can be shown that valid inference would be jeopardized if the additively linear forms of each variable in the interaction were not included in the regression (Brambor et al., 2005).) The investment, consumption, and remittances data come from the World Banks World Development Indicators database, and spans the years 1973-2006. We did not want to swamp the interesting effect of remittances impact in developing nations, so we dropped all nations from our dataset that have been in the Organization for Economic Co-operation and Development since 1973[1]. Our dataset has unbalanced panels and covers 130 countries with 2,318 observations. We give a more detailed explanation of the civil liberties proxy below. Civil liberties In the search for answers regarding the impact of institutions, scholars have devised many measures of institutional development and it is reasonable to attribute at least part of the disparity of results to the ingenuity of scholars trying to answer similar questions using these measures. Acemoglu et al. (2001) uses a protection against risk of expropriation index compiled using data from Political Risk Services as a proxy for institutions. Hall and Jones (1999) attribute cross-country differences in output per worker to institutional and policy differences, which they term social infrastructure. They measure social infrastructure by combining the International Country Risk Guide, an index of government anti-diversion policies from Political Risk Services, and an index of openness to international trade created by Sachs and Warner (1995). Glaeser et al. (2004) look at whether institutions cause growth and discuss many of the available measures. Not satised with existing measures of institutions, some authors devised composite indexes to measure various factors deemed important to the institutional framework. Rodrik et al. (2004) look at the relative importance of institutions, geography and trade, concluding that institutions rule. They measure institutions using a composite indicator of property rights and the rule of law. Scully and Slottje (1991) provide an index of economic liberty that covers only a limited time. The Heritage Foundation (2011), a well-known conservative think tank in Washington, DC also developed an index of economic freedom beginning in 1995[2]. These various measures aim to quantify the institutional environment and/or the policies in place. The methods and design of measures of institutions are as plentiful as the authors who employ them. Some focus on economic freedom, while others measure political freedom and democracy. Two of the more ambitious efforts to create a composite measure of institutions are the Economic Freedom of the World Index (EFW) by the Frasier Institute (2010)[3] and the FWI by Freedom House (2008)[4], known commonly as the Gastil Index. These provide more comprehensive composite measures that consider numerous factors, cover a longer time period and more countries than efforts by other authors.

23

IJDI 10,1

24

The EFW measures a variety of factors related to economic freedom at ve-year intervals from 1970 to 1995, and annually since 2000. Ratings are available for 100 countries continuously since 1980. Gwartney (2009) provides a basic overview of some research based on the EFW. One problem with this particular variable for our purposes is that there is no set of component parts that make up civil liberties. Furthermore, annual data only exist since 2000; prior data are at ve-year intervals. This creates an obvious econometric shortcoming when employing a dynamic model. But the FWI, another widely used composite index developed by Freedom House, specically measures:
[. . .] the opportunity to act spontaneously in a variety of elds outside the control of the government and other centers of potential domination according to two broad categories: political rights and civil liberties.

Freedom House cites the Universal Declaration of Human Rights as the basis for its denitions of political rights and civil liberties, going beyond simply rating government or government performance to evaluate the freedom of individuals faced with the behavior of non-state actors as well. As dened by the Freedom House web site Civil liberties allow for the freedoms of expression and belief, associational and organizational rights, rule of law, and personal autonomy without interference from the state. The FWI has the longest history of all indicators. Work on the index began in the 1950s with Raymond Gastil; thus, many still refer to the index as the Gastil Index. Freedom House expanded the survey in 1972 and Gastil continued to produce the annual survey until 1989 when Freedom House expanded the team of analysts. Another expansion in the mid-1990s shaped the comprehensive survey currently produced annually. The FWI has been used by numerous authors in their empirical research into institutions. For example, Scully (1988, 2002) use FWI to examine the relationship between the institutions and economic growth. Knack and Keefer (1995) use the FWI political freedom index to measure institutions in their study of economic performance. Islam and Winer (2004) investigate the impact of economic growth on civil liberties and political rights using the sum of these two measures in the FWI. Weller and Singleton (2004) use the FWI to test the impact of civil liberties and political freedoms on nancial stability, and nd that greater levels of civil liberties reduce the chance of banking and currency crises in relatively closed economies. The 2008 edition of the FWI includes ratings and reports for 193 countries and 15 territories. Each country and territory receives two numerical ratings. Political rights are rated on a scale of one to seven with one being the highest degrees of freedom and seven being the lowest. Civil liberties are rated separately, also on a scale of one to seven, with one being the highest degrees of freedom in this area. Civil liberties are rated based on the answers to 15 questions regarding various aspects of civil liberties, including associational and organizational rights, rule of law, personal autonomy and individual rights. Adding particular importance, then, is the interpretation that countries and territories with this rating enjoy free economic activity and tend to strive for equality of opportunity. In other words, a civil liberties rating of one implies that economic agents can do what they want with their money without fear of government intervention while agents that live in a country with a rating of seven cannot spend as they desire. The implications here being that in the former economy,

