Firm productivity: The Role of Competition and of the Initial Firm Efficiency. Evidence from the Czech Republic

la Cour, Lisbeth; Ionascu, Delia Working Paper Firm productivity: The Role of Competition and of the Initial Firm Efficiency. Evidence from the Czech Republic Working paper, No. 9-2007 Provided in Cooperation with: Department of Economics, Copenhagen Business School (CBS) Suggested Citation: la Cour, Lisbeth; Ionascu, Delia (2007) : Firm productivity: The Role of Competition and of the Initial Firm Efficiency. Evidence from the Czech Republic, Working paper, No. 9-2007, Copenhagen Business School (CBS), Department of Economics, Frederiksberg, https://hdl.handle.net/10398/7653 This Version is available at: https://hdl.handle.net/10419/208541 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/ Department of Economics Copenhagen Business School Working paper 9-2007 FIRM PRODUCTIVITY: the Role of Competition and of the Initial Firm Efficiency. Evidence from the Czech Republic. Delia Ionascu Lisbeth la Cour ____________________________________________________ Department of Economics -Porcelænshaven 16A, 1.fl. - DK-2000 Frederiksberg Firm Productivity: the Role of Competition and of the Initial Firm Efficiency. Evidence from the Czech Republic° Delia Ionaşcu Lisbeth la Cour* Copenhagen Business School Abstract: It has been argued that the effect of competition on a company’s incentive to innovate and to reduce managerial slack depends on the initial level of efficiency. For example, while firms close to the technology frontier invest more in innovation if competition increases, backward firms reduce innovation. On a panel data of Czech companies, for the years 1993-2005, we empirically assess the impact of increased competition on firm productivity and the importance of the initial firm efficiency level. We depart from the empirical literature on emerging markets by taking into account both domestic and foreign competition. In line with the theory, our results show that there is an inverted U-relationship between domestic competition and firm productivity. Our results also confirm that trade liberalization has a positive impact on productivity. However, the effect is less significant if domestic competition is not taken into account. In addition, we find that both domestic and foreign competition have an effect on productivity in companies close to the technology frontier but not in backward companies. JEL: D24, F10 Keywords: Firm productivity; trade liberalization; competition; initial productivity. ° This paper has benefited from extensive discussions and feedback from Pascalis Raimondos-Møller. We gratefully acknowledge his help. Also we thank Steffen Andersen, Jan Hanousek, Lubomir Lizal, Stephan Juraida, and Anders Sørensen for valuable comments and insights. We also thank the Danish Social Science Research Council for financial support. * Corresponding author: Lisbeth Funding la Cour, Copenhagen Business School, Department of Economics, Porcelænshaven 16A, 2000 Frederiksberg, Denmark; Tel.: +45 3815 2488; Fax:+45 38 15 25 76; E-mail address: [email protected] 1 1. Introduction After the collapse of the communist regimes in 1989 in Central and Eastern Europe, the countries in the region opened up their economies to foreign trade and foreign direct investment (FDI), and introduced reforms to encourage domestic competition. Among other benefits, reforms that promote competition have been expected to lead to higher productivity. The bulk of theoretical and empirical research backs up this view, pointing for example to the disciplinant effect of competition and to benefits steaming from improved access to intermediate goods (Romer 1994). Yet, there are also theoretical arguments which suggest that in certain circumstances increased competition has a negative impact on productivity. For example, in industries characterized by increasing returns to scale, increased competition might force domestic firms to scale down production and hence to move to higher average costs. In addition, competition might decrease the expected gains from innovation, and thus the innovation level, in companies that are far away from the technology frontier as these companies expect to loose market share to more efficient entrants (Boone 2000 and Aghion et al 2005b). Also an increase in competition might exacerbate the managerial slack problem if managers are highly responsive to monetary incentives (Scharfstein 1988). The slack problem also worsens if owners’ benefits from a marginal increase in efficiency decrease with the number of competitors as could be the case in Cournot competition (Martin 1993). All these theoretical results seem to indicate that the actual impact of increased competition on firm productivity dynamics might be ambiguous and, furthermore, that it might be context dependent. Whether the context plays a role in determining the actual impact of increased competition on firm productivity is an empirical question. Results emerging from the empirical literature emphasize the role played by policies and institutional aspects (see Winters 2004 for a survey). Less studied is the contribution that the industrial context has to this impact. In particular, the fact that the initial firm efficiency level affects innovation and therefore the firm productivity in a context of increased competition (Boone 2000, and Aghion et al 2005b) has been addressed