The paper founded on the theory of the degree of monopoly (Kalecki, 1954) and is combined with the traditional industrial organisation’s Structure-Conduct-Performance (SCP) paradigm. The empirical analysis is based on a panel of the annual firm level data for Slovenian manufacturing industries for the 1994-2003 period. We apply the regression analysis, using the GMM dynamic panel model estimator as proposed by Arellano and Bond (1991), which efficiently overcomes the autocorrelation due to the model dynamics, the impact of unobserved individual effects and the possible endogeneity of some of the right-hand regressors due to simultaneity bias. The analysis is conducted at both the industry and firm levels, which enables us to make a clear distinction between the set of firm- and industry-specific markup factors. Further, the model of markup determination is estimated separately for each of the investigated 2-digit NACE manufacturing industries in order to identify the variability of the direction and strength of the impacts of markup determinants across various manufacturing industries.

Results of the empirical study show that for a firm to have a high markup compared to its rivals (i.e. within a particular industry), it has to have the ambition to grow fast and the ability to hold a relatively large market share. This fast growth as well as large market power should be the consequence of its competitive advantage in comparison to its rivals, resulting from high capacity utilisation, low labour costs and high labour productivity. A high capital intensity of production and a low export orientation are also characteristics of a firm with a high markup compared to its competitors within an industry, while a firm’s size is not. An average industry-level markup would be high in industries with high sellers’ concentration and low import penetration, with a high capital intensity of production and where entry barriers are present. These industries have to be export-oriented, have growing market demand and to utilise their production capacities well. The time persistence is found in the pricing behaviour of the analysed firms both on the firm- as well as on the industry-level, showing that the previous realisations of markups are a relatively strong driving force of current markups.

The analysis also indicates that in some cases the estimates of the aggregate models might conceive true relationships between investigated variables. We find that while firm-specific factors’ coefficients remain relatively stable in their direction of their impact and statistical significance, the array, direction of impact and statistical significance of industry-level factors varies a great deal among manufacturing industries. This indicates that while similar factors are important for a firm’s markup determination relative to its rivals within the industry, the industry-specific markup determinants vary a great deal.