摘要：We explore the effect of remittances on the output per worker in Lithuania over the sample period 1980–2012. We use the augmented Solow approach and the ARDL bounds procedure to examine the cointegration relationship and subsequently estimate short-run and long-run effects, and the causality nexus. The data available in the World Bank database for the required variable are relatively small for the analysis. To overcome this constraint, we plot data of the respective variables and use appropriate trend functions to approximate data for missing years. For remittances, we use the exponential trend function as the best fit to approximate data for remittances over the period of 1980–1992 and 2012. Similar approach is used to approximate data for GDP (at current prices) for the period of 1980–1989 and gross fixed capital formation for the period of 1980–1994 to build the capital stock data. Further, we use the polynomial trend function to approximate data for GDP (constant 2005 USD) for the period of 1980–1994. Within the necessary caveats, our results show that output per worker, capital per worker and remittances are cointegrated, and remittances contribute about 0,02 % and 0,04 % to output per worker in short-run and long-run, respectively, and the capital share is around 0.50. We note a bidirectional causation and hence a mutually reinforcing effect between capital per worker and output per worker, and a unidirectional causation from remittances to output per worker, duly supporting the remittances-led growth hypothesis in Lithuania. However, our results are not unambiguous due to data constraints. In this regard, further research will supplement and give more insights to the results and outcomes of this paper.
其他摘要：We explore the effect of remittances on the output per worker in Lithuania over the sample period 1980–2012. We use the augmented Solow approach and the ARDL bounds procedure to examine the cointegration relationship and subsequently estimate short-run and long-run effects, and the causality nexus. The data available in the World Bank database for the required variable are relatively small for the analysis. To overcome this constraint, we plot data of the respective variables and use appropriate trend functions to approximate data for missing years. For remittances, we use the exponential trend function as the best fit to approximate data for remittances over the period of 1980–1992 and 2012. Similar approach is used to approximate data for GDP (at current prices) for the period of 1980–1989 and gross fixed capital formation for the period of 1980–1994 to build the capital stock data. Further, we use the polynomial trend function to approximate data for GDP (constant 2005 USD) for the period of 1980–1994. Within the necessary caveats, our results show that output per worker, capital per worker and remittances are cointegrated, and remittances contribute about 0,02 % and 0,04 % to output per worker in short-run and long-run, respectively, and the capital share is around 0.50. We note a bidirectional causation and hence a mutually reinforcing effect between capital per worker and output per worker, and a unidirectional causation from remittances to output per worker, duly supporting the remittances-led growth hypothesis in Lithuania. However, our results are not unambiguous due to data constraints. In this regard, further research will supplement and give more insights to the results and outcomes of this paper. DOI: http://dx.doi.org/10.5755/j01.ee.25.3.6421
关键词：remittances;capital per worker;cointegration;the ARDL approach;the Granger causality tests;economic growth;Lithuania
其他关键词：remittances;capital per worker;cointegration;the ARDL approach;the Granger causality tests;economic growth;Lithuania