agents are free to invest as they please (e.g. spend money on opening a business knowing that their prots are safe from the government), while agents in the latter economy are more likely to spend their money on consumables. In short, this description of civil liberties faced by individuals within an economic system ts with the general understanding of the institutional arrangements necessary to encourage economic growth. The FWI enjoys a long track record, comprehensive measures, and widespread use among scholars; thus, in our empirical analysis in this paper, we employ the measure of civil liberties within the FWI as our measure of institutions. Econometrics The methodology we employ to estimate equation (1) is a system generalized method of moments (GMM) procedure that not only accounts for the inherent endogeneity problems embodied in any dynamic panel model, but also accounts for possible feedback from yit to rit. It would certainly be reasonable to assume that the level of remittances sent back to a home country could be inuenced by that countrys existing level of investment or consumption. For instance, if a family in a home country has relatively low levels of consumption, a family member working abroad may try to send a greater percentage of their pay back to their family. Or, if the economy of the home country has an unexpected increase in the amount of domestic investment, possibly increasing the level of employment in that country, someone working abroad may send less money home. The specication problem that exists in the lagged yit s is obvious when equation (1) is rst-differenced to purge the equation of the country effects, ai0 . For example, rst-differencing a simple form of equation (1) whereby only the yit s are included in the equation, we get yit 2 yit21 a1 yit21 2 yit22 a2 yit22 2 yit23 eit 2 eit21 . It is easy to see that the rst lagged difference, yit21 2 yit22 , is in fact correlated with the rst differenced errors, eit 2 eit21 ; therefore, the difference yit21 2 yit22 must be instrumented. The procedure employed in this paper is a two-stage estimation method whereby in the rst stage, system GMM estimates of the endogenous variables are generated using a GMM construct that relies on both lagged differences and levels of the endogenous variables as instruments. Anderson and Hsiao (1982) recommend using the level, yit22 , or the second lag difference, yit22 2 yit23 , as instruments for yit21 2 yit22 . Arellano and Bond (1991) suggest using the level in a GMM platform that would incorporate a larger number of moments and start at time period zero, and Arellano and Bover (1995) extend this to highly persistent series whereby the differenced series is estimated using levels as instruments, and a levels series is estimated using differences as instruments. This is the so-called system GMM (Blundell and Bond, 1998). While the optimal methodology, the system GMM has one drawback it can quickly gain in the number of instruments and therefore the possibility of over-specication is a concern. In general, it is desirable to keep the number of instruments no more than the number of countries (Roodman, 2006). To address this concern, we only use the rst through third lags of each endogenous variable as instruments for our rst-stage estimation. We also employ specication corrections to the two-step covariance matrix (Windmeijer, 2005), test for second-order serial correlation (rst-order is expected given the design of the method) using an Arellano and Bond (AB) test, and test for independence between the residuals and the instruments using a Hansen test.

