• Nem Talált Eredményt

While the previous subsections have shown some interesting relationships between global- ization variables and markups, all these variables are interconnected in many ways, hence it is important to analyze models with all of these variables included (Table 10).

Let us start with discussing the joint explanatory power of these variables. As the base model in column (1) shows, 2-digit industry-country effects, market concentration and firm size and age explain 14.5 percent of the markup variation. Adding all the globalization variables and market share to it (Column 3) explains about 4 additional percentage points.

Importantly, however, the explanatory power of TFP and market share is much larger than that of the globalization variables. If one adds only these variables to the base model, the explanatory power increases to 0.28 (column 2). Adding the globalization variables to this model increase the explanatory power by only 1 percentage point. However, as we have already discussed, TFP and markups may be spuriously correlated, hence the explanatory power of TFP is overestimated in this calculation, while that of globalization variables may be underestimated.

Turning our attention to the coefficient estimates suggests that most of our results are robust when all of the globalization variables are included. In particular, exporting, import- ing, FDI-making, quality certification, and new IP generation are all positively and—both in an economic and statistical sense—significantly correlated with markups. Most of these variables, with the notable exception of the exporter dummy, remain significant even after controlling for TFP.

The combined effect of these variables shows that globalized firms can enjoy much larger markups than non-internationalized firms. Take, for example a non internationalized firm and compare it with one which exports, imports from emerging countries, has FDI, quality certification, and has generated IP. The combined effect of these variables is 23.9 percentage points when not controlling for TFP and 14.6 percentage points when compared to firms with similar TFP levels. These numbers should be compared with the 24.1 mean markup and its 30 percentage point standard deviation.

5 Discussion and Conclusion

In this paper we have linked a rich firm-level survey database from four European countries with a panel of balance sheet data to investigate the relationship between five dimensions of globalization and markups at the firm level. Our main aim has been to analyse these variables

Table 10: Joint regressions

(1) (2) (3) (4)

Depvar: Markup in 2009

Exporter (dummy) 0.0364∗∗∗ 0.0143

[0.0117] [0.0122]

Importer of inputs (dummy) 0.0256∗∗∗ 0.0131

[0.0077] [0.0069]

Imports from emerging areas (dummy) 0.0459∗∗∗ 0.0426∗∗∗

[0.0132] [0.0094]

International outsourcer (dummy) 0.0767∗∗ 0.0516∗∗

[0.0371] [0.0235]

Group member (dummy) 0.0228 0.0101

[0.0134] [0.0123]

Controlling firm (dummy) 0.0504∗∗∗ 0.0284∗∗

[0.0126] [0.0117]

Foreign owned (dummy) -0.0059 -0.0352∗∗

[0.0141] [0.0148]

Firm has foreign competitor (dummy) 0.0068 0.0037

[0.0133] [0.0138]

Firm has competitor in emerging areas (dummy) -0.0330∗∗∗ -0.0171

[0.0091] [0.0087]

Quality certification (dummy) 0.0561∗∗∗ 0.0326∗∗∗

[0.0105] [0.0110]

IP creation (dummy) 0.0260∗∗∗ 0.0156

[0.0088] [0.0086]

R&D activity (dummy) 0.0045 0.0042

[0.0068] [0.0080]

Market share in 2008 0.2945∗∗∗ 0.6221∗∗∗ 0.2591∗∗∗

[0.0780] [0.1240] [0.0809]

Log TFP lagged 0.2273∗∗∗ 0.2128∗∗∗

[0.0236] [0.0234]

R-squared 0.144 0.278 0.185 0.291

F-stat (size, age, hhi) 8.628 13.479 6.449 28.983

F-stat (market share) 14.267 25.166 10.263

F-stat (tfp) 92.973 82.843

F-stat (all other) 11.179 10.551

Observations 6,830 6,830 6,830 6,830

Notes: All specifications include market share, market concentration, age, size dummies and country-industry (2-digit) dummies. Robust standard errors, clustered by industry, in brackets.

Observations are weighted by the sampling weights of firms. *** p<0.01, ** p<0.05, * p<0.1.

in a symmetrical way from a unified database and to present cross-sectional stylized facts.

Regarding exporting, we find a significant positive association with markups even when controlling for TFP. This is in line with predictions both from self-selection and learning models. This finding confirms the results of De Loecker and Warzynski (2012) who find that Slovenian exporters charge significantly higher markups than their non-exporting counter- parts and that a large part of this markup premium is due to the productivity advantage of exporters. In contrast, Marin and Voigtl¨ander (2013) find on Chilean data that firms experienced a fall in their marginal cost following export entry, which then they fully passed on to lower prices, leaving the markup unchanged.

In terms of importing, we also find a positive association, which may also result from self-selection or the causal effect of access to cheaper or higher quality intermediate inputs.

De Loecker et al. (2016) show that import tariff liberalization increased firm markups in India, as firms did not fully pass through the cost savings resulting from lower input tariffs to their output prices. Related to this, we have also found that, for Western European firms, importing from low wage countries such as China is key to higher markups. Finally, note that we found evidence of a similar channel by outsourcing—a novel finding in this literature.

