The establishment-size wage premium in Greece

PRELIMINARY VERSION-not to be quoted (Draft March 2010)

Daouli J. J., M. Demoussis*, N. Giannakopoulos and I. Laliotis Department of Economics, University of Patras, Greece

Abstract In this paper we examine the establishment-size wage premium in Greece using a matched employee-employer dataset. The results of the econometric estimation suggest that indeed such a premium is present in the Greek market sector, too. Its magnitude is in line with those reported in other economies. Human capital, compensating wage differentials/incentive payment schemes and internal labor markets seem to be primarily responsible for the presence of the premium in question. Furthermore, the premium in manufacturing is substantially higher than the one in the services sector, indicating that employers in the former enjoy economics of scale, which eventually lead to higher wages, ceteris paribus. Nevertheless, a sizeable part of the premium remains unexplained.

Keywords: establishment-size, wages, Greece JEL codes: J31

* Address for correspondence: [email protected]. Department of Economics, University of Patras, University Campus, 26504, Rio, Patras, Greece, Tel.: (30)2610996134.

Electronic copy available at: http://ssrn.com/abstract=1592247

I Introduction Many studies worldwide have identified firm size as an important correlate of individual wage variation. Apparently, the investigation of wage differentials between different firm sizes has turned out to be a very fertile process for advancing our understanding of the manner in which workers and firms interact in the labor market. In a typical competitive labor market, workers are paid their value of marginal product and thus observationally equivalent workers are not expected to face any firm size-pay differentials. Nevertheless, there is ample evidence suggesting that significant wage premia exist in favor of workers in larger firms (Mellow, 1982; Blanchflower, 1986; Brown and Medoff, 1989; Gerlach and Schmidt, 1990; Schmidt and Zimmermann, 1991; Idson and Feaster, 1991; Oi and Idson, 1999; Troske, 1999). The aim of the present paper is to contribute in the relevant literature by providing evidence based on matched employer-employee data from the Greek labor market. One distinctive feature of the Greek economy is the small, in comparison with other European economies, average firm size (Kumar, et al. 2001; Burtless, 2002). This characteristic is evident in the data reported at Table 1 (accounting for the total number of firms with a registry record, including self-employed and family enterprises). More than 98 percent of the total number of operating firms falls in the two lowest employment categories [0-9 employees]. This distribution of firm sizes clearly undermines the capability of Greek private enterprises to exploit existing economies of scale (e.g., large firms employ more specialized workers, secure easier access to financial capital and resources for investment in R&D and worker training). The resulting loses in labor productivity due to the small size of Greek firms are markedly and clearly depicted in Figure 1. In a more formal manner, Nicolitsas (2007) using data from the 2002 Structure of Earnings Survey (SES) has regressed 1

Electronic copy available at: http://ssrn.com/abstract=1592247

monthly earnings on firm-size, while controlling for a variety of individual, industry and regional characteristics in Greek manufacturing. According to her results, labor productivity in firms with over 20 employees appears to be 4 percent higher than firms with over 10 employees. --Insert Table 1 about here---Insert Figure 1 about here-In their seminal work, Brown and Medoff (1989) provide several explanations of the positive relationship between wages and firm size, but still a sizeable unexplained size-wage premium remains unaccounted for. Several systematic and rigorous attempts to provide consistent explanations of the observed wage premia have followed. Using French matched employer-employee data, Abowd et al. (1999) argued that the observed wage variation between firm size categories can be almost fully explained by individual rather than firm heterogeneity. Along the same line, Winter-Ebmer and Zweinmuller (1999) provide evidence from the Swiss labor market suggesting that almost half of the (OLS) firm size premium can be explained by worker heterogeneity. Troske (1999) tested seven possible hypotheses about the firm size-wage relationship, using also matched employer–employee data, but he was not able to account adequately for the observed (OLS) premia. Furthermore, Bayard and Troske (1999) using matched employer–employee data from the US identified positive and significant firm size-related wage premia across industries. Controlling for individual characteristics and working conditions, Lallemand et al. (2005) using Belgian data identified a strong positive size-wage premium, while Barth and Dale-Olsen (2005) using Norwegian data reported a dwindling firm size effect on wages when they controlled for the number of workers in the same-skill group. Utilizing harmonized matched employer-employee data (SES,

2

1995), Lallemand et al. (2007) identified positive and significant size-wage premia across five European countries (Belgium, Denmark, Ireland, Italy and Spain), which however were volatile and negatively correlated with the degree of corporatism in each country. In this paper we explore the establishment size-wage differential utilizing the 2002 SES matched employer-employee dataset and following the methodological approach suggested by Troske (1999). We investigate the private sector as a whole and we report results separately for manufacturing and services. More specifically, we test the following hypotheses regarding the establishment size-wage premium: a) that it reflects the fact that more skilled workers tend to work together and thus more productively in larger establishments (Barron et al. 1987; Kremer and Maskin, 1996; Troske, 1999; Barth and Dale-Olsen, 2005), b) that larger establishments are more likely to follow rent-sharing policies due to greater market power (Weiss, 1966; Mellow, 1982; Stewart, 1990; Troske, 1999), c) that capital and labor are complementary in production and larger establishments are more capital intensive (Hamermesh 1980, 1993; Reilly, 1995; Troske, 1999), d) that skilled managers of larger establishments are more effective in hiring better matched workers, i.e., more suitable, (Lucas, 1978; Troske, 1999), e) that better matching leads to higher wages since it is associated with lower monitoring costs (Eaton and White, 1983; Brown and Medoff, 1989; Reilly, 1995; Albaek et al. 1998; Troske, 1999; Lallemand et al. 2007), f) that establishment-level bargaining between employers and employees results in higher wages relative to those resulting from collective agreements (Rycx, 2003; Plasman et al. 2007; Lallemand et al. 2007), g) that compensating for unpleasant working conditions and/or incentive payments schemes account for part of the observed size-wage gap (Brown and Medoff, 1989; Albaek et al. 1998; Lallemand et

