Impact of Worker’s Remittances on Kyrgyz Economic Growth1
A. Stylized facts
1. Worker’s remittances have represented by far the most important financing flow in the balance of payments of the Kyrgyz Republic in recent years but they remain quite volatile. Representing on average about 29 percent of GDP over the past five years they have covered a wide part of the large trade and service balance deficit. However, since they are almost entirely originating from Russia2 and widely denominated in rubles, they are quite sensible to Russian economic development. In particular, the marked slowdown that Russia experienced since the end 2013 has resulted in a significant drop in remittances inflows since the beginning of 2015.
Current Account and its Components, 2010–15
Sources: Authorities data and IMF staff calculations.
2. The significant decline in remittances inflows since the beginning of the year seems to be essentially due to the large ruble depreciation, the volume of transfers remaining broadly constant. Indeed, while about 25 percent of remittances originating from Russia are denominated in dollars, the rest is denominated in rubles which have depreciated by about 42 percent since the end 2013. In 2014, the evolution of the ruble has been widely compensated by an increase in the volume of remittance leading to only a marginal decrease of their inflows of about 2 percent. The relative resilience of remittances inflows can probably be explained by the sectoral repartition of Kyrgyz workers in Russia. Indeed most of them work in the sector of trade and services which have been comparatively less affected by the economic slowdown than sectors such as construction which employs a more marginal part of Kyrgyz workers. However since expatriate workers cannot indefinitely compensate for the ruble depreciation, the drop in remittance appears much more marked since the beginning of the year: they have declined by 28 percent by the end of September on a year-on-year basis.
3. The magnitude of remittances coming from Russia makes the evolution of the som highly correlated with the evolution of the ruble but the som recent depreciation has been much less pronounced than the ruble one. The coefficient of correlation between the evolutions of the two currencies vis à vis the dollar reached 90 percent over the past 10 years. In this context, the Kyrgyz republic nominal effective exchange rate has decreased since the end of 2013 (by 5.6 percent by the end of July 2015) albeit in a significantly lesser extent than the Russian one which declined by about 30 percent over the same period. The Kyrgyz Republic real effective exchange rate, on the other hand, has remained more stable since the end of 2013 reflecting only a depreciation of 1.5 percent (versus 18 percent for the ruble). This recent appreciation in real term of the som vis à vis the ruble could harm the country competitiveness at a time when the country just joined the Eurasian Custom union.
Sources: Authorities data and INS.
4. In view of the magnitude of the remittances inflows and their volatility, the long-run impact of remittances on the domestic economy warrants investigation. On the one hand, remittances are expected to alleviate poverty, smooth consumption and allegedly reduce the financial constraints for underserved segments of the population and such favoring small-scale investments projects. On the other hand, beyond being volatile and rather unpredictable, remittances can support excessive private demand during an extended period of time distorting prices and preventing an efficient allocation of labor, undermining the country’s competitiveness and its potential growth.
B. Macroeconomic impact of remittances: some hints from the literature
5. The literature on the macroeconomic impact of remittances is well furnished but remains an open field. In particular, their effects on long term growth remain controversial. On the one hand, remittances may enhance investment in physical capital thus increasing economic growth. The effect is expected to be all the more so big face significant financial constraints that the remittances may ease. Moreover, remittances can also facilitate human capital formation either through higher spending in education or, more indirectly, through improved nutrition and shelter. However, the effects on growth will crucially depend on the economic conditions that prevail in the country and in particular in the share of remittances which will be actually invested rather than consumed. And on the other hand, remittances can, independently on these specific circumstances, have negative effects on growth through a Dutch disease effect. Indeed remittance inflows can generate an appreciation of the real exchange rate that could harm the tradable good sector and the country competitiveness.
