Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Chapter 16

Impact of Remittances on Frontier Markets’ Exchange Rate Stability

H.G. Keefe^*,^**

E.W. Rengifo^†

R. Trendafilov^**,^††
^*    Department of Economics, Fairfield University, Fairfield, CT, United States
^**    Center for International Policy Studies (CIPS), Fordham University, New York, NY, United States
^†    Department of Economics and the Center for International Policy Studies (CIPS), Fordham University, New York, NY, United States
^††    St. Thomas Aquinas College School of Business, Sparkill, NY, United States

Abstract

Most of current research devoted to frontier markets has stressed many different aspects of these economies, ranging from the importance of economic fundamentals to the roles of politics, policies, corruption, and foreign aid. However, little or no attention has been given to the impact of remittances on the local economies in terms of a source of stable funding with clear repercussions on financial and real markets. In this chapter we analyze the impact of remittances on several frontier markets and show their importance in the development and stability of capital markets.

Keywords

frontier markets

exchange rates

remittances

foreign direct investments

foreign portfolio investments

JEL classification

G11

G15

1. Introduction

The research goal of this chapter is to analyze the impact of remittances on exchange rate stability in currency markets in frontier economies. Frontier market economies are more developed than the least developed economies, but are still too small to be considered emerging markets. Many of them lack proper financial development and institutional strength, making them vulnerable to factors that can deteriorate economic stability or exacerbate fluctuations.

For many investors and practitioners, frontier markets are related to the level of development of their equity markets. These markets are typically small, less developed, and limited in terms of liquidity and investment opportunities, and they have limited informational flow and persistent information asymmetries. Though possessing greater risks than their “emerging” counterparts, these markets tend to provide strong growth prospects and have been deemed the last frontier in investments. It is important to note that economies that fall into the frontier market definition range widely in terms of wealth, size, and structure of the economy. We refer the readers to Speidell (2011) for a list of countries considered frontier markets. In terms of remittances, some are net remittance receiving economies (Colombia, Ecuador, Jordan, Latvia, and Pakistan), whereas others are the source of remittance sending (the United Arab Emirates).

In these frontier markets, one of the concerns for many central banks and investors alike is to understand factors that can influence exchange rate movements. Drastic movements in a country’s currency can lead to destabilizing effects on the domestic economy and investors’ returns. Volatile exchange rates lead to a decrease in trade volumes, slow economic growth, and diminishing wealth. Therefore it is necessary to understand how important sources of foreign currency inflows, such as remittances, impact exchange rates in these economies.

For many developing economies and a handful of frontier markets, remittances represent a significant and stable source of foreign currency entering into the domestic economy. Therefore it is important to understand how this stable and notable inflow can affect exchange rates and, in turn, investments into these economies.

Surges or sudden stops in foreign currency inflows can lead to greater volatility in nominal and real exchange rates. Surges in these flows lead to appreciation pressure on the domestic currency, whereas stops or reversals of flows can lead to depreciation and, in some cases, devaluation of local currencies. Combes et al. (2011) find that significant increases in such flows may cause the financial system to be more fragile and lead to an appreciation of the real exchange rate. As presented in Keefe (2013), the level of development is also an important factor in determining how such markets react to changes in foreign currency inflows. Keefe (2013) finds that in fact greater levels of trade openness, financial development, and institutional quality allow developing economies to be more effective in absorbing the changes in foreign currency inflows, thereby mitigating their effects on exchange rate volatility.

According to the World Bank, remittances are now the second largest resource inflow for developing countries behind foreign direct investment (FDI). In 2013, remittances reached $324 billion, whereas foreign direct investment flows were $736 billion, and foreign portfolio investment (FPI) flows were $65 billion in low- and middle-income economies. Remittance inflows have also been more stable than both FDI and FPI flows. Foreign investment flows react to both domestic and global economic conditions that can affect business profits and investment returns, such as changes in the world’s interest rates, loosening of monetary policy, slowing of growth in places like China and the United States, or institutional risk factors in the developing economies.

In this sense, it is important to note that remittance flows do not respond to the same incentives and motives that drive foreign investors to increase or decrease flows into developing economies. Remittance flows remain steady for two reasons. First, when remittances enter an economy, they enter as an income transfer from one household abroad to another one in the home country. The transfer of funds can be used for consumption, savings, or investment purposes, and the funds usually do not leave the economy (Chami et al., 2008). Second, immigrants tend to send a steady flow of remittances and may even increase the amount sent in times of economic downturns in the home country (World Bank, 2011). The familiar relations underpinning the flow of remittances make them unique in their reliability as a stable transfer of funds.

Motives to send remittances hinge on host country economic conditions that may result in a rise in migrant incomes as well as altruistic and insurance motives to help and maintain family ties in the home country. Therefore factors that may lead to diminishing foreign investment flows do not necessarily have the same impact on remittance flows. This can be seen in Fig. 16.1, particularly observing the flows during the global financial crisis from 2007 to 2009. As can be seen, during this period remittances fell by only 5%, whereas FDI and FPI flows into developing economies fell by 40 and 45%, respectively.

Figure 16.1 This figure depicts the inflow of FDI, FPI, and remittances into low and middle-income economies.
The majority of frontier markets would fall into this category. It is evident that remittances play a notable role as a source of foreign currency for many of these economies. (Data from World Bank World Development Indicators.)

In comparison to other types of capital inflows, remittances have been found to contribute to the stabilization of current account positions, while reducing the volatility of capital flows and output volatility in the home country (World Bank, 2005; Ratha, 2003, 2007; Bugamelli and Paterno, 2009; Chami et al., 2009; Gupta et al., 2009; Keefe, 2013). This impact on the domestic economy provides improved debt sustainability and better creditworthiness in developing countries, as well as a vital source of foreign currency when other flows dry up.

