1. The Effect of Privatization and Competitive Pressure on Firms’ Price-Cost Margins Micro Evidence from Emerging Economies Jozef Konings, Patrick Van Cayseele and Frederic Warzynski LICOS, University of Leuven, Belgium

2. Motivation Legacy of Central Planning: Many large firms: Legacy of Central Planning: Many large firms: Size Distribution 1990 in Selected Countries (Roland, 2000) Size Distribution 1990 in Selected Countries (Roland, 2000) 0-100100-500500+ W-GermanyFranceItaly14.1%22.5%32.3%23.9%24.9%27.3%62%52.6%40.4% GDPCzechoslov.HungaryPoland1%0.1%4.5%1.4%11.1%3.4%16.3%18.2%87.9%96.5%79.3%80.4%

3. Motivation (contd.) Rapid institutional changes in Central and Eastern Europe: Rapid institutional changes in Central and Eastern Europe:  Privatization  Liberalizing markets (hence increased competitive pressure)  Transition countries are getting close to a natural experiment to evaluate these effects on firm performance, measured here as price-cost margins.

4. Effects of Privatization? Traditional Argument Against Privatization:  Firms increase pricesLoss of Allocative Efficiency (Tirole, 1991; Li, 1999; Joskow and Schamelensee, 1995) Ignores dynamic effects of privatization DDDDue to changed ownership, incentives to engage in restructuring and innovation improve, which lowers costs (Schmidt, 1996; Hart, Shleifer, Vishny, 1997; Bennett and Maw, 2000)

5. Effects of Competitive Pressure? Contestability argumentprice-cost margins will reduce (evidence: Levinsohn, 1993…) But: In a dynamic context: Trade protection may help to invest in catching-up (Rodrik, 1992) Complementarity between competition and privatization?  Schmidt (1996) shows there is due to the hardening of budget constraints when there is more competition.

6. Contributions Previous studies had to rely on small samples of firms, this paper uses a representative panel of firms in Bulgaria and Romania Estimate price-cost margins using the Roeger (1995) method, which yields consistent estimates, which is in the spirit of recent work that estimates production functions (Olley and Pakes, 1995).

7. Methodology Roeger (1995):method for estimating markups using company accounts Other methods: require price data. Assume a CRTS production function

8. Methodology (contd.) Under perfect competition, the Solow decomposition is: Under perfect competition, the Solow decomposition is:

9. Methodology (contd.) Under Imperfect competition (Hall, 1988) Under Imperfect competition (Hall, 1988) with μ = P/MC or this can be written as: (primal Solow residual (1)) where β = (P-MC)/P = 1-1/μ = the lerner index

10. DATA Data Source: Amadeus data base of Bureau Van Dijck Data Source: Amadeus data base of Bureau Van Dijck  Company accounts of all medium and large sized enterprises of Bulgaria and Romania  Company accounts of all medium and large sized enterprises of Bulgaria and Romania  Harmonized by BvD across countries  Harmonized by BvD across countries 1701 firms in Bulgaria and 2047 firms in Romania 1701 firms in Bulgaria and 2047 firms in Romania Background: Background:Bulgaria: GDP per capita in 1999: 1513 USD/capita GDP per capita in 1999: 1513 USD/capita Population: 8 million Population: 8 million Privatization started late, after 1996 Privatization started late, after 1996 Competition Policy: weak (EBRD) Competition Policy: weak (EBRD)Romania: GDP per capita in 1999: 1512 USD/ capita GDP per capita in 1999: 1512 USD/ capita Population: 22.3 million Population: 22.3 million Privatization started late, after 1996 Privatization started late, after 1996 Competition Policy: weak (EBRD) Competition Policy: weak (EBRD)

11. DATA (contd) Sample period : 1994-98 Sample period : 1994-98 Ownership information only for 1997- 1998 Ownership information only for 1997- 1998 We dropped those firms for which we could not trace ownership information. We dropped those firms for which we could not trace ownership information. Table 1: Comparison between Amadeus and National Statistics, 1998 BulgariaRomania Employment coverage.66.70 Sales Coverage.82.69 Sales coverage ratio = total sales in Amadeus / total manufacturing sales. Employment coverage ratio = total employment in Amadeus / total manufacturing employment.

12. Table 3: Average Ownership Shares in Sample Type of firm BulgariaRomania Average fraction of shares held by private domestic owners in entire sample Average fraction in private firms only 0.65 (0.38) 0.80 (0.24) 0.50 (0.39) 0.58 (0.36) Average fraction of shares held by foreign owners in entire sample Average fraction held in foreign firms only 0.04 (0.17) 0.65 (0.26) 0.10 (0.27) 0.67 (0.29) Average fraction of shares held by the state in entire sample Average fraction held in state firms only 0.30 (0.36) 0.44 (0.36) 0.31 (0.33) 0.58 (0.22) Note: standard deviations in parentheses

13. Table 4: Types of Ownership (percentage of firms in the sample) Type of firm BulgariaRomania Majority Domestic Private 74 % 47% Majority Foreign 6%11% Majority State 20 % 42%