Remittances and civil liberties

25

IJDI 10,1

Having formatted our primary model, the specication of interest actually lies in the marginal effect of remittances on yit derived from equation (1); this marginal effect takes the form: yit a1 a5 cl it : 2 r it Equation (2) states that the marginal effect of remittances on investment/consumption is now a linear function of civil liberties. Even though our primary interests lie in the mean value of equation (2), perhaps of equal importance is the statistical signicance of that effect. The standard errors of the marginal effect are non-linear and calculated as: q ^ ^ ^ ^ ^ sy=r vara1 x2 vara5 2xit cova1 a5 it with our condence interval set at 90 percent. 3. Results Table I lists the estimates from the rst-stage regressions for investment, consumption, and remittances, respectively, for all developing countries. We nd that consumption and remittances have an AR(2) process, but investment has an AR(1); albeit, the second-order component for remittances is actually rather weak as reected in the marginally signicant p-value. The AB statistic indicates that all regressions are second-order serially independent, and the Hansen statistics indicate that the lags of all variables are independent of the errors lending support that the estimates of these variables should be valid instruments. Last, over-specication should not be an issue as the number of instruments in each regression fall well below the number of countries. Having established the instruments, we will move on by incorporating each on the right-hand side in their respective second-stage regressions from equation (1) above. Table II lists the pooled second-stage estimates from equation (1). (We are not yet distinguishing between emerging and developing south economies.) The second and third columns are estimates for the investment equation and the fourth and fth columns are the same for consumption. We nd that there are AR(2) processes for both variables, and the sample correlation coefcient is quite high (the sample correlation coefcient, or R 2, does not reect the impact of the xed effects)[5]. But the interesting outcome lies in the sign of the interaction of civil liberties and remittances. This coefcient is negative
Investment Coefcient p-value yit2 1 yit2 2 No. observations No. countries No. instruments AB test Hansen test 0.709 20.043 2,581 133 125 0.171 0.393 0.000 * * 0.259 Consumption Coefcient p-value 0.678 0.116 2,581 133 125 0.199 0.434 0.000 * * 0.007 * * Remittances Coefcient p-value 1.021 20.081 2,581 133 125 0.774 0.330 0.000 * * 0.093 *

26

Table I. First-stage GMM estimations of investment and consumption

Notes: Signicance at: *10 and * *5 percent; the AB and Hansen test numbers reect the p-values attained from their respective test statistics

Investment Coefcient estimate yit-1 yit-2 Remittancesit Civil libertiesit Remitit *CLibit No. observations No. countries R2 0.491 0.149 0.382 20.199 20.054 2,318 130 0.184

p-value 0.000 * * 0.000 * * 0.000 * * 0.163 0.001 * *

Consumption Coefcient estimate 0.529 0.077 2 0.072 2 0.221 0.085 2,318 130 0.308

p-value 0.000 * * 0.026 * * 0.438 0.164 0.000 * *

Remittances and civil liberties

27
Table II. Second-stage investment and consumption regression estimates from equation (1)

Notes: Signicance at: *10 and * *5 percent; Remit is short for remittances and CLib is short for civil liberties; the R 2-value does not reect the t of the model due to the xed effects, only the t of the model due to the slope parameters

in the case of investment, but positive in the case of consumption. In other words, the marginal effect of the former is decreasing in civil liberties, but the reverse is true for the latter. This is the rst indication that our hypothesis has validity. But Figure 1 lends greater insight into the marginal effect of remittances on investment and consumption than does Table I. The left-hand graph depicts the estimated marginal effect from equation (2) for investment as derived over all levels of civil liberties; the right-hand graph depicts the same but for consumption. Evaluating the marginal effect of remittances on investment and consumption over all levels of civil liberties, we certainly seem to nd dynamics that are opposite one another as countries become less free. Specically, remittances positively affects investment up to a civil liberties level of 6 whereby the lower bound crosses zero, with this effect declining by 0.054 in value for each level of liberty. On the other hand, remittances positively affects consumption from a civil liberty level of 2 onward, and increases at a greater rate of 0.085 in value for each level of liberty. While interesting, the results so far may be inuenced by a countrys level of development. For instance, is the remittance effect on investment the same for an emerging economy as it is for a less developed economy? Especially with regard to investment, it can easily be argued that there could be signicant obstacles to the ease at which investment and saving can ow into sectors that could use it most. It has been shown that less developed nations tend to have a greater degree of separation between
Marginal effect of remittances on domestic invesment as a function of civil liberties 0.8 Marginal effect Marginal effect 0.6 0.4 0.2 0 0.2 0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free Marginal effect of remittances on domestic consumption as a function of civil liberties 0.8 0.6 0.4 0.2 0 0.2 0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free

Figure 1. Graphs of the investment and consumption marginal effects from equation (2)

