Do Workers' Remittances Bring Economic Growth to Receiving Countries?

1 Introduction

Besides foreign direct investment (FDI) and capital market flows, workers’ remittances are another external channel for capital flows. According to the OECD, remittances to developing countries amounted to US$ 149.4 billion in the year 2002. However, whereas FDI and capital market flows are subject to variation due to recessions in home countries, remittances are steadily rising every year (OECD, 2006), reaching an amount of about UD$ 300 billion in the year 2007 (Barajas et al., 2009). To give a brief definition, remittances are money transfers from migrants working abroad to their families in their home countries. Yet, the question is, do these remittances contribute to or boost economic growth in receiving countries or are they only a means to increase the migrants’ families’ welfare by directly reducing their poverty and raising the living standard (Rao and Hassan, 2011). In other words, are remittances mostly used for consumption or do they rather flow in education, and thereby contribute to the human capital, and in investments, thus increasing the capital stock in the economy (Giuliano and Ruiz-Arranz, 2009)? From the growth theory we know that consumption does not have any impact on growth, only investments, either in production or in human capital, can affect long-run growth. Evidence from Indonesia, Ecuador, and Argentina (Sayan, 2006) shows that remittances indirectly reduce volatility of growth of output in times of crises and increase the growth rate thereby (Rao and Hassan, 2011). In contrast, Sayan (2006) found that remittances are moving procyclically with out in recipient countries, boosting incomes during booms, but reducing them even more during recessions and thus magnifying the economic crisis.

This paper examines the relationship between remittances and GDP growth using in a macro panel with 67 countries and a time period of 28 years, from 1975 through 2002, as well as a cross-section analysis for comparison. The goal of this analysis is to determine whether, and to what extent, remittances have an impact on long-term economic growth and, if so, whether this relation is significant or not. The paper is structured as follows. Section 2 briefly gives an overview of the theoretical framework of the growth theory. Section 3 presents and describes the data. Section 4 provides the empirical analyses, consisting of a cross-sectional and panel analysis, and presents the results of these. Section 5 concludes.

2 Idea of the Growth Theory

The following section should provide a short overview of the neoclassical growth theory and especially the exogenous long-run growth model developed by Robert Solow (1956).

The real income is given by T(t), which is partly spent for consumption and partly saved and invested. Every period a constant fraction s of output is saved, such that the savings rate is sT(t). The capital stock is K(t) and net investment I equals aggregate savings:

[illustration not visible in this excerpt] (1)

Investment raises the capital stock every period, at the same time, a constant share ð C (0; 1)

of capital depreciates. Thus, the law of motion of capital is:

illustration not visible in this excerpt

[illustration not visible in this excerpt] (2)

Furthermore, the production function is assumed to have constant returns to scale, using

[illustration not visible in this excerpt] (3)

where A is the labour-augmenting productivity, i.e. after a while more things can be done, and

illustration not visible in this excerpt

grows at a constant rate: [illustration not visible in this excerpt]

. Since population grows exogenously at rate n, the growth

A A

rate of raw labour L is n as well. Hence the effective labour, AL, is growing at rate [illustration not visible in this excerpt].

illustration not visible in this excerpt

Let [illustration not visible in this excerpt]

illustration not visible in this excerpt

AL

be capital per effective unit of labour, and dividing (2) by AL on both sides, we

get:

illustration not visible in this excerpt

K

illustration not visible in this excerpt

AL

Deriving (4) with respect to K/AL and plugging back into (4), we get the growth rate of

capital per unit of effective labour:

illustration not visible in this excerpt

[illustration not visible in this excerpt] (5)

illustration not visible in this excerpt

As income per capita is given by T/L = Aƒ(k), income per capita grows at rate gA. Similarly, consumption and capital per capita grow at rate gA. Total income T = ALƒ(k), as well as total consumption and total capital stock is growing at a rate gA+ n.

All in all, it could be shown that changes in investment rate s or in the population growth rate n affect the long-run level of output per worker. Consumption does not contribute

to the economy’s long-run growth, thus if remittances are supposed have an impact on growth, they will have to be invested to a large part and not spent for consumption.

3 Data

The dataset consists of 67 developing countries1, which I have chosen on the basis of available data from the sample used by Giuliano and Ruiz-Arranz (2009), for the period 1975

– 2002. The dependent variable is GDP per capita growth (in annual %). For the depending variable I use the measure Workers’ Remittances and Compensation of Employees in % of GDP . Both variables are taken from WDI. The remainder of the control variables is also taken from the WDI, if not otherwise indicated, and is the following:
- Inflation – as annual percentage change in consumer price index
- Openness 2 – obtaining from the sum of imports and exports as percentage of GDP
- Population growth
- Investments – as ratio of gross fixed capital formation to GDP
- Government fiscal balance – cash surplus/deficit as percentage of GDP
- Human capital 3 – as percentage of total population aged 15 and over with a complete second level

illustration not visible in this excerpt

Table 1 reports the descriptive statistics of all variables used in this analysis. For the extremes I added the country and the year in parentheses to get a better idea of some country characteristics.

illustration not visible in this excerpt

1 see Appendix for the whole sample of countries
2 Data only for imports and exports in % of GDP from WDI
3 Data taken from the Barro - Lee (2000) series. It was available for the period 1960 – 2000 only in 5-year intervals, therefore I took over the data for each following 4-year gap, respectively, i.e. the years 1975 – 1979 have the same data, and so on.

[...]