In this chapter, we present how remittance inflows impact variations in exchange rates by studying their impact during times of net capital outflows. Using an interaction term between remittance inflows and other capital flows, we determine whether the change in exchange rates is mitigated with high and stable levels of remittance inflows. The goal is to show the importance of these inflows in frontier markets, highlighting the fact that investors must take them into account when determining their investment decisions.

The chapter is organized as follows. Section 2 presents the model and methodology used, Section 3 describes in detail the data used and the results of the model, and Section 4 concludes and presents future venues of research.

2. Model and Methodology

It is important to note that remittances are inflows of foreign currency into a local economy. As such, during periods of significant capital inflows the impact on exchange rates is procyclical in the sense that it contributes to local currency appreciation. However, and as we describe later, in periods when other types of capital (FDI and FPI) leave the local economy, remittances continue coming into the economy and consequently it behaves countercyclically, providing a force that goes against the depreciation pressure caused by other funds exiting the local markets.

The variable of interest in this chapter is represented by changes in real effective exchange rates (REERs). In particular, we are interested in the behavior of this variable when there are large capital outflows and the presence of a consistent inflow of remittances.

As documented by the literature, diminishing capital flows result in a depreciation of local currency, and drastic changes in these capital flows may result in significant fluctuations in real effective exchange rates.

We use a fixed effects panel data model to analyze the relationship between changes in remittances and changes in REERs, considering the interaction between changes in remittances and changes foreign capital flows, defined as the sum of FDIs and FPIs.

The interaction between remittances and net foreign capital flows measures whether remittances counter depreciation pressures in the real effective exchange rate caused by large outflows in foreign direct investment or foreign portfolio investments. In this sense, an interaction term between remittances and net capital flows captures whether remittances, as a stable and reliable foreign currency inflow, provide a cushion in times of diminishing foreign capital flows. The model used is presented in the following equation:

$∆ R E E R_{i, t} = β_{0} + g (R e m i t_{i, t}, F l o w s_{i, t}) + β_{1} T r a d e + β_{2} T O T_{i, t} + β_{3} I n s t_{i, t} + ρ ∆ R E E R_{i, t - 1} + η_{i, t}$ $∆ R E E R_{i, t} = β_{0} + g (R e m i t_{i, t}, F l o w s_{i, t}) + β_{1} T r a d e + β_{2} T O T_{i, t} + β_{3} I n s t_{i, t} + ρ ∆ R E E R_{i, t - 1} + η_{i, t}$

(16.1)

The function g takes on the following specification:

$g (R e m i t_{i, t}, F l o w s_{i, t}) = β_{r m} ∆ R e m i t_{i, t} + β_{c f} ∆ F l o w s_{i, t} + β_{r f} D u m m y * ∆ R e m i t_{i, t}$ $g (R e m i t_{i, t}, F l o w s_{i, t}) = β_{r m} ∆ R e m i t_{i, t} + β_{c f} ∆ F l o w s_{i, t} + β_{r f} D u m m y * ∆ R e m i t_{i, t}$

(16.2)

In Eq. 16.1, the variable of interest is the changes in real effective exchange rate for country i at time t (∆REER_i,t); term of trade (TOT) and Trade represent the terms of trade and trade openness, respectively. The TOT is calculated as exports divided by imports, and trade openness is defined as the sum of exports and imports as a share of gross domestic product (GDP). The variable Inst represents institutional quality, as calculated by the World Bank World Governance Indicators.

In Eq. 16.2, ∆Remit_i,t represents the quarterly change remittance flows for country i at time t. ∆Flows represents the quarterly change of the net capital flows, measured as the sum of FDI plus FPI as a share of GDP. Also in this equation, Dummy*∆Remit_i,t provides a measure of how the interaction between remittances and net capital flows affects changes in exchange rates.

The dummy variable takes on the value of 1 during net capital outflows (∆Flows_i,t < 0) and zero otherwise.^a

When the change in flows is increasing, there is a greater supply of foreign currency entering into the economy, resulting in an appreciation of the currency characterized as a negative value of the change in real effective exchange rates. In this scenario, β_rm, β_cf, and β_rf are expected to be negative.

To understand the sign of the coefficients, assume that a change in remittances is positive (as it is in most of the cases presented later). If there are other capital inflows (ie, changes in other types of capital are also positive), the impact of remittances (β_rm), FDI, and FPI (β_cf) and the cross-effect (β_rf), all with negative signs, exert an appreciation pressure (the cyclical component of remittances). In the other hand, when other types of capital (FDI and FPI) exit the local economy, they exert a depreciation pressure (the coefficient β_cf < 0, multiplied by decreases in FDI and FPI, becomes positive, implying a depreciation pressure); meanwhile, remittances (that most of the time enter the economy when other types of capital are exiting) exert a direct counter effect given by its coefficient (β_rm < 0) and an indirect effect given by the cross-effect coefficient (β_rf < 0).

In this sense, it is important to measure these coefficients since they can provide us some light on the relationship and importance of changes in remittances in changes in real effective exchange rates, a crucial variable for the economy in general and for investors in particular.

3. Data and Results

This section presents the results for each of the regions we study: Africa, Asia, Eastern Europe, and Latin America. We present the fixed effects analysis to test the stabilizing effect of remittances on REERs when other capital flows (FDI and FPI) leave the country. We also consider other relevant variables such as TOT trade openness (Trade), nominal GDP, and institutional quality (Inst).

Quarterly data on remittance FDI, FPI, trade, and GDP are sourced from the International Financial Statistics database from the International Monetary Fund (IMF) and complemented with data from the central banks and statistics offices of each country or the World Bank data sets. The FDI and FPI data are consistent with the balance of payments manual (BPM6) reporting standards of the IMF and the BPM6 sign convention. Only in the case of Nigeria the data is based on the BPM5 reporting convention; however, the signs for FDI and FPI have been corrected to be aligned with BPM6. Although the difference in compilation of BPM5 and BPM6 goes beyond the sign convention, we believe that in the case of Nigeria that difference does not significantly impact our analysis. The availability of data consistent with BPM6 dictates the time frame on which we focus—first quarter of 2005 to last quarter of 2013.