14. Table 5: Summary statistics Summary Statistics Bulgaria: Sample Means and Standard Deviations Full sample Majority Private domestic Majority Foreig n Majority State Employment493(981)392(641)730(648) 595 (1377 ) Sales 6634 (42850) 4934(18918) 12312 (1917 2) 9085 (3960 4) Wage Bill 876(3250)784(2615) 1726 (2066) 1414 (4609 ) Material Costs 4162 (31967) 2817(13705) 7592 (1303 2) 5851 (2756 0) Tangible Fixed Assets 2664 (12017) 2333(11588) 4784 (5991) 4672 (1434 9) Depreciation rate 0.14 (0.22) (0.22)0.16(0.22) 0.18 (0.15) 0.11 (0.10)

15. Summary Statistics Romania: Sample Means and Standard Deviations Full sample Majority Private domestic Majority Foreign Majority State Employment469(1028)378 (699) (699)690(558)624 (1525) (1525) Sales 7853 (52524) 5173 (21441) (21441)10786(11962)9231(43535) Wage Bill 1012 (3901) (3901)829(2962)1856(2135)1541 (5109) (5109) Material Costs 5170 (39348) 3030(15625)7066(8995)6117(30418) Tangible Fixed Assets 2975 (14032) 2464 (13060) (13060)5211(6316)4717(15603) Depreciation rate 0.08(0.17)0.10 (0.28) (0.28)0.14(0.15)0.06(0.08) Note: Standard deviations in parentheses; values expressed in thousands of $

16. Size Distribution of the Sample in 1997 0-100100-500 More than 500 employees Bulgaria1%43%56% Romania1%18%81%

17. Table 3 Estimates of Price-Cost Margins in Different Sectors NACE Code DescriptionBulgariaRomania 15Food and Beverages0.19** (0.017)0.11** (0.006) 16Tobacco0.21** (0.030)- 17Textiles0.19**(0.016)0.10** (0.007) 18Wearing apparel; fur0.20**(0.022)0.20**(0.015) 19Leather, luggage and footwear0.19** (0.039)0.16**(0.013) 20Wood, straw and plaiting materials0.06* (0.036)0.006 (0.004) 21Pulp, paper and paper products0.14**(0.017)0.15**(0.033) 22Publishing, printing and media0.42 (0.330)0.33**(0.046) 23Coke, refined petroleum products, nuclear fuel -0.15**(0.013) 24Chemicals and chemical products0.19**(0.021)0.13**(0.015) 25Rubber and plastic products0.24**(0.038)0.14**(0.012) 26Other non metallic mineral products0.15**(0.016)0.16**(0.006) 27Basic metals0.21**(0.028)0.12**(0.009) 28Fabricated metal products0.17**(0.023)0.17**(0.009)

18. Table 4 Results Baseline ModelRobustness Checks Sample 1994-1998 Robustness Checks Capital Fixed Bulgaria (1) Romania (2) Bulgaria (3) Romania (4) Bulgaria (5) Romania (6) Price-Cost Margin0.124** (0.029) 0.015** (0.003) 0.17** (0.01) 0.069** (0.004) 0.16** (0.03) 0.21** (0.016) The effect of import penetration 0.059 (0.042) 0.035** (0.006) -0.05 (0.03) 0.018** (0.003) -0.15** (0.058) 0.014* (0.009) The effect of the Herfindahl index 0.226** (0.087) 0.356** (0.046) 0.18** (0.08) 0.29** (0.04) -0.04 (0.12) 0.11 (0.10) The effect of Private Domestic Firms 0.037* (0.018) 0.118** (0.008) 0.03** (0.01) 0.085** (0.006) 0.04** (0.02) 0.055** (0.014) The effect of Foreign Firms0.071** (0.032) 0.133** (0.023) 0.05* (0.03) 0.165** (0.02) 0.09** (0.04) 0.148** (0.02) The effect of Import Penetration x Herfindahl -0.560** (0.195) -0.18** (0.034) -0.14 (0.16) -0.07** (0.04) 0.05 (0.30) -0.013 (0.06) Year dummiesYes F-test (private=foreign)0.260.540.460.0010.250.0004 R 2 within0.780.610.730.590.320.58 R 2 between0.360.390.240.380.100.27 R 2 overall0.590.520.460.520.200.42 # observations108417481454293918772877

19. Table 5 Results for Highly versus Lowly Competitive Sectors BulgariaRomania Highly Competitive Lowly Competitive Highly Competitive Lowly Competitive Price-cost Margin0.08** (0.03) 0.18** (0.03) 0.02** (0.003) 0.12** (0.006) The effect of Private Domestic Firms 0.095** (0.026) -0.0001 (0.022) 0.14** (0.009) 0.062** (0.01) The effect of Foreign Firms 0.058 (0.12) 0.041 (0.035) 0.157** (0.031) 0.097** (0.033) Year dummiesYes F-test (Private=Foreign) 0.760.180.610.31 R 2 within0.750.790.560.72 R 2 between0.430.440.510.31 R 2 overall0.600.630.530.55 # observations541660955874

20. Conclusions Privatization is associated with higher price-cost margins Privatization is associated with higher price-cost margins The main effect of privatization is on cost reduction, rather than price increases The main effect of privatization is on cost reduction, rather than price increases The effects of privatization are stronger in highly competitive sectors The effects of privatization are stronger in highly competitive sectors International competition disciplines firms’ pricing behavior, in highly concentrated sectors, in lowly concentrated sectors the international competition is associated with higher price-cost margins International competition disciplines firms’ pricing behavior, in highly concentrated sectors, in lowly concentrated sectors the international competition is associated with higher price-cost margins Privatization and competitive pressure go together Privatization and competitive pressure go together