The third channel of globalization is through ownership. Our finding of an FDI pre- mium in markups may be explained by self-selection into setting up affiliates (Helpman et al., 2004) or by being able to leverage knowledge assets to a larger degree. This premium remains significant even after controlling for productivity, suggesting a direct relationship between having affiliates and markups. Interestingly, foreign ownership is not significantly associated with markups in our sample. To some degree, this is in contrast with many papers showing a positive relationship between foreign ownership and productivity (Arnold and Javorcik, 2009) and markups (Clementi, 2015). This may partly be explained by dif- ferences between developed and developing countries (Indonesia and Romania in these two studies, respectively) but it is also possible that foreign ownership has a different effect on markups than on productivity. One reason for this can be that most of the profits from the productivity premium of foreign-owned firms is passed through their owners.

The fourth channel is competition with global firms. We find that import competition is also important. Firms that report more emerging market competitors experience lower

average markups.

Finally, it is important to take stock of these channels in a combined model. A great advantage of the EFIGE database is that it allows to combine various modes of being affected by globalization and review the joint effect. We contribute to the literature (similarly to (Mayer and Ottaviano, 2008) in the case of productivity) to show that the combined effect of these variables is quite large, consequently globalized firms can enjoy much larger markups than non-internationalized firms.

Note that our exercise has three main limitations. First, the cross-sectional nature of our database precludes us from estimating causal effects. Second, we can only estimate firm- level markups rather than firm-market specific ones. Third, it would be important to have a control for physical productivity, but we can only measure revenue TFP, which includes markups to some extent.

In general we find that globalized firms tend to charge significantly—both in the economic and statistical sense—higher markups than their non-globalized counterparts. This is likely to result, to a large extent, from self-selection of more competitive firms into different global- ization activities. Globalization, on the other hand, can be an important source of sustained competitiveness and higher markups for these firms. Innovation, for example is correlated strongly with markups even when controlling for many other globalization activities.

Our results also provide some evidence for a significant relationship between globalization activities and markups even when we control for TFP, a measure of productivity. This suggests that innovation and globalization activities are not only associated with increased productivity but also with generating more distinct, higher quality and more differentiated products.


Ackerberg, D., K. Caves, and G. Frazer (2006). Structural identification of production functions. Technical report.

Altomonte, C. and T. Aquilante (2012). The EU-EFIGE/Bruegel-Unicredit dataset. Tech- nical report.

Altomonte, C., T. Aquilante, G. B´ek´es, and G. I. Ottaviano (2013). Internationalization and innovation of firms: evidence and policy. Economic Policy 28(76), 663–700.

Antoniades, A. (2015). Heterogeneous firms, quality, and trade. Journal of International Economics 95(2), 263–273.

Antr`as, P. (2003). Firms, Contracts, and Trade Structure.The Quarterly Journal of Eco- nomics 118(4), 1375–1418.

Antr`as, P. and S. R. Yeaple (2014). Multinational firms and the structure of international trade. Volume 4, Chapter Chapter 2, pp. 55–130. Elsevier.

Arnold, J. M. and B. S. Javorcik (2009). Gifted kids or pushy parents? Foreign di- rect investment and plant productivity in Indonesia. Journal of International Eco- nomics 79(1), 42–53.

Atkin, D., A. Chaudhry, S. Chaudry, A. K. Khandelwal, and E. Verhoogen (2015). Mark- up and Cost Dispersion across Firms: Direct Evidence from Producer Surveys in Pak- istan. Technical report.

Aw, B. Y., S. Chung, and M. J. Roberts (2000). Productivity and turnover in the export market: Micro-level evidence from the republic of korea and taiwan (china).The World Bank Economic Review 14(1), 65–90.

Burstein, A. and G. Gopinath (2014). International prices and exchange rates. Volume 4, Chapter Chapter 7, pp. 391–451. Elsevier.

Clementi, F. (2015). Market power, foreign ownership and competitive pressure. mimeo.

Retrieved from http://www.etsg.org/ETSG2015/Papers/018.pdf, July 18, 2016.

De Loecker, J., P. K. Goldberg, A. K. Khandelwal, and N. Pavcnik (2016). Prices, markups, and trade reform. Econometrica 84(2), 445–510.

De Loecker, J. and F. Warzynski (2012). Markups and Firm-Level Export Status.Amer- ican Economic Review 102(6), 2437–71.

Feenstra, R. C. and D. E. Weinstein (2010). Globalization, Markups and U.S. Welfare.

Technical report.

Feng, L., Z. Li, and D. L. Swenson (2012). The Connection between Imported Intermediate Inputs and Exports: Evidence from Chinese Firms. Technical report.

Foster, L., J. Haltiwanger, and C. Syverson (2008). Reallocation, Firm Turnover, and Effi- ciency: Selection on Productivity or Profitability? American Economic Review 98(1), 394–425.

Goldberg, P. K., A. Khandelwal, N. Pavcnik, and P. Topalova (2008). Imported Interme- diate Inputs and Domestic Product Growth: Evidence from India. Technical report.

Hall, R. E. (1986). Market Structure and Macroeconomic Fluctuations.Brookings Papers on Economic Activity 17(2), 285–338.