The monthly data for the REERs come from Breugel.org. We later transform this monthly information into a quarterly basis. Last, the institutional quality is measured as the average of five governance indicators from the World Bank Governance Index. The index is collected annually and does not vary drastically in a short period of time. Therefore, given that the institutional factors remain relatively stable on a quarter-to-quarter basis, the annual indicators are used as quarterly proxies. To ease comprehension of the results, we assign values ranging from 0 to 5 for the institutional-quality variable, instead of the traditional –2.5 to 2.5 that is assigned by the World Bank. The five indices used in this research are Government Effectiveness, Voice and Accountability, Political Stability and Absence of Violence, Regulatory Quality, and Control of Corruption. Table 16.1 presents a summary of the institutional quality of the countries in the sample.

Table 16.1

Institutional Quality Index Computed as Average of Government Effectiveness, Voice and Accountability, Political Stability and Absence of Violence, Regulatory Quality, and Control of Corruption

	Mean	First observation	Last observation
Africa
Jordan	2.386	2.486	2.231
Mauritius	3.241	3.163	3.289
Namibia	2.863	2.761	2.872
Nigeria	1.382	1.440	1.353
Asia
Bangladesh	1.560	1.349	1.565
Kyrgyzstan	1.677	1.535	1.775
Pakistan	1.389	1.521	1.326
Sri Lanka	2.040	2.020	2.131
Eastern Europe
Bulgaria	2.760	2.800	2.675
Croatia	2.921	2.893	2.965
Estonia	3.514	3.481	3.568
Latvia	3.131	3.182	3.197
Lithuania	3.215	3.235	3.332
Macedonia	2.388	2.064	2.474
Romania	2.630	2.544	2.651
Serbia	2.385	2.282	2.448
Slovenia	3.438	3.454	3.331
Ukraine	2.028	2.051	1.807
Latin America
Argentina	2.285	2.328	2.212
Colombia	2.120	1.989	2.222
Ecuador	1.789	1.759	1.996
Panama	2.627	2.528	2.621

The time frame under study spans from the first quarter of 2005 to the last quarter of 2013.

A list of 47 countries fitting the definition of frontier market countries is available in Speidell (2011). The list includes countries in Africa (Botswana, Côte d’Ivoire, Gabon, Ghana, Kenya, Mauritius, Namibia, Nigeria, Senegal, Tanzania, Zambia); Asia (Bangladesh, Kazakhstan, Kyrgyzstan, Papua New Guinea, Sri Lanka, Vietnam); Eastern Europe (Bosnia, Bulgaria, Croatia, Cyprus, Estonia, Latvia, Lithuania, Macedonia, Malta, Romania, Serbia, Slovak Republic, Slovenia, Ukraine); Latin America (Argentina, Barbados, Colombia, Ecuador, Jamaica, Panama, Trinidad and Tobago); and Middle East (Bahrain, Jordan, Kuwait, Lebanon, Oman, Pakistan, Qatar, United Arab Emirates). In this chapter we study 22 of them based on data availability. The list of countries included is presented in Tables 16.1–16.3. Table 16.2 summarizes socioeconomic data related to the size of these economies and proxies for their living standards. At the end of Table 16.2 we also present information from USA, European Union, and China for comparison.

Table 16.2

Information About GDP, GDP Growth, GDP Per Capita, Unemployment, and Population Living Below the Poverty Line, for the Countries in Our Sample

	GDP (PPP) in billions	GDP growth (%)	GDP per capita (PPP)	Unemployment rate (%)	Population below poverty line (%)
Africa
Jordan	$79,770	3	$11,900	12.3	14.2
Mauritius	$23,420	3.3	$17,900	8.0	8.0
Namibia	$23,590	4.3	$10,800	27.4	28.7
Nigeria	$1,058,000	7.0	$6,100	23.9	70.0
Asia
Bangladesh	$535,600	6.2	$3,400	5.0	31.5
Kyrgyzstan	$19,290	4.1	$3,400	8.6	33.7
Pakistan	$884,200	4.1	$4,700	6.8	12.4
Sri Lanka	$217,100	7.0	$10,400	4.2	8.9
Eastern Europe
Bulgaria	$123,300	1.4	$17,100	11.0	21.8
Croatia	$87,300	−0.8	$20,400	21.0	21.1
Estonia	$35,400	1.2	$26,600	8.6	17.5
Latvia	$48,590	2.7	$23,900	9.5	NA
Lithuania	$78,950	3.0	$26,700	11.1	4.0
Macedonia	$27,410	3.4	$13,200	28.0	30.4
Romania	$386,500	2.4	$19,400	7.0	22.2
Serbia	$90,320	−0.5	$12,500	26.1	9.1
Slovenia	$60,540	1.4	$29,400	13.6	13.5
Ukraine	$373,100	−6.5	$8,200	8.8	24.1
Latin America
Argentina	$927,400	−1.7	$22,100	7.7	30.0
Colombia	$642,700	5.0	$13,500	9.2	32.7
Ecuador	$182,000	4	$11,400	5.0	25.6
Panama	$76,950	6.6	$20,300	4.5	26.0
China, EU, USA
China	$17,630,000	7.4	$12,900	4.1	6.1
EU	$17,610,000	1.4	$38,300	10.0	NA
USA	$17,460,000	2.4	$54,800	6.2	15.1

Data corresponds to 2014 that was published by the CIA World factbook.

