Abstract
Are the forces of market selection at work in Africa? How successful are markets in these economies in sorting out firms on an efficiency basis following the sequence of reforms to liberalize and particularly to transform some of the previous command economies to market oriented ones? What is the pattern of entry and exit in the manufacturing sector and how does it affect industry productivity growth? This study examines these issues using firm-level industrial census data from the Ethiopian manufacturing sector. It is the first attempt to analyze firm turnover and productivity differentials using industrial census data in sub-Saharan Africa. The Ethiopian manufacturing sector exhibits a high firm turnover rate that declines with size. Exit is particularly high among new entrants; 60% exit within the first 3 years in business. Our study consistently shows a significant difference in productivity across different groups of firms, which is reflected in a turnover pattern where the less productive exit while firms with better productivity survive. We also found higher aggregate productivity growth over the sample period, mainly driven by firm turnover.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The exchange rate was fixed in the period 1975–1991 and according the official exchange rate of the Birr versus US Dollar at that time (2.08 Birr/1USD) the ceiling on private investment was roughly about a quarter of a million USD.
The Central Statistics Agency collects data annually on manufacturing firms with 10 and above employment customarily identifying them as Medium and Large firms and some times Formal Manufacturing Sector.
Between 1994 and 2002 GDP per capita grew with an average growth rate of about 2.6%. The service sector share to GDP in terms of value added increased from 35% to 48% while the agriculture shrank from 55% to 40% in the same period. The industrial sector share remained almost constant around 11% (source WDI 2004 but own calculation).
The capital stock is calculated as \(K_{it}=K_{it-1} +\frac{I_t }{p^{t}}-\delta K_{it-1} -sK_{it},\) where K it−1 denotes the beginning year capital, p t investment deflator, δ depreciation rate, and sK it sold assets in year t.
The labor quality index is constructed as follows. We identify the average wage of the production and non-production workers by firm. Then we divide the average wage of non-production workers to average wage of production workers in each firm and aggregate by year, \(\sum\nolimits_{i=1}^N {\frac{WNP_i}{WP_i}},\) where WNP i represents average wage of non-production workers at firm i and WP i average wage of production workers at firm i. The number of non-production workers is then multiplied by the year averages of this ratio to construct quality index for non-production workers. We follow the same approach to construct the quality index for seasonal workers since they are found to have on average lower wage than the permanent production workers (i.e. the number of seasonal workers is multiplied by \(\sum\nolimits_{i=1}^N {\frac{WP_i }{WS_i}}\) at each year, where WS i represents average wage of seasonal workers at firm i.
A firm entering the data base might be due to either expansion of employment to 10 or more persons or “green field” investment. At the same time the firm exit from the data could be due to either shutdown or contraction of employment to less than 10 persons. Our data does not identify whether the entry is due to “green field” investment or expansion or whether the exit is due to shutdown or contraction. The exit record from contraction might bias the exit rate, and this is expected to be particularly pronounced for small firms that employ a number of persons around the cut-off point, 10 employees.
For further comparison see Table A1 in the Appendix.
One alternative view is the capital vintage effect that new firms acquiring better technology are more efficient than old firms, thus the probability of exit is higher among older firms.
The simultaneity bias arises when the firms’ knowledge of their own productivity levels affects their choice of inputs, thus the unobserved fixed effect is correlated with the observed inputs. The selection bias on the other hand arises because firm exit is not exogenous since smaller firms with less capital intensity are more likely to exit.
When x it is endogenous (i.e. x it is correlated with v it and earlier shocks), lagged values dated (t − 2) and earlier will be valid additional instruments. When x it is predetermined (i.e. x it and v it are not correlated but x it might be correlated with v it-1 and earlier shocks), lagged values dated (t − 1) and earlier will be valid instruments in the first differenced equation. If x it is strictly exogenous then the complete time series of x it will be valid instruments in each of the first differenced equations in addition to the dependent variable (t − 2) and earlier instruments. These relations are easily testable using standard GMM tests of over-identifying restrictions: the Sargan–Hansen test and the Difference-Sargan test.
If the error terms v it are correlated over time, then the GMM estimator is inconsistent. Thus, for the error term to be serially uncorrelated, the serial correlation of the differenced residual should be first-order, but not second-order. This is also testable with the null of no second-order serial correlation in the first-differenced equations.
Olley and Pakes (1996) proposed a semi-parametric approach using observable micro information, for example investment, as a proxy to controls for the part of the error correlated with inputs. Levinsohn and Petrin (2003) extended this approach by introducing the possibility of using intermediate inputs as a proxy rather than investment. Ackerberg and Caves (2003) and Bond and Söderbom (2004) criticized the proxy method, on the basis of problems of identifying the parameters.
In estimating the system-GMM for industry production functions we used xtabond2 in Stata 9. Unlike the Sargan statistics which is a minimized value of the one-step GMM criterion, the Sargan–Hansen statistic reported by xtabond2 is the minimized value of the two-step GMM criterion function, and is robust to heteroskedasticity or autocorrelation.
The TFP for the whole manufacturing sector is constructed from the sector specific models coefficients of the production function estimation.
The effect of exit and entry on productivity growth will be discussed in detail in Sect. 7.
A successful entrant here is defined as a firm that entered the data set after 1996 and remained in the data set until the end of the sample period i.e. 2003.
This categorical variable is time invariant in the sense that those firms exiting in any year before the end of the sample period are treated equally and assigned a value equal to one for all periods they are present in the data set.
