Journal Issue

Research Summaries: Migration: An Attractive Insurance Option in African Countries?

International Monetary Fund. Research Dept.
Published Date:
June 2015
  • ShareShare
Show Summary Details

Ahmat Jidoud

Over the last two decades, officially recorded remittances (migrants’ transfers) toward African countries have reached substantial amounts. This paper assesses whether these flows help to cushion the impact of macroeconomic shocks in these countries and explores the channels through which these flows affect macroeconomic volatility. The empirical results show that remittances significantly reduce output and consumption volatilities by absorbing a significant amount of GDP shocks, but their effect on consumption fluctuations is less pronounced. A small, open–economy model augmented with remittances shows that the stabilizing effects of remittances are higher in economies where (i) remittances induce no wealth effect on labor supply, and (ii) help with promoting financial development by lessening financial frictions.

There has been widespread empirical work on the impact of remittances on African countries, both from a microeconomic and macroeconomic perspective in recent years. Although significantly underestimated, officially recorded remittance flows to Africa have increased from $9.1 billion in 1990 to nearly $40 billion in 2010, outweighing other financial flows (see Figure 1a). In addition, remittances tend to be more stable and persistent than other financial flows (see Figure 1b). An extensive empirical literature has recognized the microeconomic benefits of these financial flows, while a growing number of studies have looked at their macroeconomic effects. More precisely, recent research has extensively examined the impact of migrants’ transfers on growth (Adam Jr. and Page 2005; Gupta, Pattillo, and Wagh 2009). This research found that both international migration and remittances significantly reduce the level, depth, and severity of poverty in developing countries. Remittances reduce poverty through their positive impact on financial development. Remittances give financially constrained households access to credit markets by collateralizing assets they build using their remittances. Hence, they can contribute to an increase in aggregate investment and savings. Another widely explored topic in this literature is the capacity of remittances to reduce macroeconomic volatility. Remittances are thus viewed as automatic stabilizers which buffer shocks and smooth aggregate fluctuations. This insurance role of remittances and migration hinges on the countercyclicality of these financial flows with respect to recipient countries’ fundamentals and their resilience to economic developments within the originating countries.

Figure 1a:Remittances and Other Financial Flows in SSA

Figure 1b:Coefficient of Variation

Source: Author’s calculations based on African Development Indicators (2011).

Note: ODA = Official Development Aid.

The present paper fits into the second strand of the literature as it assesses the ability of remittances to stabilize business cycle volatility in a sample of African countries with available relevant data. The paper takes a double approach. First, the paper empirically evaluates the correlation between aggregate fluctuations (output and private consumption volatilities) and the size of remittances as percentage of GDP. From this perspective, our analysis follows the existing work on developing countries; nonetheless, this paper focuses only on African countries. Second, the paper probes the theoretical mechanisms through which the stabilizing or destabilizing effects of remittances spread into the economy. Although the existing literature has brought evidence of the smoothing impact of remittances in recipient countries (Chami, Hakura, and Montiel 2009; Combes and Ebeke 2011, among others), it has been silent about the theoretical mechanisms at work. This paper intends to fill this gap. Understanding these mechanisms is critical for policymakers for designing appropriate policies to effectively manage these financial flows. This theoretical exercise is the main contribution of the paper that most clearly distinguishes the analysis from similar work, in addition to tentative policy conclusions that may be derived.

Do Remittances Help Stabilize Macroeconomic Fluctuations? How Much of Shocks Are Absorbed by Remittances?

In responding to the first question, our working paper follows the existing work in assessing the correlation between business cycle volatility and remittances by estimating the semi-elasticity of volatility (measured by the standard deviation of GDP or consumption) on cross-section data. The sample comprises 27 African countries with yearly data covering 1980–2005 all from the African Development Indicators (2011) from the World Bank. Data are averaged over this period. The model controls for other relevant determinants of volatility identified in the literature such as per capita income, financial development, government size, development aid, and terms of trade volatility. Standard endogeneity tests (Correlation and Nakamura and Nakamura tests) rule out the endogeneity of remittances in the model. The model is estimated using generalized least squares method to account for potential heteroskedasticity in the residuals. The empirical results show that remittances reduce both output and consumption fluctuations. Specifically, an increase in workers’ remittances-to-GDP of one percentage point reduces the log of the standard deviation of per real capita GDP growth by 3.2 percent while the volatility of real per capita consumption decreases by two percent. This small stabilizing effect on consumption seems rather counterintuitive, but at least two arguments can be put forward to explain it. First, households’ consumption in developing economies encompasses both consumption of durables and non-durables. It might be that remittances actually reduce the consumption instability, but only of the non-durable component of aggregate consumption. If this is the case, the inability to disentangle these two components makes it harder to observe a significant effect on aggregate consumption fluctuations. Second, surveys at microeconomic levels in some African countries have brought evidence that more than half of households receiving remittances from outside Africa are in the top two consumption quintiles (Mohapatra and Ratha 2011), minimizing the consumption smoothing effect of remittances may, since these households have access to alternative means (e.g., credit markets), help to smooth out consumption shocks.

Our analysis also quantifies how much of GDP shocks are absorbed by remittances, that is the risk-sharing through migration using a very simple econometric panel framework designed by Asdrubali, Sorensen, and Yosha (1996). The results suggest that around 4 to 22 percent of GDP shocks are smoothed through remittances channel. In a context where remittances data are severely underestimated due to informal channels, misreporting or misclassification in other revenues (e.g., non resident’s deposits), the stabilizing virtues of remittances are potentially underestimated as well.

How Do Remittances Affect Macroeconomic Volatility? A DSGE Model Approach

We present a small open economy dynamic stochastic general equilibrium (DSGE) model to provide a theory consistent with the preceding empirical results. In particular, we probe into the channels through which remittances affect macroeconomic fluctuations. For this purpose we used a small open economy a la Schmitt-Grohe and Uribe (2003) enriched with financial frictions which play a significant role in driving aggregate fluctuations, in particular, the excess volatility of consumption with respect to output. The model is calibrated to replicate the average economy formed in the sample of the economies. The quantitative experiments suggest that the stabilization benefits of remittances depend on two key factors: (i) the size of the wealth effect of remittances on labor supply and (ii) the ability of remittances to promote financial development. The first mechanism is highlighted by comparing the predictions of the model with two types of utility specifications: Greenwood, Hercovitz, and Huffman (1988), or GHH, which induce a zero wealth effect and King, Ploser and Rebelo (1988), or KPR, with a negative wealth effect on labor supply. This channel implies that the impact of remittances on macroeconomic fluctuations is higher in countries where the labor supply does not decrease following an increase in response to an increase in remittances (no wealth effect). In this case, the elasticity of labor supply is constant and irresponsive to changes in households’ income, implying a moderate response of output and consumption in response to shocks. However, in an environment where migrants’ transfers induce a negative wealth effect on the recipients’ labor supply, an increase in remittances deepens the business cycle as macroeconomic variables respond strongly to exogenous shocks. In order to highlight the second channel, we assume that remittances could reduce households’ borrowing costs through a lower risk-premium. This mechanism suggests remittances are more stabilizing in countries with a shallow financial sector while remittances are channeled through this sector. The reason is that remittances help credit-rationed borrowers to access financial services as they build physical assets which they can use as collateral. This finding underscores the need for recipient countries to promote financial development so as to channel remittances through the financial system and to help the poor have greater access to credit.


Other Resources Citing This Publication