by few empirical studies (Bernard et al 2006, and Konings and Vandenbussche 2007). The studies that have addressed this issue focused on the impact of a specific reform – trade liberalization – on firm productivity and the role played by the firm efficiency level. However, especially in emerging economies, trade reforms that promote openness are typically accompanied by other reforms that promote competition in general. Therefore, to assess the overall impact on competition on firm productivity one has to consider both domestic and import competition. In addition, if other reforms are not taken into 2 account one might attribute too much of changes in productivity to foreign competition. This paper analyzes the impact of competition on firm productivity, with a special focus on the degree by which the initial firm efficiency affects the relation between increased competition, on the one hand, and firm productivity, on the other hand. The analysis is done using a panel data of Czech firms for the period 1993-2005. In the Czech Republic’s transition process, fast and comprehensive trade and FDI liberalization has been accompanied by other reforms that have spurred competition (e.g. the reform of the financial system, the introduction of bankruptcy laws). As already indicated, we depart from the empirical literature on emerging markets by taking into account both domestic and foreign competition. We find that trade liberalization has a positive impact on firm productivity. Yet, this effect weakens with the increase in the initial level of tariff protection. Our results also show that an increase in market concentration induces higher productivity in markets in which domestic competition is tough. The latter result is consistent with the theoretical literature that suggests that both innovation and managerial efficiency lead to an inverted U-shape relationship between productivity and competition. With respect to the role of the initial level of cost efficiency for the impact of competition on firm productivity, we find that the above results regarding the impact of competition, both domestic and foreign, are not present in companies that are far away from the technological frontier, where there is practically no effect of competition on firm productivity. Furthermore, we assessed the extent to which the absence of a control for domestic competition biases the estimated effect of trade liberalization on firm productivity. We find that the effect of trade liberalization on firm productivity is seriously understated if a control for domestic competition is absent. Moreover, the fact that the initial firm efficiency affects the impact of trade liberalization is not anymore fully confirmed in these regressions. Thus, our results indicate that to assess the extent to which trade liberalization affects firm productivity one needs to control for domestic competition and to take into account the initial firm efficiency levels. The paper proceeds in the following way. Section 2 reviews the theoretical research that links competition and firm productivity and the existent empirical work. Section 3 describes the data we use and the methodology on which we base our empirical analysis. Section 4 presents 3 the empirical results and the results of several robustness checks. Section 5 concludes. 2. Competition and productivity. Theory and empirical results. That increased competition in general has an effect on firm performance is already well established in the theoretical literature, though the sign of this effect might be ambiguous.1 One of the perceived benefits from increased competition stem from the effect competition has on managerial slack. In companies in which managers have more information than owners about productivity shocks and own effort, if monopoly rents exist, managers can capture part of them in the form of slack. Yet, faced with higher competition, managers have to increase their effort to fulfil targets specified by incentive schemes (Hart 1983, and Scharfstein 1988). In addition, since unobserved productivity shocks are likely to be correlated across firms, higher competition from more efficient companies increases owners’ opportunities to compare or to assess the actual performance of their companies, and thus to design sharper incentive schemes (Nalebuff and Stiglitz 1983, Hermalin 1992, Holmstrom 1982, Meyer and Vickers 1995, and Nickell 1994). However, designing incentive schemes that induce high performance might become too costly when competition is strong as the additional effort necessary to deliver a good rather than a bad performance increases with competition, hence increasing managers’ incentives to underreport their productivity. Therefore, the effect competition has on slack might be non-linear, of a U-shape (Scharfstein 1988, Hermalin’s 1992, Meyer and Vickers’s 1997, among others). The fact that competition might have a U- shape relation with managerial slack has been backed up by Schmidt (1997), who showed that competition not only raises the probability of liquidation, but also reduces profits. Therefore when competition becomes too intense, managerial effort might in fact decrease with further increase in competition. Another benefit from increased competition steams from its effect on innovation, as a monopolist tends to “rest on his laurels” (Arrow 1962) or might get trapped in bureaucratic structures (Schumpeter 1934). Yet, Schumpeter (1942) has noticed that most of the innovation is done in big firms that have the necessary resources to invest and are able to accrue the associated benefits due to their position in the market. Therefore, much as the impact of competition on effort, the effect of competition on innovation could have an inverted U- 1 For instance, see Winters (2004) for the impact of openness on firm performance. Also see Djankov and Murrell (2002) for a survey of literature for transition economies. 