IJDI 10,1

28

savers and investors in the sense that these countries tend to have a less nancially savvy and connected demographic (Aghion et al., 1999). Along this same line, it has also been argued that in more developed countries, a society will exist whereby sophisticated economic agents will spend quite a bit of effort neutralizing their portfolios from economic volatility thereby increasing the ease at which monies ow to and from the sectors that require liquidity injections with the highest returns (Edwards and Thames, 2010). To this end, Table III, like Table II, lists estimates from equation (1), and Figure 2, like Figure 1, shows the remittance effects for investment and consumption, but both are evaluated separately for emerging and developing economies along the lines of Akin and Kose (2007). When the countries are delineated by emerging and developing south economies, we nd that the results from Table II and Figure 1 were dominated by the developing south economies. In fact, the respective decline in the investment effect and increase in the consumption effect with increasing civil liberties seem to have actually been somewhat muted by the inclusion of the emerging economies. For investment, the effect of remittances declines at a more robust 0.06 points for each level of civil liberty, and the effect on consumption increases at 0.091 for every civil liberty level. With Figure 2, it becomes obvious that the developing south economies are dominating the outcomes. In fact, the marginal effect of remittances on investment is nowhere signicant over the respective levels of civil liberties for emerging economies, and only marginally signicant for the consumption variable from civil liberty level 4 to about level 6. Exactly why our hypothesis only holds for developing south economies is not clear at this point. It very well could be the larger degree of separation between investors and savors as argued above, but that argument would only hold for investment
Investment Coefcient estimate Emerging south yit2 1 yit2 2 Remittancesit Civil libertiesit Remitit *CLibit No. observations No. countries R2 Developing south yit2 1 yit2 2 Remittancesit Civil libertiesit Remitit *CLibit No. observations No. countries R2 0.792 2 0.136 2 0.173 0.014 0.009 490 22 0.226 0.453 0.183 0.427 2 0.293 2 0.060 1,828 108 0.185 Consumption Coefcient estimate 0.919 20.014 0.169 0.044 0.001 490 22 0.466 0.494 0.075 20.077 20.264 0.091 1,828 108 0.298

p-value 0.000 * * 0.109 0.474 0.958 0.840

p-value 0.000 * * 0.862 0.406 0.843 0.981

0.000 * * 0.000 * * 0.000 * * 0.077 * 0.001 * *

0.000 * * 0.051 * 0.460 0.165 0.000 * *

Table III. Second-stage investment and consumption regression estimates from equation (1)

Notes: Signicance at: *10 and * *5 percent; Remit is short for remittances and CLib is short for civil liberties; the R 2-value does not reect the t of the model due to the xed effects, only the t of the model due to the slope parameters

Marginal effect of remittances on domestic invesment for emerging south economies 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free Marginal effect of remittances on domestic invesment for developing south economies 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 Marginal effect Marginal effect 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 Marginal effect Marginal effect

Marginal effect of remittances on domestic consumption for emerging south economies

Remittances and civil liberties

29
0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free Marginal effect of remittances on domestic invesment for developing south economies

0 1 2 3 4 5 6 7 8 1-2.5 Free; 3-5 Partly free; 5.5-7 Not free

Figure 2. Graphs of the investment and consumption marginal effects from equation (2) as delineated by development level

and not consumption. It could be that remittances themselves are swamped by the relative size of an emerging economy and therefore the effects are themselves minor, but if this were the case, the dynamic over the different levels of civil liberties should be the same, just that the effects are insignicant. But, whatever the reason, it would be beyond the scope of this particular study and will be left for future research. 4. Conclusion Remittances, increasingly important as a source of external nance for developing countries, have been a topic of interest to growth economists. This literature has not reached a consensus regarding the direct effect of remittances on growth, yet remittance ows certainly have an indirect impact on consumption and/or investment in the home country. In the institutional growth literature, it is well-established that civil liberties positively affects a countrys level of investment. In this paper, we investigated the interactive effect of remittances and civil liberties on investment and consumption. Individual recipients of remittances from abroad are more likely to invest remittance payments received if they feel that their home government will respect their civil liberties. With increasing violations of individual civil liberties, returns to investment decline and so does the incentive to invest. We use a two-stage system GMM procedure, annual data that cover over 30 years and 130 developing countries, and the civil liberties subset of the FWI as a proxy for civil liberties. We nd that, in general, the marginal effect of remittances on investment is positive and statistically signicant, but decreasing in civil liberties up to level