Table 16.3

Presents the p-Values of the t-Test of the Mean of the Capital Flows (FDI and FPI) and the Capital Flows Plus Remittances

	t-test of mean flows to total flows: p-value	t-test of mean flows/GDP to total flows/GDP: p-value	Correlation remit/GDP and REER	Correlation d(remit/GDP) and d(REER)
Africa
Jordan	0.000	0.000	−0.771***	−0.103
Mauritius	1.000	0.999	−0.469**	0.029
Namibia	0.977	0.976	0.038	0.544***
Nigeria	0.000	0.000	−0.637***	0.116
Asia
Bangladesh	0.000	0.000	0.338**	0.160
Kyrgyzstan	0.000	0.000	0.708***	−0.322*
Pakistan	0.000	0.000	0.124	−0.078
Sri Lanka	0.000	0.000	0.342**	−0.196
Eastern Europe
Bulgaria	0.174	0.173	−0.874***	−0.033
Croatia	0.501	0.385	−0.751***	0.173
Estonia	0.520	0.518	−0.656***	−0.096
Latvia	0.090	0.065	0.538***	0.198
Lithuania	0.036	0.040	0.414**	−0.373**
Macedonia	0.073	0.072	−0.171	−0.234
Romania	0.004	0.006	0.340**	−0.279
Serbia	0.001	0.001	−0.351**	−0.060
Slovenia	0.766	0.754	−0.231	0.157
Ukraine	0.005	0.013	−0.385**	−0.427**
Latin America
Argentina	0.839	0.853	0.825***	−0.101
Colombia	0.013	0.000	−0.872***	−0.590***
Ecuador	0.000	0.000	−0.389**	−0.336**
Panama	0.630	0.661	−0.060	0.013

The first two columns of this table present the p-values of the t-test for testing the mean of capital flows (FDI Plus FPI) versus the mean of total flows (FDI plus FPI plus remittances). The third column presents the correlation between REER and remittance, and the fourth column presents the correlation between the differenced REER and differenced remittance. *, **, *** Represent significance at the 10, 5, and 1% levels, respectively. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013.

Looking at the GDP growth, many of the countries have grown at rates higher than that of the USA (2.4%) in 2014, with highest growth rates of 7% (Nigeria and Sri Lanka), 6.6% (Panama), and 6.2% (Bangladesh). Few of them have negative growth rates (Croatia, Serbia, Ukraine, and Argentina). As can be seen in this table, the frontier European countries and also the EU countries have overall lower growth rates.

With respect to GDP per capita, the European and Latin American frontier countries have higher values for this economic indicator (with highest in Slovenia at $29,000). However, these countries are still far from the average of the European Union ($38,300) and USA ($54,800).

The unemployment rates are quite high across the African and European frontier countries (with highest rates in Macedonia at 28% and Namibia at 27.4%). In comparison, for 2014, the unemployment rate in the USA was 6.2% and in the European Union 10%.

Finally, it is important to note that many of the frontier countries in the sample have large parts of their populations living in poverty, with highest numbers in Nigeria (70%) and Kyrgyzstan (33.7%). The low standards of living coupled with high unemployment rates and high percentage of the populations living in poverty emphasize the importance of remittance inflows as a source of income. That fact also underpins remittances as an important currency inflow component and the possibility that remittances can have a significant impact on the exchange rates.

A primary interest of this chapter is to determine the relationship between REERs and remittance flows. We have arranged the data in four regions: Africa and Middle East (Mauritius, Namibia, Nigeria, Jordan); Asia (Bangladesh, Kyrgyzstan, Sri Lanka, Pakistan); Eastern Europe (Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Serbia, Slovenia, Ukraine); and Latin America (Argentina, Ecuador, Colombia, Panama). We also have divided the countries in two groups depending on the remittances received in relation to other capital flows: high and low remittance countries.

From existing literature we know that the capital flows (FDI + FPI) have a significant effect on REER. Also, if a country has high-enough remittances compared with the size of the other capital flows, then remittances also have a significant effect on the REER.

In order to determine if a given country can be categorized as a high remittance one, we have compared the mean of the capital flows (FDI + FPI) and the mean of the capital flows plus remittance flows. If the means are statistically different, we categorize the country as high remittance country. To compare the two means we use a t-test, where the null hypothesis is that the means are the same. Following this categorization, we use panel regression analysis to examine in depth the effects of remittance inflows on the REER in the high remittance frontier markets.

The first column of Table 16.3 presents the p-values of the t-test of the mean of the capital flows (FDI and FPI) and the capital flows plus remittances. The second column of Table 16.3 presents the p-values of the t-test of the mean capital flows per GDP and the capital flows plus remittance per GDP. As can be seen from this table, the hypothesis results are very similar using both variable definitions (in levels and as a share of GDP), and out of the 22 countries 14 can be categorized as high remittance countries based on the t-test and a significance level of 10%.

In the third column of Table 16.3 we have calculated the correlation between the remittance per GDP and REER, and the fourth column presents the correlation between the difference in the remittance per GDP and the difference in REER. From this table, in most cases where the remittances are significantly large (high remittance countries), the correlation between the remittance and REER is negative (except for Nigeria, Bangladesh, and Latvia). This is a result that is confirmed by the literature in the sense that remittance flows have a negative relationship with REER.

In what follows we present a detailed analysis first by geographic region and later by the level of remittances of the countries (high remittance countries).

3.1. Africa and Middle East

Due to data restriction issues, we focus our attention on four frontier markets in the Africa and Middle East regions: Mauritius, Namibia, Nigeria, and Jordan.

As can be seen in Fig. 16.2, Jordan and Nigeria have sizable remittance inflows that remain stable in time. In contrast, their FDIs and FPIs are very volatile, both having periods of large surges and outflows. Further, from t-tests of the mean in the first two columns in Table 16.3 we can see that remittance inflows are large enough to significantly change the mean capital flows for these two frontier economies. Therefore, Jordan and Nigeria are categorized as high remittance countries.