4 shape.2 There is also an emerging literature on the role played by the initial level of firm efficiency in defining the actual impact of competition on firm productivity (Aghion et al 2005b, and Boone 2000). Thus, the threat of entry encourages incumbent advanced firms to invest in innovation in order to retain their market but discourages innovation in firms far from frontier as those companies expect, under any circumstances, to loose markets to more efficient entrants (Aghion et al 2005b, and Boone 2000). These results suggest that the relationship between competition and innovation (thus, firm productivity) is influenced by the initial firm productivity. To sum-up, theoretical results indicate that the relation between competition and firm productivity might have an inverted U-shape (either due to the U-shape relationship between competition and managerial slack and/or due to the inverted U-relationship between competition and innovation). Furthermore, the above results suggest that this relationship might be different in firms that are close to the technology frontier than in backward firms. Turning to the empirical literature, several empirical results support the theoretical conjecture of a U-shape impact of competition on managerial slack (Green and Mayes 1991) and of an inverted U-shape relationship between innovation and competition (Scherer 1967, and Aghion et al 2005a). In addition, the empirical literature has often addressed these relations indirectly, by searching for a relation, usually linear, between competition and firm productivity. Thus Nickell et al (1992) and Nickell (1996) found that an increase in import competition had no impact on firm productivity in UK manufacturing. Yet, an increase in market share had a negative effect on the level of productivity, probably due to the negative effect that an increased in monopoly power has on managerial and workers’ effort. More generally, decreases in trade costs also spur competition, and to that account there are numerous papers that look at the effect of openness on productivity. Substantial evidences show that changes in openness lead to a reallocation of resources and market shares from inefficient to efficient firms, and therefore had a positive impact on the evolution of industry productivity (for a review see Tybout 2003). There are also evidences that trade cost 2 Aghion et al (2005a)’s theoretical model confirms this fact. 5 reductions enhance productivity at the plant level both in developing and developed countries (see Pavcnik’s 2002, Bessonova et al 2003, or Sabirianova et al 2005a for studies on emerging economies, and Lawrence 2000 and Bernard et al 2006b for studies on the US). Thus, apparently, empirical analyses that consider only one aspect of competition – trade cost reductions – yield different conclusions regarding the impact of trade liberalization on productivity than those papers that consider both foreign and domestic competition. As indicated by the theoretical results, this might be due to differences in the context in which trade liberalization has taken place (e.g. the initial level of firm efficiency in the industry) but could also be due to the fact that in the absence of a control for changes in domestic competition induced by other market reforms that typically (although not always) accompany trade reforms, the coefficient of trade liberalization picks up their effects. This is one aspect that we investigate in our empirical analysis. As the impact of competition on firm productivity seems to be contextual dependent, several studies have tried to unveil the underlying conditions that favour a positive relation between increased competition and firm productivity. Emerging results emphasize the role played by policies and institutional aspects (see Winters 2004 for a survey) or by industrial aspects. Among the studies that have focused on the latter factors, results show that a positive and a stronger relationship between competition and firm productivity is likely to exist in highly concentrated industries (MacDonald’s 1994), in low-skill intensive industries (Lawrence 2000), or in non-multinational companies (Bernard et al 2006). Also, some empirical studies address the role the initial efficiency level plays in moderating the impact of competition on firm productivity, as suggested by the theoretical studies of Boone (2000) and Aghion et al (2005b). Aghion et al (2005b) show that entry liberalization lead to a rising inequality in the regulated manufacturing sector in India, and moreover, that productivity increased by more in industries that were close to the Indian productivity frontier. Sabirianova et al (2005) find that increased foreign presence in Czech and Russian industries lead to a rise in the efficiency of foreign firms, which are assumed to be at the outset closer to the technological frontier, but had a negative effect on productive