6. Empirical findings show little evidence of a long term impact of remittances on growth but support the idea that remittances tend to reduce output volatility. Most cross country and panel analysis find no or at best a mild impact of remittances inflows on growth, conditioned on the degree of financial development and the improvement of business environment. Guilanao and Ruiz-Arranz (2009) for instance, argue, using a panel of about 100 developing countries, that remittances boost investment and long term growth in economies with underdeveloped financial sectors by helping local entrepreneurs circumvent collateral constraint due to inefficient credit market. This result is somewhat confirmed by Chami et al (2008) whose results albeit not supporting any positive impact of remittances on growth highlight the negative interaction between remittances and financial deepening on growth. Controlling for a various set of institutional variables, Catrinescu and others (2006) using both cross-sectional and panel regressions, find a significant, albeit mild, positive relationship between remittances inflows and GDP growth. The impact of remittances on output volatility appears to be more conclusive. Both cross section analysis (see for instance IMF 2005 or Chami et al 2008) and panel regressions (World Bank 2006) find a negative correlation between output volatility and remittances inflows. This result can be explained the desire among migrants to compensate their relative for negative impact of economic fluctuation in the home country enabling the recipient households to smooth consumption. As mentioned in Chami et al (2008), this result could also comes from smoother investment if the remittances flow through the financial system making it easier for them to borrow and hence easier to smooth their investment expenditures.
7. The literature provides a broader evidence of the impact of remittances inflows on the equilibrium real exchange rate. The conventional view that higher remittance receipts is associated with an appreciation of the real exchange rate has been confirmed by Bourdet and Falck (2003) for Cape Verde over the period 1980–2000, by Hyder and Mahbood (2005) for Pakistan over 1978–2005 or by Saadi-Sedik and Petri (2006) for Jordan over 1964–2005. Amuedo-Dorantes and Pozo (2004) found similar results using a panel of 13 Latin American and Caribbean countries over 1978–98. So, if the literature is far from having studied all single country cases, most of the research to date seems to confirm that higher remittances tend to appreciate the real exchange rate. So if this appreciation indeed implies Dutch disease effects, harming the competitiveness of the tradable sector, large remittances inflows could have negative effects on the long term growth. Thus if remittances are not sufficiently oriented toward investment their beneficial short run effects on the recipient country welfare through higher and more stable consumption and poverty reduction could come at the expense of long-term growth.
C. Macroeconomic impact of remittances: an empirical analysis
8. We use a VAR model for our analysis of Kyrgyz’s remittances. We estimate an unrestricted vector autoregression (VAR) in which real GDP, remittances inflows and the real effective exchange rate evolve jointly and endogenously of the form:
Where Yt = [Real GDP growth; Remittance inflow growth in dollar; Real effective exchange rate growth] and all growth rate are month over month of the preceding year. The data span covers the period 2006M03 to 2015M06. Lags criterion criteria being rather inconclusive, we choose to have 2 lags included, striking balance between higher complexity and better fit of the model.3 Our results are displayed in Table 1.
|Real GDP growth||REER growth||Remit. Inflows growth|
9. Our results appear to be in line with the literature on the macroeconomic impact of remittances. In column (1), it is worth noting that as various empirical studies we do not find any evidence of a positive impact of remittances on growth suggesting that these flows are probably not enough oriented toward investment. However, in column (2) our results highlight a positive and very significant impact of remittances inflows on appreciation of the real effective exchange rate in Kyrgyz Republic supporting the idea the higher flows of remittances receipt could generate Dutch disease effects. A careful analysis of the impact of remittances on tradable industries in the manufacturing sector would however be necessary to confirm these preliminary findings.