IJDI 10,1

30

6 out of 7 (a civil liberty level of 1 indicates a country that is free and 7 indicates not free). For consumption, this effect increases and is signicant after a civil liberty level of 2. Together, both dynamics indicate that there may indeed be a substitution effect from investment to consumption as civil liberties deteriorate. However, this effect only holds for developing south economies, not for emerging economies. We also nd that the marginal effect on investment diminishes less quickly as economies become less free than it increases for consumption indicating that the substitution is not quite one-for-one. Given the positive impact of investment on growth, our results indicate that developing south economies with low levels of civil liberties could benet by improving civil liberties in ways that would encourage recipients to channel remittances into investment rather than consumption.
Notes 1. These countries include Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Israel, Italy, Japan, Luxembourg, The Netherlands, New Zealand, Norway, Portugal, Great Britain, and the USA. 2. The current version is available online at: www.heritage.org/index/ranking.aspx 3. The current version is available at: www.freetheworld.com 4. The current version is available at: www.freedomhouse.org 5. It may be a concern to some that there is an AR(2) process for investment in the second-stage estimation, but not in the rst-stage estimation. However, one must also realize that the conditioning sets are different in each case. References Abdih, Y., Chami, R. and Montiel, P. (2008), Remittances and institutions: are remittances a curse?, IMF Working Papers WP/08/09, IMF, Washington, DC. Acemoglu, D., Johnson, S. and Robinson, J.A. (2001), The colonial origins of comparative development: an empirical investigation, American Economic Review, Vol. 91 No. 5, pp. 1369-401. Acosta, P.A., Lartey, E.K.K. and Mandelman, F.S. (2009), Remittances and the Dutch disease, Journal of International Economics, Vol. 79, pp. 102-16. Adams, R.H. Jr (1998), Remittances, investment, and rural asset accumulation in Pakistan, Economic Development and Cultural Change, Vol. 47, pp. 155-73. Adelman, I. and Taylor, J.E. (1990), Is structural adjustment with a human face possible? The case of Mexico, Journal of Development Studies, Vol. 26, pp. 387-407. Aghion, P., Banerjee, A. and Piketty, T. (1999), Dualism and macroeconomic volatility, Quarterly Journal of Economics, Vol. 114 No. 4, pp. 1359-97. Akin, C. and Kose, M.A. (2007), Changing nature of north-south linkages: stylized facts and explanations, IMF Working Paper No. WP/07/280, IMF, Washington, DC. Ameuedo-Dorantes, C. and Pozo, S. (2006a), Migration, remittances and make and female employment patterns, American Economic Review, Vol. 96 No. 2, pp. 222-6. Ameuedo-Dorantes, C. and Pozo, S. (2006b), Workers remittances and business ownership in the Dominican Republic, The World Economy, Vol. 29 No. 7, pp. 939-56. Anderson, T.W. and Hsiao, C. (1982), Formulation and estimation of dynamic models using panel data, Journal of Econometrics, Vol. 18 No. 1, pp. 47-82.