efficiency of domestic firms, which are less efficient. Konings and Vandenbussche (2007) find that a decrease in competition due to antidumping protection helps more laggard EU companies than cost efficient ones. Topalova (2004)’s results, however, show that trade liberalization did 6 not lead to divergence in productivity within Indian industries as similar productivity improvements can be noticed in firms with both high and low productivity prior to the trade reform. As well, Bernard et al (2006) find no evidence that the impact of trade costs reduction on productivity was different for firms with different productivities in US manufacturing. Thus, there are few but mixed evidences of a differential impact of competition on firm productivity according to the initial efficiency level. The present study analyzes the impact of competition on firm productivity, and compares the response to competition of firms that are closer with those that are farther away from the technological frontier. Among the studies we have mentioned above, only three directly tackle the discrepancy in response to competition of firms with different initial efficiency: Bernard et al’s (2006), Topalova (2004), and Konings and Vandenbussche (2007). These papers, however, look only at the effect of trade cost reductions on firm productivity. Given the inferences we drawn from Nickell et al (1992) and Nickell’s (1996) results, unlike these studies, we look at the impact of both domestic and outside competition on firm productivity. This allows us to assess in this study the extent to which the absence of a control for domestic competition biases the effect of trade liberalization on firm productivity and the conjectures regarding the differing impact of trade liberalization on productivity with respect to the initial firm efficiency. 3. The empirical methodology and data description 3.1. Methodology We do our empirical analysis in two steps. First, we estimate firm productivity. Second, we study the impact of competition both on all firms and on sub-samples that have at the outset high/low productivity. 3.1.1. Estimating firm productivity We assume a Cobb-Douglas production function itititlitkit lky υ+ω+β + β+β= 0 (1) where i is the firm index, t is the time index, y is log of output, k is log of capital, l is log of labour, and the residual term is decomposed into a time varying productivity shock, ω, that is not observed to the econometrician, and a white noise, υ. To estimate this production function we use the semiparametric approach developed by Olley and Pakes (1996) that allows us to obtain a time-varying measure of plant productivity that accounts for the simultaneity bias. If 7 Table 3. OLS and Olley and Pakes estimates of production function OLS Olley and Pakes Ind. Log Empl. Log Capital Returns to scale No. Obs. R2Log Empl. Log Capital Returns to Scale 15 0.547*** 0.428*** 0.975 2837 0.75 0.501*** 0.510*** 1.011 17 0.639*** 0.290*** 0.929 832 0.83 0.562*** 0.230*** 0.792 18 0.705*** 0.203*** 0.908 267 0.87 0.687*** 0.261*** 0.948 19 0.693*** 0.196*** 0.889 156 0.79 0.729*** 0.204*** 0.933 20 0.617*** 0.355*** 0.972 654 0.79 0.564*** 0.207*** 0.771 21 0.446*** 0.486*** 0.932 315 0.81 0.448*** 0.201*** 0.649 22 0.546*** 0.282*** 0.828 525 0.73 0.518*** 0.158*** 0.676 24 0.523*** 0.420*** 0.943 786 0.84 0.457*** 0.377*** 0.834 25 0.538*** 0.379*** 0.917 1123 0.84 0.499*** 0.390*** 0.889 26 0.500*** 0.466*** 0.966 1294 0.81 0.487*** 0.292*** 0.779 27 0.617*** 0.327*** 0.944 630 0.85 0.551*** 0.442*** 0.993 28 0.582*** 0.307*** 0.889 2467 0.75 0.549*** 0.341*** 0.89 29 0.675*** 0.213*** 0.888 2557 0.79 0.617*** 0.237*** 0.854 31 0.677*** 0.266*** 0.943 1061 0.8 0.643*** 0.300*** 0.943 32 0.579*** 0.272*** 0.851 284 0.74 0.558*** 0.561*** 1.119 33 0.649*** 0.188*** 0.837 442 0.75 0.623*** 0.248*** 0.871 34 0.668*** 0.334*** 1.002 595 0.83 0.626*** 0.327*** 0.953 35 0.667*** 0.184*** 0.851 299 0.78 0.702*** 0.213*** 0.915 36 0.652*** 0.322*** 0.974 940 0.86 0.609*** 0.306*** 0.915 37 0.451*** 0.309*** 0.76 181 0.56 0.372*** 0.355*** 0.727 Industries: 15-Food products and beverages; 17-Textiles; 18-Wearing apparel; 19-Leather manufacturing; 20- Wood and wood and cork products, except furniture; 21-Pulp, paper and paper products; 22-Publishing and printing; 24-Chemicals and chemical products; 25-Rubber and plastic products; 26-Non-metallic mineral products; 27-Basic metals; 28-Fabricated metal products, except machinery and equipment; 29-Machinery and equipment n.e.c.; 31-Electrical machinery and apparatus n.e.c.; 32-Radio, television and communication equipment and apparatus; 33-Medical, precision and optical instruments; 34-Motor vehicles, trailers; 35-Other transport equipment; 36-Furniture; manufacturing n.e.c.; 37-Recycling Figure 1. The marginal effect of a change in competition on firm productivity 510 15 20 25 -0.04 -0.03 -0.02 -0.01 0.01 0.1 0.2 0.3 0.4 0.5 0.6 -6 -4 -2 2 tariff Herfindal (a) The marginal effect of an increase in tariff (b) The marginal effect of an increase in on firm productivity domestic competition on firm productivity 14 Regarding the impact of an increase in domestic competition, its effect has been negative in markets with already tough competition but has been positive in concentrated markets. This result is in line with theoretical results that point to an inverted U-shape relationship between competition and firm productivity. Given that