The robust variance estimate adjusts for within-firm correlation where the observations within cluster (firm) may not be treated as independent, but the clusters themselves are independent. This is because the dependent variable, exit-decision, is time invariant dummy where as, the explanatory variables are allowed to vary by year, thus could result in serial correlation in the residuals and downward biased standard errors.
The whole manufacturing productivity can also be aggregated in the same manner using each industry output share as weights.
Note that there are other modified versions of this decomposing method. For example Foster et al. (2001) separate the within and between effects from the cross/covariance effect.
The definition here is a bit different from the previous definition in Sect. 3, since in this current section we are dealing with year by year turnover and productivity change. Liu and Tybout (1996) use a similar approach based on annual data. This decomposition method is basically that of Baily et al. (1992), except our data is annual and theirs was 5-year interval.
The average productivity growth for 14 industries and the aggregate manufacturing sector is calculated from the year by year productivity growth. For brevity we only report the averages but not the year by year.
References
Ackerberg DA, Caves K (2003) Structural identification of production functions. Department of Economics, UCLA, Los Angeles mimeo
Arellano M, Bond S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev Econ Stud 58:277–297
Aw BY, Chen X, Roberts MJ (1997) Firm-level evidence on productivity differentials, turnover, and exports in Taiwanese manufacturing. NBER Working Paper No. W6235
Baily MN, Hulten C, Campbell D (1992) Productivity dynamics in manufacturing plants. Brookings papers on economic activity: microeconomics, 187–267
Bartlesman E, Doms M (2000) Understanding productivity: lessons from longitudinal micro databases. J Econ Lit 38(3)
Bartelsman E, Haltiwanger J, Scarpetta S (2004) Microeconomic evidence of creative destruction in industrial and developing countries. Tinbergen Institute Discussion Paper 04–114/3, Tinbergen Institute
Blundell R, Bond S (1998) Initial conditions and moment restrictions in dynamic panel data models. J Econometr 87:115–143
Bond S, Söderbom M (2004) Adjustment costs and the identification of Cobb Douglas production functions. Centre for Studies of African Economies, Oxford University
Central Statistical Authority of Ethiopia (CSA) (1990) “The survey of manufacturing and electricity industries” CSA, Addis Ababa
Dunne T, Roberts MJ, Samuelson L (1989) The growth and failure of US manufacturing plants. Quart J Econ 104:671–698
Foster L, Haltiwanger J, Krizan CJ (2001) Aggregate productivity growth: lessons from microeconomic evidence. In: Dean E, Harper M, Hulten C (eds) New developments in productivity analysis. University of Chicago Press, Chicago
Griliches Z, Regev H (1995) Firm productivity in Israeli industry. J Econometr 65:175–203
Grilliches Z, Mairesse J (1995) Production functions: the search for identification. NBER Working Paper No, W5067
Gunning JW, Mengistae T (2001) Determinants of African manufacturing investment: the microeconomic evidence. J Afr Econ 10(AERC supplement 2):48–80
Hahn CH (2000) Entry, exit, and aggregate productivity growth: micro evidence on Korean manufacturing. OECD Economics Department Working Paper 45
Harding A, Söderbom M, Teal F (2004) Survival and success among African manufacturing firms. Centre for studies of African Economies, CSAE Working paper 2004–2005, Oxford University
Hopenhayn HA (1992) Entry, exit and firm dynamics in long-run equilibrium. Econometrica 60:1127–1150
Jovanovic B (1982) Selection and the evolution of industry. Econometrica 50:649–670
Levinsohn J, Petrin A (2003) Estimating production functions using inputs to control for unobservables. Rev Econ Stud 70:317–342
Liu L (1993) Entry-exit, learning and productivity change: evidence from Chile. J Dev Econ 42:217–242
Liu L, Tybout JR (1996) Productivity growth in Colombia and Chile: panel-based evidence on the role of entry, exit and learning. In: Roberts M, Tybout J (eds) Industrial evolution in developing countries: micro patterns of turnover, productivity and market structure. Oxford University press, New York
Mairesse J, Hall BH (1996) Estimating the Productivity of Research and Development in French and US Manufacturing Firms: an Exploration of Simultaneity Issues with GMM Methods. In: Wagner K, Van Ark B (eds) International Productivity Differences and Their Explanations. Elsevier Science, pp 285–315
Olley GS, Pakes A (1996) The dynamics of productivity in the telecommunications equipment industry. Econometrica 64:1263–1297
Roberts MJ, Tybout JR (1996) A preview of country studies. In: Roberts M, Tybout J (eds) Industrial evolution in developing countries: micro patterns of turnover, productivity and market structure. Oxford University press, New York
Tybout JR (2000) Manufacturing firms in developing countries: how well do they do, and why? J Econ Lit 38:11–44
Windmeijer F (2000) A finite sample correction for the variance of linear two-step GMM estimators. Working paper 00/19, Institute of Fiscal Studies, London
World Bank (2004) World development indicators 2004, CD-Rom, Washington, DC
Acknowledgments
The author would like to thank Arne Bigsten, Lennart Hjalmarsson, Måns Söderbom, Francis Teal, two anonymous referees and the editor for insightful comments. I am also grateful for comments from seminar participants at the Department of Economics, Göteborg University.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendices
Rights and permissions
About this article
Cite this article
Gebreeyesus, M. Firm turnover and productivity differentials in Ethiopian manufacturing. J Prod Anal 29, 113–129 (2008). https://doi.org/10.1007/s11123-007-0076-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11123-007-0076-0