The rise of firms' product market power is pervasive around the globe (De Loecker et al. 2020, Dıez et al. 2018, 2021). However, firms can exert market power in both product and labour markets. This is crucial to understanding firms' market power because the two markets are deeply connected (Dobbelaere and Mairesse 2013, Yeh et al. 2022), with important welfare implications (Eeckhout et al. 2022). On the one hand, product market power implies that firms can produce less and set prices above the marginal costs of production, which can reduce labour demand and affect wages. On the other hand, firms with monopsony power can hire workers at lower wages, ultimately affecting firms' pricing behaviour. Despite their obvious relationship, the joint characterisation of firms' power in both the product and labour markets still lacks systematic research.
In recent research (Ding et al. 2023), we add to this literature by documenting the dynamics of price markups and wage markdowns in Lithuania since its accession to the EU. The Lithuanian economy provides an interesting environment for our exercise. In the product market, rapid economic growth fostered significant firm entry, intensifying competition among producers. In the labour market, the right to live and work in other EU member states led to a wave of mass emigration. This high emigration episode, combined with the high firm entry rate, led to a rise in the number of firms per worker, with potential implications for labour market power.
In this column, we present the evolution of Lithuanian firms' power in the product and labour markets between 2004 and 2018 and discuss the forces that drive these trends.
Gaining power in the product market but losing it in the labour market
The top panel in Figure 1 plots the evolution of aggregate markup and aggregate markdown relative to their baselines in 2004. The aggregate markup increased by about 2% during this period, with a contraction from 2004 to 2008 and a steady increase since the Great Recession. Over the same period, the aggregate markdown declined by about 4%, with a sharp reduction until the onset of the Great Recession, followed by a sudden adjustment from 2009 to 2011, and since then, a steady decline, reflecting the rise in the aggregate labour share. At the micro level, firms that charge higher markups in the product market tend to be those firms that charge lower markdowns in the labour market. This is portrayed on the bottom panel in Figure 1, where we provide a binned scatter plot from firm-level regressions of (log) markups on (log) markdowns. This finding suggests firms with high product market power may share rents with their workers to a greater extent, which is in line with the recent evidence in the literature.
Figure 1 Markups and markdowns at the macro (top) and micro (bottom) level
Source: Annual survey of enterprises in Lithuania 2004-2018.
Note: The top panel plots the aggregate markup (weighted by intermediate input costs) and aggregate markdown (weighted by labour cost). Each series is normalised to its 2004 (base) value. The bottom panel is a binned scatter plot from firm-level regressions of (log) markup on (log) markdown, controlling for year and industry fixed effects.
Reallocation across sectors
Panels (A) and (B) in Figure 2 show a simple decomposition of aggregate markup into two terms: an average sector-level markup and a covariance term between sector size and the sector markup. Since the EU accession, the average markup at the sector level has declined. However, the covariance term has been increasing, which implies sectors with higher markups are expanding over time. These observations are consistent with the significant economic transformation experienced by the Lithuanian economy in the last two decades. For example, between 2004 and 2018, the contribution of most service activities to total employment and gross value-added increased. Consistent with these dynamics, two-digit service sectors exhibited the highest markup levels at the beginning of our period and have expanded the most.
Panels (C) and (D) show the same decomposition for aggregate markdown. Since 2004, the average markdown at the sector level has declined. In addition, the covariance term has been decreasing, which implies sectors with lower markdowns are getting larger over time. These observations indicate that both within-sector and between-sector components are pushing down the aggregate markdown, suggesting that the whole economy is experiencing a reduction in monopsony power. For instance, the ‘manufacture of paper and paper products’ industry had a markdown of 0.49 in 2004 (while the aggregate markdown in 2004 was 2.03), and this industry has increased its size by almost 40% during the same period.
Figure 2 Decomposition of markup and markdown at the sector level
Source: Annual survey of enterprises in Lithuania 2004-2018.
Note: The graph plots the OP (Olley and Pakes 1996) decomposition for the intermediate input cost-weighted aggregate markup and labour cost-weighted aggregate markdown at the two-digit NACE2 level, with the mean normalised to its 2004 (base) value.