at the level of industry aggregation that we have in our data, the markets are highly competitive (as indicated by the sample mean of 0.013 for the Herfindal index and by its 0.039 standard deviation), in most of the industries the impact of an increase in domestic competition on firm productivity has been positive (see Figure 1, (b)). There are, however, examples of the opposite effects as well, mostly among subindustries of textile manufacturing (ISIC 1700) and in the manufacture of aircraft and spacecraft (ISIC 3530). Estimates of equation (6) show that companies with different levels of efficiency respond differently to trade liberalization and to an increase in domestic competition, as all the interaction terms between competition and the level of firm inefficiency are highly significant (see Table 4, column (2)). Moreover, we can see from Table 4, column (2), that for highly inefficient firms (firms with a dit-1 closed to 1), both the first and the second order effects of a tariff reduction and of a decrease in competition on firm productivity decrease substantially. This indicates that the productivity of laggard companies does not to respond to changes in competition. However, frontier companies are affected by both, domestic and foreign competition (these effects should be close to the effects inferred from competition measures that are not interacted with distance as the highly efficient companies have a dit-1 closed to 0). The fact that changes in competition levels have different impacts for frontier than for laggard companies is further confirmed by our estimations of equation (5) on subsamples of efficient and inefficient companies (see Table 4, columns (3) and (4)). Thus, our results are in line with the predictions of the theoretical models developed by Boone (2000) and Aghion et al (2005b). Regarding the marginal effects of an increase in competition on firm productivity, for the highly productive firms, the positive relation between a decrease in tariff and firm productivity is less likely to hold now (Table 4, column (4)) than in the case when we estimate this effect using the entire sample (Table 4, column (1)). The opposite is true, however, for the positive impact of an increase in competition on firm productivity. 15 Table 4. The impact of competition on firm productivity8 (1) (2) (3) (4) All firms All firms Low Prod High Prod lag Tariff rates -0.041*** -0.080*** -0.036 -0.076** (3.80) (4.69) (1.16) (2.68) square 0.001*** 0.003*** 0.001 0.003** (3.43) (4.67) (1.09) (2.53) * lagged inefficiency (dit-1) 0.062** (2.02) square * lagged inefficiency (dit-1) -0.002** (2.13) lag Herfindal 2.433** 5.457*** -0.250 5.161** (2.04) (4.09) (0.15) (2.43) square -6.720** -11.248*** -3.474 -9.697** (2.42) (3.54) (0.80) (2.30) * lagged inefficiency (dit-1) -5.437*** (3.09) square * lagged inefficiency (dit-1) 9.274*** (2.74) lagged inefficiency (dit-1) -1.177*** (8.15) Constant 0.059 0.620*** -0.401** 0.571*** (0.98) (6.40) (2.57) (4.31) Observations 7158 6988 725 1096 R-squared 0.04 0.34 0.17 0.23 t-statistic in parentheses; *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. Robust errors adjusted for clustering at 3-digit level in regressions (1) and (2) and to two digit level, due to smaller number of observations, in regressions (3) and (4). Year and industry dummies at 2 digit level. Base year: 1996 We have further estimated the models (5) and (6) using only the level of tariff protection as a competition measure to see if in the absence of a control for domestic competition, the effect of trade liberalization on firm productivity is biased. The results are given in Table (5). When comparing these results with the ones in Table 4 we see that in the former estimates the impact of trade liberalization is underestimated (see results in columns (1) in both tables). Yet, the importance of the initial firm efficiency is seriously downplayed, the effects of the interaction terms between trade liberalization and the level of firm inefficiency being insignificant in Table 5, column (2), and having coefficients much closer to zero and lower 8 We got the same results when using industry dummies at 3 digit level in regressions (1)-(4), or when interacting industries dummies with the productivity distance in regressions (3) and (4). The results in regressions (5)-(8) do not change when we define low (high) productive companies as being the ones with a productivity distance less (higher) than 0.1 (0.9). We have experimented with dummies for exporting industries, for companies that exit the market, and with variables that differentiate between companies that produce goods within different industries at NACE 4-digit. Since none of these variables were significant, we have dropped them. We have also normalized the competition measures on industrial output rather than on the sum of industrial output and imports. The results are similar with the ones in the table above. 16 significance in regressions run on the subsample of highly productive firms (Table 5, column (4)). Table 5. The impact of trade liberalization on firm productivity (without domestic competition) (1) (2) (3) (4) All All Low High lag Tariff rates -0.009** -0.016** 0.005 -0.019* (2.28) (2.30) (0.65) (1.95) square 0.000* 0.000*** -0.000 0.000** (1.90) (2.65) (0.15) (2.36) * lagged inefficiency (dit-1) 0.009 (1.07) square * lagged inefficiency (dit-1) -0.000 (1.25) lagged inefficiency (dit-1) -0.943*** (14.45) Constant -0.038 0.394*** -0.569*** 0.399*** (0.97) (7.27) (13.84) (7.24) Observations 12742 12338 1370 2029 R-squared 0.03 0.29 0.12 0.14 t-statistic in parentheses; *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. Robust errors adjusted for clustering at 3-digit level in regressions (1) and (2) and to two digit level, due to smaller number of observations, in regressions (3) and (4). Year and industry dummies at 2 digit level. Base year: 1996 In general our results are robust to the two alternative measures of domestic competition we use, Herfindal and Mkt concentration, as the results in Appendix 2, Table 2.1 closely match the results in Table 4, the only difference being that although the square term of market concentration remains negative, it looses its significance. Also, the results in Appendix 2, Tables 2.2 and 2.3 show that using OLS and CRS_OP productivities we get the same results, with slightly less significance in the coefficients for the latter case. 5. Conclusions In this paper we have studied the impact of competition on firm performance in the Czech Republic. We have found that laggard and frontier firms respond differently to an increase in domestic competition and to trade liberalization. Firms that are close to the technological frontier benefit from trade liberalization. Also, they are affected by changes in domestic competition: an increase in competition has a positive impact on firm productivity in concentrated markets but has the opposite effect on firms with tough competition. We found 17 no effect of trade liberalization or competition on firm productivity in laggard companies. The results also indicate a non-linear effect from competition to productivity and are in line with the prediction of the theoretical models developed by Boone (2000) and Aghion et al (2005b). Furthermore, our results indicate that in the absence of a control for domestic competition, the impact of trade liberalization on firm productivity is understated. Meanwhile, in this case the role played by the initial firm efficiency in determining the effect of an increase in competition on firm productivity is seriously downplayed. References Ackerberg, D., L. Benkard, S. Berry, and A. Pakes (2005): Econometric Tools for Analyzing Market Outcomes, to appear in the Handbook of Econometrics Volume 6, J.J. Heckman. Aghion, P., N. Bloom, R. Blundell, R. Griffith, and P. Howitt (2005a): Competition and Innovation: An Inverted U Relationship, Quarterly Journal of Economics 120(2): 701-728 Aghion, P., R. Burgess, S.J. Redding, and F. Zilibotti (2005b): Entry Liberalization and Inequality in Industrial Performance, Journal of the European Economic Association 3(2-3): 291-302. Amiti, M. and J. Konings (2007): Trade Liberalization, Intermediate Inputs and Productivity: Evidence from Indonesia, American Economic Review, forthcoming. Arrow, K. (1962): Economic Welfare and the Allocation of Resources for Inventions, in The Rate and Direction of Inventive Activity, ed. R. Nelson, Princeton University Press. Bernard, A.B., J.B. Jensen and P.K. Schott (2006): Trade Costs, Firms and Productivity, Journal of Monetary Economics 53(5): 917-937. Bessonova, E., K. Kozlov and K. Yudaeva (2003): Trade Liberalization, Foreign Direct Investment, and Productivity of Russian Firms, CEFIR Working Papers w0035. Blundell, R. and S. Bond (1998): Initial Conditions and Moment Restrictions in Dynamic Panel Data Models, Journal of Econometrics 87: 115-143. Blundell, R. and S. Bond (2000): GMM Estimantion with Persistent Panel Data: An Application to Production Functions, Econometric Reviews 19(3). Boone, J. (2000): Competitive Pressure: The Effects on Investments in Product and Process Innovation, RAND Journal of Economics 31(3): 549-569. Djankov, S. and P. Murell (2002): Enterprise Restructuring in Transition: A Quantitative Survey, Journal of Economic Literature 40: 739-792. Good, D.H., M.I. Nadiri and R.C. Sickles (1996): “Index Number and Factor Demand Approaches to the Estimation of Productivity,” NBER Working Paper 5790. Green, A. and D. Mayes (1991): Technical Inefficiency in Manufacturing Industries, The Economic Journal 101: 523-538. Hart, O.D. (1983): The Market Mechanism as an Incentive Scheme, The Bell Journal of Economics 14: 366-382. 18 Hermalin, B.E. (1992): The Effects of Competition on Executive Behavior, RAND Journal of Economics 23(3): 350-365. Keller, W. (2004): International Technology Diffusion, Journal of Economic Literature 42(3): 752-782. Konings, J. and H. Vandenbussche (2007): Antidumping Protection and Productivity of Domestic Firms: A Firm Level Analysis, mimeo. Lawrence, R.Z. (2000): Does a Kick in the Pants Get You Going or Does it Just Hurt? The Impact of International Competition on Technological Change in US Manufacturing, in Feenstra, R.E. ed., The Impact of International Trade on Wages, University of Chicago Press for the National Bureau of Economic Research, pp. 197–224. Levinsohn, J. and A. Petrin (2003): Estimating Production Functions Using Inputs to Control for Unobservables, Review of Economic Studies 70(2): 317-341. MacDonald, J.M. (1994): Does import competition force efficient production? Review of Economics and Statistics 76(4): 721–727. Martin, S. (1993): Endogenous Firm Efficiency in a Cournot Principal-Agent Model, Journal of Economic Theory 59: 445-450. Meyer, M.A. and J. Vickers (1997): Performance