Arellano, M. and Bond, S. (1991), Some tests of specication for panel data: Monte Carlo evidence and an application to employment equations, The Review of Economic Studies, Vol. 58 No. 2, pp. 277-97. Arellano, M. and Bover, O. (1995), Another look at the instrumental-variable estimation of error-component models, Journal of Econometrics, Vol. 68 No. 1, pp. 29-52. Barro, R.J. (1997), Determinants of Economic Growth: A Cross-country Empirical Study, Harvard University Press, Cambridge, MA. Blundell, R. and Bond, S. (1998), Initial conditions and moment restrictions in dynamic panel data models, Journal of Econometrics, Vol. 87 No. 1, pp. 115-43. Bourdet, Y. and Falck, H. (2006), Emigrants remittances and Dutch Disease in Cape Verde, International Economic Journal, Vol. 20 No. 3, pp. 267-84. Brambor, T., Clark, W.R. and Golder, M. (2005), Understanding interaction models: improving empirical analyses, Political Analysis, Vol. 14, pp. 63-82. Brown, R.P.C. (1997), Estimating remittance functions for Pacic Island migrants, World Development, Vol. 25, pp. 613-26. Burki, S.J. (1984), International migration: implications for labor exporting countries, The Middle East Journal, Vol. 38, pp. 668-84. Catrinescu, N., Leon-Ledesma, M., Piracha, M. and Quillin, B. (2009), Remittances, institutions, and economic growth, World Development, Vol. 37 No. 1, pp. 81-92. Chami, R., Fullenkamp, C. and Jahjah, S. (2003), Are immigrant remittance ows a source of capital for development?, IMF Working Paper 01/189, IMF, Washington, DC. Chandavarkar, A.B. (1980), Use of migrants remittances in labor-exporting countries, Finance & Development, June, pp. 36-9. Cole, J.H. (2003), The contribution of economic freedom to world economic growth: 1980-99, Cato Journal, Vol. 23, pp. 189-98. Dawson, J.W. (1998), Institutions, investment, and growth: new cross-country and panel data evidence, Economic Inquiry, Vol. 36, October, pp. 603-19. Durand, J.W., Kandel, W., Parrado, E.A. and Massey, D.S. (1996), International migration and development in Mexican communities, Demography, Vol. 33, pp. 249-64. Edwards, J. and Thames, F.C. (2010), Growth volatility and the interaction between economic and political development, Empirical Economics, Vol. 39 No. 1, pp. 183-201. Faini, R. (2001), Development, Trade, and Migration, IMF, Washington, DC. Fajnzylber, P. and Lopez, H. (2008), Remittances and Development: Lessons from Latin America, The World Bank, Washington, DC. Feng, Y. (2001), Political freedom, political instability, and policy uncertainty: a study of political institutions and private investment in developing countries, International Studies Quarterly, Vol. 45 No. 2, pp. 271-94. Frasier Institute (2010), Economic Freedom of the World, available at: www.Freetheworld.com Freedom House (2008), The Freedom of the World Index, available at: www.FreedomHouse.org Funkhouser, E. (1992), Migration from Nicaragua: some recent evidence, World Development, Vol. 20, pp. 1209-18. Funkhouser, E. (1995), Remittances from international migration: a comparison of El Salvador and Nicaragua, The Review of Economics and Statistics, Vol. 77 No. 1, pp. 137-46. Giuliano, P. and Ruiz-Arranz, M. (2008), Remittances, nancial development, and growth, Journal of Development Economics, Vol. 90, pp. 144-52.

Remittances and civil liberties

31

IJDI 10,1

32

Glaeser, E.L., La Porta, R.L., Lopez-de-Silanas, F. and Shliefer, A. (2004), Do institutions cause growth?, Journal of Economic Growth, Vol. 9, pp. 271-303. Glytsos, N.P. (1993), Measuring the income effects of migrant remittances: a methodological approach applied to Greece, Review of Economics and Statistics, Vol. 77, pp. 137-46. Gwartney, J. (2009), Institutions, economic freedom, and cross-country differences in performance, Southern Economic Journal, Vol. 75 No. 4, pp. 937-56. Hall, R.E. and Jones, C.I. (1999), Why do some countries produce so much more output per worker than others?, Quarterly Journal of Economics, Vol. 114, February, pp. 83-116. Hanson, G. (2005), Emigration, labor supply, and earnings in Mexico, NBER Working Paper No. 11412, NBER, Cambridge, MA. Hanson, G. (2007), Emigration, remittances, and labor force participation in Mexico, Integration and Trade Journal, Vol. 27, pp. 73-103. Helliwell, J. (1994), Empirical linkages between democracy and economic growth, British Journal of Political Science, Vol. 24 No. 2, pp. 225-48. Heritage Foundation (2011), Index of Economic Freedom, available at: www.heritage.org/index/ ranking.aspx IMF (2005), World Economic Outlook, International Monetary Fund, Washington, DC. Islam, M.N. and Winer, S.L. (2004), Tinpots, totalitarians (and democrats): an empirical investigation of the effects of economic growth on the civil liberties and political rights, Public Choice, Vol. 118, pp. 289-323. Jongwanich, J. (2007), Workers remittances, economic growth and poverty in developing Asia and the Pacic countries, UNESCAP Working Paper No. WP/07/01, ESCAP, Bangkok. Kapur, D. and McHale, J. (2005), Give Us Your Best and Brightest: The Global Hunt for Talent and Its Impact on the Developing World, Center for Global Development, Washington, DC. Knack, S. and Keefer, P. (1995), Institutions and economic performance: cross country tests using alternative institutional measures, Economics and Politics, Vol. 7 No. 3, pp. 207-27. Kormendi, R. and Meguire, P. (1985), Macroeconomic determinants of growth: cross country evidence, Journal of Monetary Economics, Vol. 16 No. 2, pp. 141-63. Le, T. (2008), Trade, remittances, institutions, and economic growth, International Economic Journal, Vol. 23 No. 3, pp. 391-408. Mundaca, B.G. (2009), Remittances, nancial market development, and economic growth: the case of Latin America and the Carribean, Review of Development Economics, Vol. 13 No. 2, pp. 288-303. Oberai, A.S. and Singh, H.K.M. (1980), Migration, remittances and rural development: nding of a case study in the Indian Punjab, International Labour Review, Vol. 119, pp. 229-41. Ozden, C. and Schiff, M. (Eds) (2005), International Migration, Remittances, and the Brain Drain, The World Bank, Washington, DC. Pastor, M. and Hilt, E. (1993), Private investment and democracy in Latin America, World Development, Vol. 21, pp. 489-507. Pastor, M. and Sung, J.H. (1995), Private investment and democracy in the developing world, Journal of Economic Issues, Vol. 29 No. 1, pp. 223-43. Pradhan, G., Upadhyay, M. and Upadhyaya, K. (2008), Remittances and economic growth in developing countries, The European Journal of Developmental Research, Vol. 20 No. 3, pp. 497-506.