Figure 16.2 This figure presents the flows of foreign currency from FDI, FPI, and remittances in selected African and Middle Eastern countries: Mauritius, Namibia, Nigeria, and Jordan.
The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

It can be also noted from third column in Table 16.3 that the correlations between the remittance per GDP and REER are negative and significant for these two countries (Jordan and Nigeria). In the case of the differentiated series (column four) the correlation is negative for Jordan although not significant, and positive and not significant for Nigeria.

On the other hand, Mauritius and Namibia have low levels of remittance inflows, though this may be attributed to low official reported remittance inflows. Most countries, especially those that have less developed financial markets, have a notable portion of remittance inflows that are unofficial and cannot be tracked, which may be the case in these two markets. The results of the t-test in Table 16.3 confirm the visual result from Fig. 16.2, positioning Mauritius and Namibia as low remittance countries.

In Fig. 16.3, we compare the changes in the real effective exchange rate and the change in remittance inflows. In Namibia, these changes are highly correlated; the correlation between the differenced remittance per GDP and the differenced REER is 0.544. In the other markets, the correlation is not strong, with the smallest correlation in Mauritius (0.029).

Figure 16.3 This figure presents the change in remittances and real effective exchange rates in select African and Middle Eastern countries: Mauritius, Namibia, Nigeria, and Jordan.
The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

Table 16.4 presents the results of using Eqs. 16.1 and 16.2. The overall impact of changes on remittances is not significant in this region, which can also be seen when we segment the high remittance receiving countries (Nigeria and Jordan) in Table 16.5. A feasible explanation may be the dynamics of the currency markets in these economies: Jordan is a highly dollarized economy, where foreign currency is used in parallel with the domestic currency; therefore foreign currency inflows may not need to be exchanged for local currency, thereby not impacting exchange rates. In the case of Nigeria and since 2009, it has had a relatively stable currency, with frequent government interventions to attempt to stabilize the nominal exchange rate since its precipitous depreciation observed from Oct. 2008 to Jan. 2009 and in Sept. 2014, according to Reuters (2014) (Fig. 16.3). Given these reasons, the effect of remittances on exchange rates is expected to be mild or not significant.

Table 16.4

Fixed Effects Panel Analysis for Africa and Middle East, Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.149	0.149	0.145	0.128	0.108
	(0.188)	(0.118)	(0.202)	(0.167)	(0.243)
dRemit (% GDP)	37.10	5.478	35.22	16.246	14.99
	(0.457)	(0.445)	(0.481)	(0.690)	(0.710)
d(Net Kflows) (% GDP)	−1.993	−2.092		−0.810	−0.682
	(0.469)	(0.453)		(0.703)	(0.746)
Dummy*d(Remit)		−170.9	−134.1	−142.4	−119.1
		(0.725)	(0.781)	(0.717)	(0.759)
Trade				3.913*	5.452**
				(0.060)	(0.016)
TOT				5.152**	5.604***
				(0.015)	(0.008)
Institutions					−9.95*
					(0.089)
R²	0.294	0.295	0.289	0.130	0.153
Adj. R²	−0.068	−0.081	−0.074	0.056	0.073
N obs.	116	116	116	116	116

The dependent variable is the quarterly difference in real effective exchange rates. There are four countries belonging to Africa and Middle East regions: Mauritius, Namibia, Nigeria, and Jordan. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels

Table 16.5

Fixed Effects Panel Analysis for Africa and Middle East Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	−0.218	−0.211	−0.207	−0.443**	−0.433*
	(0.355)	(0.385)	(0.383)	(0.041)	(0.051)
dRemit (% GDP)	21.55	21.60	27.49	38.49	43.25
	(0.661)	(0.669)	(0.566)	(0.347)	(0.308)
d(Net Kflows) (% GDP)	5.722	5.201		3.514	1.733
	(0.596)	(0.646)		(0.699)	(0.856)
Dummy*d(Remit)		109.1	151.6	−312.4	−272.1
		(0.815)	(0.735)	(0.487)	(0.549)
Trade				4.043	1.299
				(0.751)	(0.923)
TOT				8.131***	9.697**
				(0.007)	(0.014)
Institutions					−8.505
					(0.488)
R²	0.827	0.827	0.825	0.909	0.912
Adj. R²	0.471	0.443	0.466	0.668	0.656
N obs.	56	56	56	56	56

The dependent variable is the quarterly difference in real effective exchange rates. There are two countries belonging to Africa and Middle East regions that are classified as high remittance countries: Nigeria and Jordan. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in Parentheses are p-Values. *, **, *** Represent significance at the 10, 5, and 1% levels.

One thing to note from this table is the negative and significant impact that institutions have on the changes of exchange rates. According to these results, having good institutions contributes to local currency appreciation, and in times when capital exits the local economies they act countercyclically; that is, institutions counteract depreciation pressures generated by FDIs and FPIs leaving the economy.

3.2. Asia

In the Asian region we focus our attention on four frontier markets: Bangladesh, Pakistan, Kyrgyzstan, and Sri Lanka, where enough data were available. As can be seen in Fig. 16.4, all four economies have sizable remittance inflows that have been steadily increasing. The graphical results are supported by the t-test for the mean of the capital flows with and without remittance shown in the first two columns in Table 16.3. In all four countries, the remittances are sizable enough to categorize all of them as high remittance ones. The correlation between the differentiated remittance per GDP and the differentiated REER is negative for Kyrgyzstan (–0.322), Sri Lanka (–0.196), and Pakistan (–0.077), and it is positive for Bangladesh (0.160).