Comparisons and Dynamic Incentives, Journal of Political Economy 105(3): 547-81. Nalebuff, B.J. and J.E. Stiglitz (1983): Information, Competition, and Markets, American Economic Review 73: 278-283. Nickell, S.J. (1996): Competition and Corporate Performance, Journal of Political Economy 104(4): 724-746. Nickell, S.J. (1994): Competition and Corporate Performance, Discussion Paper 182, London School of Economics, Centre for Economic Performance. Nickell, S.J., S.B. Wadhwani and M. Wall (1992): Productivity Growth in U.K. Companies, 1975-1986, European Economic Review 36: 1055-1085. Olley, G.S. and A. Pakes (1996): The Dynamics of Productivity in the Telecommunications Equipment Industry, Econometrica 64(6): 1263-1297. Pavcnik, N. (2002): Trade Liberalization, Exit, and Productivity Improvements: Evidence from Chilean Plants, Review of Economic Studies 69(1): 245-276. Romer, P. (1994): New Goods, Old Theory and the Welfare Cost of Trade Restrictions, Journal of Development Economics 43(1): 5-38. Sabirianova, K., J. Svenjar and K. Terrell (2005): Distance to the Efficiency Frontier and FDI Spillovers, Journal of the European Economic Association 3(2-3): 576-586. Scharfstein (1988): Product-market competition and managerial slack, RAND Journal of Economics 19(1): 147-155. Scherer, F. (1967): Market Structure and the Employment of Scientists and Engineers, American Economic Review 57: 524-531. Schmidt, K.M. (1997): Managerial Incentives and Product Market Competition, Review of Economic Studies 64: 191-213. Schumpeter, J. (1934): The Theory of Economic Development, Cambridge, MA: Harvard University Press. 19 Schumpeter, J. (1942): Capitalism, Socialism and Democracy, New York: Harper. Topalova, P. (2004): Trade Liberalization and Firm Productivity: the Case of India, IMF Working Papers 04/28. Tybout, J. (2003): Plant- and Firm-level Evidence on ‘New’ Trade Theories, in Choi, E.K., Harrigan, J. (Eds.), Handbook of International Economics, Basil-Blackwell, Oxford. Winters, L Alan (2004): Trade Liberalization and Economic Performance: An Overview, The Economic Journal 114(2): F4-F21. 20 Appendix 1 – Data description Firm level data are from Amadeus. Amadeus is a pan-European commercial database, provided by Bureau van Dijk, which contains financial information on public and private companies. We used data from all versions of the Amadeus database since 1996 with information on medium and large firms. Most of the Czech firms included in the database produce goods in several industries at 4 digit NACE level. We have classified firms according to their main activity. We did the following modifications to the data: i. we excluded all companies that had less than 10 employees: ii. we excluded firms with non-positive investment levels when estimating firm productivity iii. since we did not have enough observations in three industries at 2-digit NACE level (16 – manufacture of tobacco, 23 – manufacture of coke, refined petroleum and nuclear fuel, 30 – manufacture of office machinery and computers) to estimate the production function we dropped companies from this sector. iv. we dropped 2 observations to exclude firms with market shares higher than 100. v. we dropped 6 observations to exclude firms with a productivity index less than -5 as they looked to be outliers (see Figure A1.4). Table A1.1. Variables Variable Definition y (log of output) Added value deflated by the producer price index (PPI). For most of the industries, we have the PPI at 3-digit NACE; for the remaining we have used PPI at 2-digit NACE.9 Sources: Added value is from Amadeus and Aspekt; PPI from the Czech Statistical Office. Coverage: 1993-2005 k (capital) Tangible fixed assets deflated by the price index for gross fixed capital formation, at a slightly more aggregated level than 2-digit NACE. Sources: Tangible fixed assets are from Amadeus and Aspekt; price index for gross fixed capital formation from AMECO. Coverage: 1993-2005 l (log of labour) Number of employees. Sources: Amadeus and Aspekt Coverage: 1994-2005 i (investment) Computed as ititit kki )1( 1 δ − − = +, where δ = 15%. Coverage: 1993-2004 Tariff rate Weighted average (by trade value) of effectively applied rates, taking into consideration applicable (and available) preferential duties. Source: WTO Coverage: 1994-2004 Herfindal Number of companies in a 4-digit ISIC industry Sources: UNIDO via Campus Solutions. Coverage: 1995-2003 9Mairesse and Jaumandreu (2005) show that deflating value added with PPI rather than a firm specific price index leads to very similar estimates of the coefficients in the production function. 21 Mkt concentration The ratio of the sales of the 4 companies with the biggest sales and the industrial output, at 4-digit ISIC level. Sources: Firm sales are from Amadeus and Aspekt; industrial output is from UNIDO via Campus Solutions. Coverage: 1995-2003 Inefficiency (dit) = (Max(productivity)-productivity) / (Max(productivity)- Min(productivity) at 4-digit NACE level Table A1.2. Descriptive statistics – observations based on which firm productivity is estimated Variable Mean Std. Dev. Min Max Observations y overall 6.197 1.335 0.118 12.359 N = 19940 between 1.312 0.442 11.880 n = 5338 within 0.389 0.342 9.088 T-bar = 3.73548 k overall 6.265 1.770 0.041 12.991 N = 19940 between 1.881 0.041 12.665 n = 5338 within 0.351 1.402 10.161 T-bar = 3.73548 l overall 5.224 1.153 2.303 10.129 N = 19940 between 1.180 2.303 