| Using an unbalanced panel data that contains rich information about manufacturing firms for the period of 2000-2016,this study adds empirical evidence about the superior performance of exporters relative to non-exporters and causal effects of exporting on firms’ performances in developing economies.I employed the semiparametric technique developed by Ackerberg,Caves and Frazer(2015)to account for the endogeneity and identification problems in estimating elasticities of a production function in calculating firm level total factor productivity(TFP).Separate production functions for each ISIC industrial classification has been estimated using Non-linear heteroskedastic GMM estimator on the value added of firms.The type of production function exhibited by each ISIC category has been identified and using the elasticity of each production function TFP at firm level has been computed.A comparison of TFP growth across different growth plans(SDPRS,PASDEP,GTPI and GTPⅡ)has been made for each industrial classification.Positive and continuous TFP growth rate was observed for Food processing including Beverage production,Textiles&Apparel but there was negative TFP growth rate in the Leather&Footwear.In most other manufacturing activities,the growth of TFP across different plan periods was volatile and inconsistent.Furthermore,in this study the dynamics of productivity and growth in value added of manufacturing firms has been carried out as part of the validity test for Kaldor Second growth Law(Verdoorn’s law)and found Kaldor’s Second growth Law has been valid in all manufacturing except the Leather and Footwear industries.The implication of these findings is that there is increasing returns to scale in the manufacturing sector with the exception of the Leather and Footwear industries.The empirical evidence is in support of the argument that manufacturing plays as an engine of growth.However,there are many challenges faced by manufacturing firms which has affected their performances explained by the fact that most of the firms are producing not at their full capacity.The main problems firms faced are lack of raw material supply,lack of market/clients for the locally produced goods associated to weaker streamlined marketing strategy,shortages of working capital due to lack of enough credit from banks to manufacturing firms and repeated breakage of machineries.There is widespread empirical evidence indicating that exporters outperform non-exporters worldwide.However,there are only few empirical studies on the exporting-productivity links in Ethiopia.In fact the proposition that export trade increases performance of domestic firms has not been straight forward.There are two different but not mutually exclusive arguments associated to the relationship between exporting and productivity;the self-selection hypothesis and learning by exporting hypothesis.I have analyzed the self-selection and the post-entry effects with respect to various firms’ characteristics such as TFP growth,output growth and employment growth.In analyzing the self-selection into exporting,I applied a linear probability model with fixed effect and probit models of panel data technique.I have also estimated a dynamic version of random probit model using Wooldridge(2005)initial condition when considering the sunk costs of exporting on the export decision of firms.The econometric estimation results confirmed that sunk costs are found to be the main driving forces of the exporting activity in Ethiopian manufacturing industry as evidenced by a statistically significant effect of lagged export status in model estimated.It can be concluded that the estimated results are consistent to the previous findings that productive firms self-select in the export activity.To put in different words,it has been well evidenced that the self-selection hypothesis holds.The implication of these empirical findings is that it is essential to encourage firms through providing skill enhancing facilities to workers which increases their productivity level and thereby enhance the export activity.Furthermore,addressing constrain factors related to raw material supply shortage,provision of credit as well as reducing sunk costs of exporting are vital to boost the export sector.In analyzing the post-entry effects,I applied the propensity score matching jointly with the differences-in-differences estimator which disentangles the causal effect of exporting on performances(productivity,employment and outputs)of exporting firms from the self-selection effect.Technically to effectively reduce the effect of the bias that comes to existence due to sample selection effect,randomizing the sample has been found crucial.One way to randomize the sample selection has been to match the sample of exporters and non-exporters using the propensity score matching technique in which the following variables were used as repressors in the logit model of matching process:firm productivity(TFP),employment,real wage and real capital per worker.By estimating the difference-in-differences model for the matched data,it has been found that the causal relationship from exporting to TFP growth rate is not in support of the learning by exporting hypothesis.More specifically,although the sign of the coefficient of the export dummy is positive,it is not statistically significant even at 10 percent level of significance implying that there is no empirical evidence of learning by exporting effect in Ethiopian manufactured products exporting firms.This findings contradicts earlier studies of Gebreeyesus.et al(2009)and Esmale(2013)that concluded in support of presence of learning by exporting effects.As a part of the robustness check,I estimated the relationship between the subsequent firms’performances and export participation by using the system GMM estimator of dynamic panel data analysis framework.The estimated result also confirmed the proposition of the superior performance of the exporters relative to non-exporters.However,the superiority is due to market selection mechanism while the learning from exporting effect is negligible in the Ethiopian manufacturing firms. |
PDF Full Text Request