Figure 16.4 This figure presents the foreign currency flows from FDI, FPI, and remittances in four Asian countries: Bangladesh, Pakistan, Kyrgyzstan, and Sri Lanka, where enough data are available.
The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

Once again, foreign direct investment and foreign portfolio investment are very volatile, indicating periods of surges and outflows, but the relative size of these flows is smaller than remittance inflows. For Sri Lanka and Pakistan, the volatility in portfolio investment is especially large.

Fig. 16.5 compares the changes in the real effective exchange rate and the change in remittance inflows. These changes are sizable across all the economies.

Figure 16.5 This figure presents the change in remittances and real effective exchange rates in four Asian countries: Bangladesh, Pakistan, Kyrgyzstan, and Sri Lanka, where enough data are available.
The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

Table 16.6 presents the results using Eqs. 16.1 and 16.2. In this case, we see a statistically significant and negative impact on changes in remittance flows on changes in the real effective exchange rate. As expected, this result implies that increases (positive changes) in remittance inflows as a share of GDP contribute to appreciation of the real effective exchange rates. In addition, we see the same impact, though much weaker, of changes in other capital flows.

Table 16.6

Fixed Effects Panel Analysis for High Remittance Countries in Asia Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.170*	0.163*	0.172*	0.164*	0.164*
	(0.080)	(0.096)	(0.078)	(0.092)	(0.095)
dRemit (% GDP)	−39.09***	−33.31**	−34.13**	−32.60**	−32.52**
	(0.004)	(0.026)	(0.021)	(0.047)	(0.050)
d(Net Kflows) (% GDP)	−7.255		−7.298	−7.579*	−7.578*
	(0.111)		(0.109)	(0.097)	(0.099)
Dummy*d(Remit)		−28.38	−29.04	−40.71	−40.99
		(0.427)	(0.413)	(0.254)	(0.258)
Trade				−1.237	−1.301
				(0.643)	(0.653)
TOT				4.445*	4.533
				(0.059)	(0.104)
Institutions					0.370
					(0.953)
R²	0.513	0.503	0.516	0.527	0.537
Adj. R²	0.313	0.299	0.311	0.327	0.319
N obs.	135	135	135	135	135

The dependent variable is the quarterly difference in real effective exchange rates. There are four countries in the analysis that are classified as high remittance countries: Bangladesh, Pakistan, Kyrgyzstan, and Sri Lanka. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

This result is really important and goes hand in hand with what was already explained. If remittances are a constant inflow of funds into local economies, they play a significant role against depreciation pressures when other forms of capital (FDI and FPI) are exiting the economy and as such, remittances have an important effect in analyzing and understanding exchange rate dynamics.

Note that even though the interaction terms are not significant, it is clear from these results that when there is a net outflow of foreign currency derived from FDI or FPI, remittances counteract the impact of those capital outflows. According to the coefficient value and sign, when there is a capital outflow, the depreciation pressure will be countered by the stable remittance inflows.

3.3. Eastern Europe

In the Eastern Europe region, we have the largest representation of frontier markets. We focus our attention on 10 of them where enough data are available. As can be seen in Fig. 16.6, the relative size of remittance inflows to all other capital inflows varies across countries. The countries included in the analysis are: Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Serbia, Ukraine, and Slovenia. Out of the 10 countries, 6 can be identified as high remittance countries according to the results in the first two columns in Table 16.3 and significance level of 10%: Latvia, Lithuania, Macedonia, Romania, Serbia, and Ukraine. In all high remittance countries, except Latvia, the correlation between the differentiated remittance per GDP and the differentiated REER is negative. The highest negative correlation is observed in Ukraine (–0.427) and the smallest in Bulgaria (–0.033). Once again, foreign direct investment and foreign portfolio investment are very volatile, indicating periods of surges and outflows.

Figure 16.6 This figure presents the inflow of foreign currency from FDI, FPI, and remittances in Eastern Europe.
The countries included in the analysis are: Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Serbia, Ukraine, and Slovenia. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

In Fig. 16.7, we compare the changes in the real effective exchange rate and the change in remittance inflows. These changes are sizable across all the economies.

Figure 16.7 This figure presents the change in remittances and real effective exchange rates in Eastern Europe.
The countries included in the analysis are: Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Serbia, Ukraine, and Slovenia. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

Table 16.7 presents the results using Eqs. 16.1 and 16.2. Once more, we see a statistically significant and negative impact on increases in remittance inflows on changes in the real effective exchange rate. This result implies that increases in remittance inflows as a share of GDP contribute to appreciation of the real effective exchange rate. In contrast, the impact of changes in other capital inflows on the real effective exchange rate is not statistically significant, in line with the discussion presented earlier; if remittances are a constant inflow of funds into local economies, they play a significant role against depreciation pressures when other forms of capital (FDI and FPI) are exiting the economy. The effects hold true for the high remittance receiving countries as well in Table 16.8.

Table 16.7

Fixed Effects Panel Analysis for Eastern Europe Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.238***	0.235***	0.228***	0.232***	0.231***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
dRemit (% GDP)	−49.39**	−56.29**	−56.95**	−55.03**	−54.95**
	(0.017)	(0.013)	(0.012)	(0.016)	(0.016)
d(Net Kflows) (% GDP)	1.446	1.432		1.448	1.455
	(0.126)	(0.130)		(0.127)	(0.126)
Dummy*d(Remit)		47.96	49.79	49.42	49.92
		(0.452)	(0.436)	(0.443)	(0.439)
Trade				−0.537	−0.543
				(0.463)	(0.460)
TOT				0.542	0.537
				(0.802)	(0.804)
Institutions					0.385
					(0.839)
R²	0.325	0.326	0.321	0.327	0.327
Adj. R²	0.169	0.168	0.164	0.164	0.161
N obs.	348	348	348	348	348

The dependent variable is the quarterly difference in real effective exchange rates. There are ten countries in this analysis: Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Serbia, Ukraine, and Slovenia. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