Rodrik, D., Subramanian, A. and Trebbi, F. (2004), Institutions rule: the primacy of institutions over geography and integration in economic development, Journal of Economic Growth, Vol. 9, pp. 131-65. Roodman, D. (2006), How to do xtabond2: an introduction to difference and system GMM in stata, Working Paper No. 103, Center for Global Development, Washington, DC. Russell, S.S. (1986), Remittances from international migration: a review in perspective, World Development, Vol. 14, pp. 677-96. Sachs, J.D. and Warner, A. (1995), Fundamental sources of long-run growth, American Economic Association Papers and Proceedings, Vol. 87, pp. 184-8. Scully, G.W. (1988), The institutional framework and economic development, Journal of Political Economy, Vol. 96, pp. 652-62. Scully, G.W. (2002), Economic freedom, government policy and the trade-off between equity and economic growth, Public Choice, Vol. 113, pp. 77-96. Scully, G.W. and Slottje, D.J. (1991), Ranking economic liberty across countries, Public Choice, Vol. 69, pp. 121-52. Spatafora, N. (2005), Two current issues facing developing countries, World Economic Outlook, Chapter 2, International Monetary Fund, Washington, DC, A survey by the staff of IMF. Stahl, C.W. and Habib, A. (1989), The impact of overseas workers remittances on indigenous industries: evidence from Bangladesh, The Developing Economies, Vol. 27, pp. 269-85. Stark, O. and Lucas, R. (1988), Migration, remittances and the family, Economic Development and Cultural Change, Vol. 36 No. 3, pp. 456-81. Taylor, J.E. (1992), Remittances and inequality reconsidered: direct, indirect and intertemporal effects, Journal of Policy Modeling, Vol. 14 No. 2, pp. 187-208. Weller, C.E. and Singleton, L. (2004), Political freedom, external liberalization and nancial stability, International Review of Applied Economics, Vol. 18 No. 1, pp. 43-61. Windmeijer, F. (2005), A nite sample correction for the variance of linear efcient two-step GMM estimators, Journal of Econometrics, Vol. 126 No. 1, pp. 25-51. Woodruff, C. and Zeneto, R. (2001), Remittances and microenterprises in Mexico, UCSD working paper, UCSD, La Jolla, CA. (The) World Bank (2006), Global Economic Prospects: Economic Implications of Remittances and Migration, The World Bank, Washington, DC. Ziesemer, T.H.W. (2009), Worker remittances and growth: the physical and human capital channels, Jahrbucher fur Nationalokonomie und Statistik, Vol. 229 No. 6, pp. 743-73. Corresponding author Jeffrey A. Edwards can be contacted at: [email protected]

Remittances and civil liberties

33

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints