VoxEU Column International trade

Globalisation and the fall in markups in Poland

There has been a lively debate concerning the dynamics of markups. This column contributes to this debate by studying the relationship between globalisation and monopolistic markups in Poland. It highlights important non-linearities leading to an uneven  distribution of the effects of global value chains for participating firms, with lowest benefits found for firms in the middle of the production chain where goods are highly standardised and substitutable. It also documents a fall in markups for Poland which can be explained by rising dependence on imported intermediates in export-oriented production and fiercer competition of domestic firms on export markets.

There has been a lively debate in recent years concerning the dynamics of markups of price over marginal costs (e.g. De Loecker et al. 2020). We add to this debate and propose globalisation trends as a factor contributing to those changes in markups. We study the case of Poland, which – together with other Central and Eastern European countries – joined the production chains of Western European countries in the 2000s, focusing on the production of intermediates. This had lasting effects on Central and Eastern European economies, reshaping their enterprise sectors and fuelling convergence of those countries. At the micro level, Hagemejer and Kolasa (2011) provide evidence on both vertical and horizontal efficiency spillovers from globalisation in Poland. Moreover, the benefits of globalisation in Central and Eastern European countries are also highlighted in Hagemejer and Mućk (2019) who show that entering global value chains (GVCs) resulted in a rapid capital deepening and a rise of trade within the chain, which contributed significantly to value added growth of these economies.

In our study (Gradzewicz and Mućk 2019), we use a unique, granular dataset of all Polish firms employing more than nine employees, covering the period from 2002 to 2016. Monopolistic markups are measured using the identification scheme proposed by De Locker and Warzynski (2012). Next, we aggregate them into 2-digit production sectors and combine them with globalisation measures calculated from the World Input-Output database.

Our results imply that the link between GVC participation and markups is quite complex, which is in line with the non-linear effects of a country's position in GVC found in the literature. These non-linearities stem from the fact that although fragmentation, outsourcing or offshoring imply efficiency improvements for the whole production process, these gains are not necessarily evenly distributed across the chain. In this vein, Timmer et al. (2014) and Ye et al. (2020) show that the benefits from GVC participation, measured by the value added, are highest for firms that are either at the beginning of the production process (e.g. R&D, design) or very close to the final use (e.g. marketing, advertising, post-sale services). On the contrary, for firms engaged in the middle of the production chain (e.g. production of intermediates, usually located in manufacturing sectors) the benefits from the participation in GVC are the lowest.  

Figure 1 Sectoral relationship between markups (on vertical axis in both graphs) and distance to final demand (left panel, horizontal axis) and import intensity of exports (right panel, horizontal axis)


We show that this U-shape relationship applies not only to value added but also to markups (left panel of Figure 1). Industries located in the middle of the production chain as measured by distance to final demand (the ‘upstreamness’ index introduced by Antras et. al. 2012) are characterized by the lowest markups. The ability of firms in these industries to set prices above marginal costs is limited by the fact that they are manufacturers of relatively standardised products and hence are subject to competition from all over the world. Moving either to the beginning or to the end of the production process makes the product less standardised and less substitutable. This gives firms more room for price setting.

Our results also indicate that the relationship between the participation in GVC (measured as the foreign value added of exports) and markups exhibits an inverted U-shape (right panel of Figure 1). In industries that are less engaged in the GVC, an increase in participation drives markups up. There are at least two reasons for that. First, for relatively less internationally integrated firms a decision to import more usually means to import more high-tech intermediates as they are not available on the domestic market, especially in emerging economies. This could lead to quality improvements of produced goods, translating into higher markups. Second, as the usage of foreign supplies is strongly related to the degree of integration of supply chains, it facilitates the adoption of more advanced technologies. Those then boost production efficiency, driving marginal costs down and leaving more room for markups. These effects explain the positive effect of more GVC integration on markups in exporting firms in the early stage of the integration with GVC.  