Table 16.8

This Table Presents the Fixed Effects Panel Analysis for Eastern Europe Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.249***	0.249***	0.243***	0.242***	0.242***
	(0.001)	(0.001)	(0.002)	(0.002)	(0.002)
dRemit (% GDP)	−49.75*	−49.64	−51.72	−47.51	−47.49
	(0.083)	(0.116)	(0.102)	(0.135)	(0.136)
d(Net Kflows) (% GDP)	2.068	2.069		2.003	2.005
	(0.183)	(0.186)		(0.202)	(0.203)
Dummy*d(Remit)		−0.684	7.003	4.019	4.139
		(0.994)	(0.939)	(0.967)	(0.965)
Trade				−1.295	−1.296
				(0.240)	(0.241)
TOT				2.221	2.220
				(0.495)	(0.496)
Institutions					0.131
					(0.961)
R²	0.347	0.347	0.339	0.354	0.354
Adj. R²	0.168	0.163	0.158	0.161	0.155
N obs.	192	192	192	192	192

The dependent variable is the quarterly difference in real effective exchange rates. There are six countries in this analysis that are classified as high remittance countries: Latvia, Lithuania, Macedonia, Romania, Serbia, and Ukraine. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

3.4. Latin America

In the Latin America region, we focus our attention on four frontier markets: Argentina, Colombia, Ecuador, and Panama. As can be seen in Fig. 16.8, the relative size of remittance flows with respect to all other capital flows varies across countries, with Ecuador having the highest level of remittance flows relative to all other flows.

Figure 16.8 This figure presents the inflow of foreign currency from FDI, FPI, and remittances in Latin America.
The countries analyzed are Argentina, Colombia, Ecuador, and Panama. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

In most of these economies, remittances make up a small portion of total foreign currency flows. Once again, foreign direct investment and foreign portfolio investment are very volatile, indicating periods of surges and outflows.

Following the results presented in the first two columns of Table 16.3, Ecuador and Colombia can be identified as high remittance countries. In both cases the correlation between the differentiated series of the remittance per GDP and REER is negative and significant (–0.590 in Colombia and –0.336 in Ecuador).

In Fig. 16.9, we compare the changes in the real effective exchange rate and the change in remittance inflows. These changes are sizable across all the economies.

Figure 16.9 This figure presents the change in remittances and real effective exchange rates in Latin America.
The countries analyzed are Argentina, Colombia, Ecuador, and Panama. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. (Data from IMF International Financial Statistics Database.)

Table 16.9 presents the results using Eqs. 16.1 and 16.2. Once more, we see a statistically significant and negative impact from changes in remittance inflows on changes in the real effective exchange rate. This result implies that increases in remittance inflows as a share of GDP contribute to appreciation of the real effective exchange rate. In contrast, the impact of changes in other capital inflows on the real effective exchange rate is not statistically significant.

Table 16.9

This Table Presents the Fixed Effects Panel Analysis for Latin America Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.269***	0.277***	0.277***	0.230**	0.217*
	(0.007)	(0.006)	(0.006)	(0.038)	(0.057)
dRemit (% GDP)	−446.2***	−492.1***	−500.1***	−516.6***	−528.0***
	(0.002)	(0.004)	(0.003)	(0.004)	(0.003)
d(Net Kflows) (% GDP)	−1.391	−1.212		−0.931	−1.026
	(0.744)	(0.777)		(0.831)	(0.815)
Dummy*d(Remit)		159.1	165.7	142.5	165.9
		(0.594)	(0.575)	(0.645)	(0.597)
Trade				−6.749	−9.224
				(0.591)	(0.487)
TOT				−3.613	−3.697
				(0.191)	(0.184)
Institutions					3.857
					(0.555)
R²	0.527	0.529	0.528	0.538	0.542
Adj. R²	0.211	0.204	0.219	0.187	0.181
N obs.	148	148	148	144	144

The dependent variable is the quarterly difference in real effective exchange rates. There are four countries in this analysis: Argentina, Colombia, Ecuador, and Panama. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

In the Latin American frontier market economies, not only is the impact of changes of remittances on changes on the exchange rates negative and significant, but its magnitude is much larger than the ones observed anywhere else. These results are consistent when we only consider the high remittances countries (Table 16.10).

Table 16.10

This Table Presents the Fixed Effects Panel Analysis for Latin America Using the Specification Presented in Eqs. 16.1 and 16.2 for High Remittance Countries Identified in the Analysis

Fixed effects: time and country
d(REER) (−1)	0.252***	0.245***	0.241**	0.208**	0.208**
	(0.007)	(0.009)	(0.011)	(0.033)	(0.037)
dRemit (% GDP)	581.1***	−552.6***	−554.1***	−584.6***	−584.4***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
d(Net Kflows) (% GDP)	6.118	6.242		6.637	6.631
	(0.482)	(0.476)		(0.451)	(0.454)
Dummy*d(Remit)		−115.8	−110.3	−39.44	−39.02
		(0.643)	(0.658)	(0.878)	(0.880)
Trade				−7.086	−7.193
				(0.484)	(0.512)
TOT				−4.751	−4.751
				(0.237)	(0.239)
Institutions					0.086
					(0.979)
R²	0.306	0.307	0.303	0.324	0.325
Adj. R²	0.272	0.265	0.270	0.265	0.256
N obs.	88	88	88	88	88

The dependent variable is the quarterly difference in real effective exchange rates. There two countries in this analysis that are classified as high remittance countries: Ecuador and Colombia. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

3.5. High Remittance Receiving Frontier Countries

Finally, we focus on the frontier markets in all regions that have remittance flows whose mean is different from the mean of total capital flows to a statistically significant degree. We refer to these countries as “high remittance receiving countries.” There are 14 countries in this analysis that are considered high remittance receiving countries (ordered alphabetically): Bangladesh, Colombia, Ecuador, Jordan, Kyrgystan, Latvia, Lithuania, Macedonia, Nigeria, Pakistan, Romania, Serbia, Sri Lanka, and Ukraine.

Fig. 16.10 presents the flows of remittances and remittances as a share of GDP in each of the high remittance receiving economies. In many of them, remittances are steadily increasing. For others, such as Ecuador, Colombia, Jordan, Nigeria, and Romania, remittances as a share of GDP have been decreasing. It is important to note that this decrease may not be due to the decrease in gross remittance inflows, but instead can be attributed to the growth in GDP. It is important to note again that in order for the remittance inflows to have impact on the REER, they have to be a significant part of the currency inflows, which is our definition of high remittance countries.

Figure 16.10 This figure presents the inflows of remittances and remittances as a share of GDP for the economies that are considered high remittance receiving countries.
The countries presented in this figure are: Bangladesh, Colombia, Ecuador, Jordan, Kyrgystan, Latvia, Lithuania, Macedonia, Nigeria, Pakistan, Romania, Serbia, Sri Lanka, and Ukraine.

Table 16.11 presents the results for the impact of changes in remittances and capital flows on the changes in real effective exchange rates. Consistent with previous results from Asia, Eastern Europe, and Latin America, we see a statistically significant and negative impact from increases in remittance inflows on changes in the real effective exchange rate.

Table 16.11

This Table Presents the Fixed Effects Panel Analysis for Countries With High Remittance Inflows Across Regions Using the Specification Presented in Eqs. 16.1 and 16.2

Fixed effects: time and country
d(REER) (−1)	0.194***	0.194***	0.191***	0.195***	0.195***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
dRemit (% GDP)	−37.48***	−35.83**	−35.74**	−32.81**	−32.80**
	(0.003)	(0.011)	(0.011)	(0.021)	(0.021)
d(Net Kflows) (% GDP)	1.520	1.526		1.522	1.525
	(0.314)	(0.313)		(0.312)	(0.312)
Dummy*d(remit)		−10.67	−10.16	−18.84	−18.87
		(0.768)	(0.779)	(0.603)	(0.603)
Trade				−0.982	−0.987
				(0.311)	(0.309)
TOT				3.148**	3.170**
				(0.029)	(0.029)
Institutions					0.221
					(0.918)
R²	0.199	0.199	0.197	0.209	0.209
Adj. R²	0.062	0.059	0.059	0.067	0.064
N obs.	471	471	471	471	471

The dependent variable is the quarterly difference in real effective exchange rates. There are 14 countries in This analysis: Bangladesh, Colombia, Ecuador, Jordan, Kyrgystan, Latvia, Lithuania, Macedonia, Nigeria, Pakistan, Romania, Serbia, Sri Lanka, and Ukraine. The time frame under study spans from the first quarter of 2005 to the last quarter of 2013. Values in parentheses are p-values. *, **, *** Represent significance at the 10, 5, and 1% levels.

This result implies that increases in remittance inflows as a share of GDP contribute to appreciation of the real effective exchange rate. In contrast, the impact of changes in other capital flows on the real effective exchange rate is not statistically significant. Once more this result supports the idea that under the assumption (as observed from the data) that remittances are a constant inflow of funds into local economies, they play a significant role against depreciation pressures when other forms of capital (FDI and FPI) are exiting the economy.

It is important to note the role of institutions in frontier markets. Throughout this analysis, there is mixed evidence of the statistical significance of institutions on changes in the real effective exchange rate, but the institutional quality in a country is crucial in helping the economies maintain economic stability and cope with sudden or drastic fluctuations in foreign currency flows. Due to the fact that the available data on institutional quality are limited to an annual frequency, at best, and that we are focusing on a limited number of frontier markets, we believe that in fact institution quality is critical even if it is not statistically significant in this analysis, as has been shown in the literature.

4. Conclusions

In this chapter we have discussed the effect of changes in remittances on the changes on real effective exchange rates, and we have concentrated our efforts on showing that remittances are a more stable form of capital entering a local economy than other capital sources (FDI and FPI) and that, as such, they play a very significant role when other sources of funds are exiting the economies. This is particularly relevant for those interested on investing in frontier markets.

We have mentioned the economic and social reasons why remittances behave differently than other forms of capital: remittances enter the economy as an income transfer from one household abroad to another one in the home country. The transfer of funds can be used for consumption, savings, or investment purposes, and the funds usually do not leave the economy. Moreover, senders (immigrants) tend to send a steady flow of remittances and may even increase the amount sent in times of economic downturns in the home country. The familial relations underpinning the flow of remittances make them unique in their reliability as a stable transfer of funds. Motives to send remittances hinge on host country economic conditions that may result in a rise in migrant incomes as well as altruistic and insurance motives to help maintain family ties in the home country. Therefore, factors that may lead to diminishing foreign investment flows do not necessarily have the same impact on remittance flows.

In comparison to other types of capital inflows, remittances have been found to contribute to the stabilization of current account positions, while reducing the volatility of capital flows and output volatility in the home country. This impact on the domestic economy provides improved debt sustainability and better creditworthiness in developing countries, as well as a vital source of foreign currency when other flows dry up.

Our results show that if remittances are a constant inflow of funds into local economies, they play a significant role against depreciation pressures when other forms of capital (FDI and FPI) are exiting the economy, and, as such, remittances have an important effect in analyzing and understanding exchange rate dynamics. These results are consistent across regions and countries and across high and low remittance receiver countries. However, the impact of the changes in remittances is significantly stronger on those high remittance economies.

Acknowledgments

The authors would like to thank Dominick Salvatore and Darryl McLeod for helpful suggestions and help with the data downloading. The usual disclaimers apply.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for Chapter 16: Impact of Remittances on Frontier Markets’ Exchange Rate Stability

Create new playlist

Sign In

Sign Up