D. Policy implications and recommendations
10. The results of our empirical analysis suggest that a higher mobilization of remittances flows toward investment would be growth enhancing. One way that governments can use public expenditures to enhance remittances’ growth impact is to improve public infrastructure, both physical infrastructure and public institutions. This recommendation echoes the migration and development literature which finds that poor physical infrastructure and poor governance discourage private investment. And because they typically support household incomes the receipts of remittances exacerbates the negative impact of poor infrastructure. The more remittances a country receives, the higher the quality of the country’s public investment must be in order to induce a given amount of investment from individuals. According to the last World Economic Forum’s Global competitiveness report government instability, corruption, policy instability and inefficiency government bureaucracy are majors impediments for doing business in the Kyrgyz Republic. Working on these institutional weaknesses would help orienting remittances toward private investment. Moreover the same report notes that the Kyrgyz Republic tends to lag behind other CIS countries in terms of infrastructure: in this context the launch of the large PIP program, if well implemented, appears welcome.
11. With regard to tax policy, empirical studies suggest that a shift toward consumption based taxation may provide the correct incentive to increase private savings and investments. On the other hand, it is not recommended to tax directly remittances that could cause a decline of remittances activity and induce a migration of the flow from formal to informal channel (Chami et al 2008). If the Kyrgyz Republic already taxes consumption through VAT, increasing its coverage through the reduction of the informal economy would help create savings incentives. According to Abdih and Medina (2013), the informal economy represented about 26 percent of the Kyrgyz GDP in 2008 and was mainly driven by the regulatory burden in financial and products markets4. In this context, it would be beneficial to increase the availability and affordability of financial services while rising local competition and reducing regulations burdensome in doing business.
12. In particular, increasing the formality of transfers is important for leveraging remittances. It can make the process more easily monitored potentially make access to finance more effective. Ultimately the securitization of remittances inflows could be considered: it would be an efficient way to combine relatively small amounts of personal savings into larger flow of capital to finance public investment. Such diaspora bonds have already been used to finance public investment in several countries such as India, Israel, Sri Lanka, Ethiopia, or Zimbabwe. Moreover, improving statistics on remitters will help in the analysis and policy making decision process.
13. Finally, governments can mitigate the effects of the real exchange rate appreciation linked to large remittances inflows by implementing reforms that favor the manufacturing sector competitiveness. Increased spending in education, reforms to make the labor market more flexible for instance should improve labor productivity and enhance external competitiveness. The last World Economic Forum global competitiveness report points in particular an inadequately educated workforce as a significant impediment of doing business in the Kyrgyz Republic.
AbdihY.MedinaL.2013. Measuring the informal economy in the Caucasus and Central Asia. IMF working paper13/137.
Amuedo-DorantesC.PozoS.2004. Workers’ remittances and the real exchange rate: a paradox of gifts. World development vol. 32 pp. 1407–17.
BourdetY.FalckH.2003. Emigrants, remittances and Dutch disease in Cape Verde. Kristiansad Univerty College working paper No. 11.
ChamiR.BarajasA.CosimanoT.FullenkampC.GapenM.MontielP.2008. Macroeconomic consequences of remittancesIMF Occasional Paper 259.
CatrinescuN.Leòn-LedesmaM.PirachaM.QuillinB.2006. Remittances, institutions and economic growth. IZA Discussion Paper No 2139.
GiulianoP.Ruiz-ArranzM.2005. Remittances, financial development and growth. IMF working paper05/234.
HyderZMahboobA.2005. Equilibrium real effective exchange rate misalignment in Pakistan. State Bank of Pakistan.
IMF2005. “Two current issues facing developing countries” in World Economic Outlook April 2015: Globalization and external imbalances.
Saadi-SedikT.Petri.M.2006. To smooth or not to smooth: the impact of grants and remittances on the equilibrium real exchange rate in Jordan. IMF working paper06/257.
Prepared by Veronique Salins.
About 97 percent of remittances inflows in the Kyrgyz Republic come from the Russian Federation.
However, it is worth noting that we obtain similar results using alternatively one lag or three.
These regulatory burden can be defined as burdensome regulations in the product markets, in the form of procedures for starting business, registering property and dealing with construction permits, as well as difficulties in the credit market (such as availability and affordability of financial services). On the other hand, nay legislation aiming at increasing local competition, and reducing monopolies and the extent of market dominance would contribute to reduce the size of the informal economy.