But our regression analysis also reveals that for firms which are highly engaged in the GVC, the relationship between the usage of imported intermediates and markups is negative. Since these firms are tightly integrated with both foreign suppliers of intermediates and with producers of finals goods within the production chain, it suggests that their degree of vertical specialization is large. Thus, their dependence on being in the production chain limits their ability to exert high market power.

Globalisation and the decline of markups in Poland

We also showed that markups in Poland fell during the last 15 years, by accumulated 18.6% in weighted terms. This finding is in contrast to widely discussed evidence for the US economy (e.g. De Loecker et al. 2020).  On the other hand, Diez et al. (2018) show that for emerging economies, markups are also stable or declining. The decline in markups we observe for Poland is unrelated to the measurement strategy used and is present in more than 70% of all firms we investigate and in most of the industries. The fall is notably more pronounced in exporting firms for Poland. Those firms still experienced relatively higher markups in the early 2000s (compared to non-exporting firms), a pattern consistent with empirical evidence from De Locker and Warzynski (2012), who document a significant exporter premium for Slovenian firms at the end of the 1990s. However, since the middle of the 2000s, the markups for exporting firms have declined substantially (black line in Figure 2) and the exporter premium evaporated. 

We run a regression with firm- and industry-level characteristics in order to better understand what contributed to the decline in markups in Poland. We then decompose the cumulated change of average markups into components related to the explanatory variables used, as depicted in Figure 2. Our analysis shows that the decline in markups is to a large extent explained by globalisation trends. When Polish firms joined the global value chains, they not only started to operate at a higher scale but also adjusted their production techniques. This led to the productivity gains that are extensively documented in related literature (Hagmejer and Kolasa 2011), but while rising productivity (measured by total factor productivity, see Figure 2) of exporting firms was indeed a factor pushing markups up there were simultaneous, stronger forces driving markups down.

Figure 2 Sources of markup declines in exporting firms

Note: FVAX = foreign value added in exports.

Figure 2 shows that there are two main forces behind the fall of markups for exporters: rising dependence on imports (measured by foreign value added in exports) and the concentration of domestic firms on export markets (measured on the industry level by the number of symmetrical firms on export markets). The increasing reliance of firms on imported components in production related to the rising participation in the GVC contributed to a fall in markups on average. In other words, most firms were located on the right side of the inverted U-curve in the right panel of Figure 1 and moved further to the right.  While using more sophisticated imported inputs tends to increase marginal costs, it does not need to result in lower markups in principle, as firms may pass these costs on through higher prices. However, this did not happen as the rising presence of domestic firms on export markets (measured by declining concentration) lead to fiercer competition for export orders and limited the ability of firms to raise prices. The combination of these two phenomena played the most important role in explaining the fall of markups in Poland.


Antras, P, D Chor, T Fally and R Hillberry (2012), “Measuring the Upstreamness of Production and Trade Flows," American Economic Review 102(3): 412-416.

De Loecker, J, J Eeckhout and G Unger (2020), “The Rise of Market Power and the Macroeconomic Implications”, The Quarterly Journal of Economics 135(2).

De Loecker, J and F Warzynski (2012), “Markups and Firm-Level Export Status”, American Economic Review 102(6): 2437-2471.

Diez, F, D Leigh and S Tambunlertchai (2018), “Global Market Power and its Macroeconomic Implications”, IMF Working Paper No. 18/137.

Gradzewicz, M and J Mućk (2019), “Globalization and the fall of markups”, NBP Working Paper 304.

Hagemejer, J and M Kolasa (2011), “Internationalisation and Economic Performance of Enterprises: Evidence from Polish Firm-level Data”, The World Economy 34(1): 74-100.

Hagemejer, J and J Mućk (2019), “Export-led growth and its determinants evidence from CEEC countries”, The World Economy 42(7): 1994-2025.

Timmer, M P, A A Erumban, R Los, R Stehrer and G J de Vries (2014), “Slicing up global value chains”, Journal of Economic Perspectives 28 (2): 99-118.

Ye, M, B Meng and S J Wei (2020), “Measuring smile curves in global value chains”, Oxford Bulletin of Statistics and Economics, forthcoming.

1,365 Reads