Reallocation across firms
The top panel in Figure 3 shows a firm-level decomposition to quantify the changes that occur within firms (within), across firms (between and covariance), as well as through firm entry and exit (net entry). The top panel suggests that although the average firm-level markup has declined over time, the reallocation process and firm dynamics have more than compensated for this decline, leading to a rise in the aggregate markup. The graph highlights a very competitive feature of the product market in the Lithuanian economy: firms have experienced persistent productivity growth (Garcia-Louzao and Tarasonis 2023), yet the increasing degree of competition in the market has consistently put pressure on the incumbents, driving down firms’ product market power.
The bottom panel shows the same decomposition of markdown at the firm level. The graph indicates the ‘within’ component dominates the reallocation forces, suggesting that the reallocation process across firms has mitigated the significant decline in aggregate markdown that would have arisen from the declining dynamics of the average markdown (20%), resulting in the observed drop of approximately 5% between 2004 and 2018. Our findings that the ‘within’ component is the main factor that drives down the aggregate markdown in Lithuania is consistent with the country's increasing degree of labour market competition and compression in the dispersion of firm-specific wage components, suggesting firms are losing wage-setting power (Garcia-Louzao and Ruggieri 2023).
Figure 3 Decomposition of markup and markdown at the firm level
Source: Annual survey of enterprises in Lithuania 2004-2018
Note: The graph plots the FHK (Foster et al. 2001) decomposition of the intermediate input cost-weighted aggregate markup and labour cost-weighted aggregate markdown. Each series is normalised to its 2004 (base) value.
Conclusion
Following Lithuania's accession to the EU in 2004 and its zone of free movement of goods, labour, and capital (Randveer and Staehr 2021), we document different firm power dynamics in product and labour markets. On the one hand, the substantial economic growth from 2004 to 2018 was accompanied by a significant increase in the stock of firms, which may partly explain the milder increase in product market power that we document for Lithuania compared to the sharp increases observed in other countries (De Loecker et al. 2020, Dıez et al. 2021). On the other hand, the large migration episode following the EU accession resulted in severe labour shortages, which likely contributed to the decline in firms’ monopsony power. Moreover, the lower margin on labour costs can also influence markups through its impact on production costs and, thus, on input-mix decisions. In this regard, we find that growing industries experienced increasing markups but decreasing markdowns, suggesting that the fall in labour market power may also help explain the trend in the aggregate markup.
References
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): 561-644.
Díez, F J, S Tambucnlertchai and D Leigh (2018), “Global market power and its macroeconomic implications”, VoxEU.org, 28 June.
Díez, F J, S Tambucnlertchai and D Leigh (2021), "Global declining competition?", Journal of International Economics 132(C).
Ding, Z, J Garcia-Louzao and V Jouvanceau (2023), "The Dynamics of Product and Labor Market Power: Evidence from Lithuania", Bank of Lithuania Working Paper Series 118, Bank of Lithuania.
Dobbelaere, S and J Mairesse (2013), "Panel data estimates of the production function and product and labor market imperfections", Journal of Applied Econometrics 28(1): 1-46.
Eeckhout, J, S Deb, A Patel and L Warren (2022), "Market Power and Wage Inequality", CEPR Discussion Papers 17569.
Foster, L, J C Haltiwanger and C J Krizan (2001), “Aggregate productivity growth: Lessons from microeconomic evidence”, in New developments in productivity analysis, pp. 303-372, University of Chicago Press.
Garcia-Louzao, J and L Tarasonis (2023), "Productivity-enhancing reallocation during the Great Recession: evidence from Lithuania," Oxford Economic Papers 75(3): 729-749.
Garcia-Louzao, J and A Ruggieri (2023), "Labor Market Competition and Inequality", Bank of Lithuania Working Paper Series 117.
Olley, G S and A Pakes (1996), “The Dynamics of Productivity in the Telecommunications Equipment Industry”, Econometrica 64(6): 1263-1297.
Randveer, M and K Staehr (2021), "Macroeconomic Trends in the Baltic States Before and After Accession to the EU", in M Landesmann and I P Székely (eds.), Does EU Membership Facilitate Convergence? The Experience of the EU's Eastern Enlargement, Vol. I, pages 211-237, Palgrave Macmillan.
Yeh, C, C Macaluso and B Hershbein (2022), "Monopsony in the US Labor Market", American Economic Review 112(7): 2099-2138.