Hall, R. E. (1988). The Relation between Price and Marginal Cost in U.S. Industry.

Journal of Political Economy 96(5), 921–47.

Halpern, L., M. Koren, and A. Szeidl (2015). Imported Inputs and Productivity.American Economic Review 105(12), 3660–3703.

Helpman, E., M. J. Melitz, and S. R. Yeaple (2004). Export Versus FDI with Heteroge- neous Firms. American Economic Review 94(1), 300–316.

Krugman, P. (1979). Increasing returns, monopolistic competition, and international trade. Journal of International Economics 9(4), 469–479.

Kugler, M. and E. Verhoogen (2009). Plants and Imported Inputs: New Facts and an Interpretation. American Economic Review 99(2), 501–07.

Kugler, M. and E. Verhoogen (2012). Prices, Plant Size, and Product Quality.Review of Economic Studies 79(1), 307–339.

Levinsohn, J. and A. Petrin (2003). Estimating production functions using inputs to con- trol for unobservables. The Review of Economic Studies 70(2), 317–341.

Marin, A. G. and N. Voigtl¨ander (2013). Exporting and Plant-Level Efficiency Gains: It’s in the Measure. Technical report.

Markusen, J. R. (2004).Multinational Firms and the Theory of International Trade. The MIT Press.

Mayer, T. and G. I. Ottaviano (2008). The happy few: the internationalisation of european firms. Intereconomics 43(3), 135–148.

Melitz, M. J. (2003). The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity. Econometrica 71(6), 1695–1725.

Melitz, M. J. and G. I. P. Ottaviano (2008). Market Size, Trade, and Productivity.Review of Economic Studies 75(1), 295–316.

Murak¨ozy, B. and C. Hornok (2015). Markup and productivity of exporters and importers.

Technical report.

Navaretti, G. B., M. Bugamelli, F. Schivardi, C. Altomonte, D. Horgos, and D. Maggioni (2011). The global operations of European firms - The second EFIGE policy report.


Novy, D. (2013). International trade without CES: Estimating translog gravity. Journal of International Economics 89(2), 271–282.

Olley, G. S. and A. Pakes (1996). The Dynamics of Productivity in the Telecommunications Equipment Industry. Econometrica 64(6), 1263–97.

Wooldridge, J. M. (2009). On estimating firm-level production functions using proxy vari- ables to control for unobservables. Economics Letters 104(3), 112–114.

Appendix A: Production function estimation

We perform the production function estimation on the value added with capital and labor inputs following Wooldridge (2009). Wooldridge (2009) shows that the two-step production function estimation procedures developed by Olley and Pakes (1996), Levinsohn and Petrin (2003) and Ackerberg et al. (2006) can be implemented in a one-step generalized method of moments (GMM) framework.

The production function estimating equation with all variables in logs is

yitllitkkit+g(ki,t−1, mi,t−1) +δt+it. (4) Value added output of firmiin year t(yit) is a function of the current labor (lit) and capital (kit) use and a functiong(.) of lagged capital and material use, which proxies for the expected (in t−1) component of the current total factor productivity, while theδt are year intercepts.

As it is customary in the literature, we specify function g(.) as a third-degree polynomial with interaction terms.16 The error term it also incorporates the (unexpected) productivity shock.

The parameters of interest, βl andβk, measure the output elasticity of labor and capital, respectively. In order to obtain unbiased estimates, however, one has to account for the possible correlation between the current variable input (lit) and the productivity shock in the error term. This is achieved by a generalized method of moments instrumental variable estimation, where lit is instrumented with li,t−1, while all other right-hand side variables are instruments for themselves.

We measure value added output as sales minus material costs, labor input by the number of employees, capital input by fixed assets and material use by material costs of the firm.

We deflate sales and material costs with industry- and country-specific producer prices and fixed assets with country-specific prices for capital goods.17

We estimate (4) on an unbalanced panel of the broadest possible set of French, German, Italian and Spanish firms in the Amadeus database over years 2004-2013. We perform the

16The terms of the polynomial are hence ki,t−1, mi,t−1, ki,t−1mi,t−1, k2i,t−1, m2i,t−1, ki,t−12 mi,t−1, ki,t−1m2i,t−1,ki,t−13 andm3i,t−1.

17The source of the price indices is Eurostat.

estimation separately by country and three-digit NACE industry.18 Figure A.1 present the histogram of the estimated output elasticities of labor for our baseline estimation sample.

The estimated industry-country elasticities fall in a reasonable range with a sample mean of 0.66.

Figure A.1: Histogram of ˆβl by country


.2 .4 .6 .8 1 .2 .4 .6 .8 1





Graphs by country

Based on the estimated output elasticities we can calculate the total factor productivity (in log) of firmi in year t as

ln TFPit =yit−βˆl(cj)lit−βˆk(cj)kit,

where ˆβk(cj) is the estimated output elasticity for capital, specific to country-industry cj.

Figure A.2 presents histograms of the estimated firm-level productivities by country.

18We made sure that the number of observations per country and industry is not smaller than 50, otherwise we merged some three-digit industries.