10.005 n = 5338 within 0.233 2.445 7.413 T-bar = 3.73548 i overall 0.505 2.826 0.000004 125.106 N = 19940 between 1.594 0.000 74.944 n = 5338 within 1.386 -45.009 87.663 T-bar = 3.73548 OP productivity overall 1.190 0.873 -5.069 5.283 N = 19940 between 0.816 -3.217 4.949 n = 5338 within 0.367 -4.478 3.982 T-bar = 3.73548 Table A1.3a. Descriptive statistics – observations based on which the impact of competition on firm productivity is estimated Variable Mean Std. Dev. Min Max Observations lav_real overall 6.354 1.249 0.118 11.748 N = 7158 between 1.277 0.118 11.343 n = 2249 within 0.296 1.495 8.458 T-bar = 3.18275 k overall 6.483 1.550 0.445 12.192 N = 7158 between 1.668 0.445 12.109 n = 2249 within 0.254 3.740 8.187 T-bar = 3.18275 l overall 5.347 1.063 2.303 9.842 N = 7158 between 1.114 2.303 9.842 n = 2249 within 0.209 3.384 6.822 T-bar = 3.18275 i overall 0.421 1.725 0.000 60.730 N = 7158 between 1.289 0.000 35.192 n = 2249 within 0.957 -22.484 31.855 T-bar = 3.18275 OP productivity overall 1.163 0.791 -4.211 5.283 N = 7158 between 0.745 -4.211 5.070 n = 2249 within 0.284 -3.283 3.190 T-bar = 3.18275 Tariff rate overall 6.702 4.657 0.000 27.530 N = 7158 between 4.403 0.005 27.530 n = 2249 within 1.028 -7.674 27.036 T-bar = 3.18275 Herfindal overall 0.010 0.027 0.000 0.626 N = 7158 between 0.022 0.000 0.360 n = 2249 within 0.017 -0.308 0.564 T-bar = 3.18275 Mkt concentration overall 0.126 0.107 0.001 1.006 N = 7158 between 0.099 0.001 0.952 n = 2249 within 0.053 -0.336 0.909 T-bar = 3.18275 22 Table A1.3b. Descriptive statistics – low productivity firms in 1995 Variable Mean Std. Dev. Min Max Observations lav_real overall 6.245 1.091 2.010 9.303 N = 725 between 1.091 3.527 9.060 n = 158 within 0.408 1.385 8.036 T-bar = 4.58861 k overall 6.747 1.163 2.670 10.933 N = 725 between 1.246 3.947 10.933 n = 158 within 0.253 5.167 8.327 T-bar = 4.58861 l overall 5.460 0.959 2.996 9.842 N = 725 between 0.993 2.996 9.842 n = 158 within 0.207 4.401 6.309 T-bar = 4.58861 i overall 0.312 0.578 0.000 5.558 N = 725 between 0.701 0.004 5.558 n = 158 within 0.307 -1.128 3.153 T-bar = 4.58861 OP productivity overall 0.916 0.774 -2.971 2.976 N = 725 between 0.707 -1.525 2.761 n = 158 within 0.378 -3.529 2.458 T-bar = 4.58861 Tariff rate overall 6.162 3.821 0.010 27.530 N = 725 between 3.231 0.010 21.035 n = 158 within 1.052 1.272 13.432 T-bar = 4.58861 Herfindal overall 0.013 0.039 0.000 0.416 N = 725 between 0.036 0.000 0.302 n = 158 within 0.023 -0.168 0.245 T-bar = 4.58861 Mkt concentration overall 0.123 0.124 0.003 1.006 N = 725 between 0.125 0.008 0.833 n = 158 within 0.066 -0.327 0.648 T-bar = 4.58861 Table A1.3c. Descriptive statistics – high productivity firms in 1995 Variable Mean Std. Dev. Min Max Observations lav_real overall 7.110 1.250 4.041 11.748 N = 1096 between 1.247 4.370 11.343 n = 211 within 0.257 5.760 8.231 T-bar = 5.19431 k overall 7.130 1.593 1.977 12.192 N = 1096 between 1.646 2.785 12.109 n = 211 within 0.266 5.428 8.478 T-bar = 5.19431 l overall 5.869 1.065 2.996 9.782 N = 1096 between 1.094 2.996 9.615 n = 211 within 0.224 4.977 6.631 T-bar = 5.19431 i overall 0.807 2.946 0.000 60.730 N = 1096 between 2.749 0.002 35.192 n = 211 within 1.613 -22.098 26.345 T-bar = 5.19431 OP productivity overall 1.458 0.830 -0.977 4.274 N = 1096 between 0.757 -0.283 3.840 n = 211 within 0.260 0.150 2.464 T-bar = 5.19431 Tariff rate overall 7.261 4.719 0.000 27.530 N = 1096 between 4.130 0.005 21.635 n = 211 within 1.281 1.366 21.329 T-bar = 5.19431 Herfindal overall 0.014 0.036 0.000 0.626 N = 1096 between 0.022 0.000 0.164 n = 211 within 0.029 -0.106 0.567 T-bar = 5.19431 Mkt concentration overall 0.144 0.130 0.005 1.006 N = 1096 between 0.108 0.007 0.633 n = 211 within 0.073 -0.132 0.926 T-bar = 5.19431 23 Table A2.3. The impact of competition of firm productivity using CRS_OP productivity (1) (2) (3) (4) (5) (6) (7) (8) All firms All firms All firms All firms Low Prod Low Prod High Prod High Prod lag Tariff rates -0.016* -0.015* -0.053*** -0.049*** -0.018 -0.019 -0.059*** -0.057*** (1.87) (1.77) (3.29) (3.11) (0.88) (0.89) (4.75) (4.26) square 0.001* 0.000 0.002*** 0.002*** 0.001 0.001 0.002*** 0.002*** (1.70) (1.60) (3.54) (3.26) (0.75) (0.77) (4.74) (4.30) * lagged inefficiency 0.062** 0.056** (dit-1) (2.37) (2.18) square *lagged -0.002*** -0.002** inefficiency (dit-1) (2.70) (2.43) lag Herfindal 1.123 2.220** -0.422 4.293 (1.56) (2.28) (0.31) (1.66) square -4.408** -6.951** -0.892 -20.721 (2.20) (2.39) (0.25) (1.66) * lagged inefficiency -2.773* (dit-1) (1.74) square *lagged 6.475 inefficiency (dit-1) (1.58) lag Mkt concentration 0.508** 1.033*** -0.201 0.477 (2.34) (2.96) (0.61) (0.83) square -0.756 -1.299** 0.066 -0.080 (1.66) (2.18) (0.14) (0.09) * lagged inefficiency -1.188** (dit-1) (1.99) square *lagged 1.291 inefficiency (dit-1) (1.21) lagged inefficiency -1.119*** -1.010*** (dit-1) (8.03) (6.76) Constant -0.090* -0.132** 0.449*** 0.355*** -0.497*** -0.474*** 0.454*** 0.410*** (1.68) (2.11) (4.43) (3.11) (3.61) (3.20) (5.89) (3.83) Observations 7156 7156 6985 6985 669 669 1052 1052 R-squared 0.03 0.03 0.31 0.31 0.18 0.18 0.15 0.15 t-statistic in parentheses; *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. Robust errors adjusted for clustering at 3-digit level in regressions (1)-(4) and to two digit level, due to smaller number of observations, in regressions (5)-(8). Year and industry dummies at 2 digit level. Base year: 1996 30