3

al. 2007), h) that the within-establishment wage dispersion is bigger in large establishments due to either more heterogeneous workforce (Davis et al. 1991; Davis and Haltiwanger, 1996) or the existence of internal labor markets (Lazear and Rosen, 1981), i) that part of the observed premium is due to the presence and operation of labor unions (Mellow, 1982; Albaek et al. 1998) and j) that specific operating characteristics of each establishment (multi-unit or exporting establishment) are correlated with establishment size (Wagner, 1995; Schank et al. 2007). The paper is organized as follows: In section II we present the data sources, the definition of the main variables, the associated descriptive statistics and preliminary evidence regarding the relationship between establishment size and wages. Section III presents the econometric specification and the empirical estimation strategy. In Section IV we present and discuss the results of the econometric analysis of successive alternative model specifications. The paper concludes with Section V which contains a summary of our major findings and implications for further research.

II Data sources Matched employer-employee data are indeed the appropriate vehicle for testing several core hypotheses about establishment size-wage differentials and this is why they have been utilized extensively in related studies. We will rely on matched employer-employee data, too. For the purposes of the present study we will employ the Greek 2002 Structure of Earnings Survey (SES021 thereafter) which contains adequate information on the characteristics of both, workers and their employers. More specifically, the SES02 includes detailed salary and job information for 49.153

1

It should be mentioned that the SES was also conducted in 1995 and 2006. However, the former is relatively old and the latter lacks- unexpectedly- information on a number of variables which are required for the purposes of the present study.

4

employees in each of some 5.281 local units (establishments with more than 10 workers). The unique design of the survey allows us to model wage outcomes at the employee level while including controls for several pertinent characteristics of surveyed workers and establishments. We focus on the subsample of workers in the SES02 with non-missing data on individual, job and establishment-related characteristics. In particular, the database contains information on a) establishment characteristics such as, sector of economic activity (see Table A1 in the Appendix), geographical area, number of local units, number of persons employed, type of employment contract, type of financial auditing, main market of economic activity, and b) worker’s characteristics such as, gender, age, occupation (see Table A2 in the Appendix), type of contract, nationality, tenure in the current employer, education level, vocational training, earnings, monthly working hours and weeks worked per year. Our final sample consists of non-agricultural, private sector salaried workers, between 18 and 64 years of age. Summary statistics on several worker characteristics are presented at Table 2 for the whole sample and by establishment size. We observe that the wage and its standard deviation increase along with the establishment size. Furthermore, workers in larger establishments are, on average older, more educated, with longer tenure and they are more likely males and Greeks. Furthermore, large establishments seem to employ more skilled workers, since only 25 percent of those working in large establishments are considered to be blue-collar workers. It is also observed that sizeable differences exist between large and small establishments with respect to additional alternative forms of payment, (i.e., compensation for working conditions, productivity bonuses, profit bonuses and piece payments). Moreover, larger establishments utilize more intense monitoring and operate more than one unit. The distribution of establishments by sector of economic

5

activity for the full sample indicate that the majority of establishments are concentrated in manufacturing (33.7 percent), wholesale/retail trade and repairs etc (26.3 percent) and in hotels and restaurants (11.6 percent). This distribution remains roughly unchanged when we consider different employment-size groups (up to 499 employees). For example, establishments with 500 to 999 employees are concentrated in manufacturing (45.5 percent), wholesale/ retail trade and repairs etc (16.5 percent) and in construction (13.8 percent). With regard to the largest employment-size group (≥1000 employees) establishments are concentrated mostly in financial intermediation (28.4 percent), wholesale/retail trade and repairs etc (27.4 percent), manufacturing (19.2 percent) and transport, storage and communication (19.0 percent). --Insert Table 2 about here --

III Estimation strategy and empirical results For estimation purposes of the wage equation we adopt a typical Mincer (1974) approach. The wage equation is estimated by OLS with White (1980) heteroscedasticity-consistent standard errors while correcting for clustering due to possible biases in the estimated standard errors, stemming from the use of aggregated establishment variables in an individual wage equation. Throughout the empirical investigation, we utilize several augmented Mincerian earnings functions by including observed characteristics of both employees and employers as explanatory correlates. More specifically, the estimated functions are of the following form:

ln wi    Si   Hik γ k  Zimδm  Cin λ n  ui

(1)

where, ln wi is the natural log of the hourly wage rate for individual i , Si is the log of the median number of employees for each of the seven establishment size classes of

6

the database2, H i is a vector of k human capital variables (such as, age, age squared, years of schooling, tenure, tenure squared, gender, dummy for the country of birth),

Z i is a vector of m establishment-specific characteristics, Ci is a vector containing controls for occupation, industry and region and u is a worker-specific error term. The parameter of interest is  which corresponds to the wage elasticity with respect to establishment size. The baseline model includes (in addition to establishment size) human capital variables, controls for occupation, industry and regional variation in the utilized sample of workers. Positive values of the estimated establishment size parameter indicate that a unit increase in establishment size raises hourly wages by ˆ . In order to test the validity of the various theoretical explanations of the establishment size-wage gap we introduce control variables in a step wise mode. In line with Troske (1999), we begin our analysis by estimating equation (1) using the full sample of the SES02 workers while setting γ=0 and δ=0. Column (1) of Table 3 shows that without controls for employer and employee characteristics, there is a sizeable effect of establishment size on wages (0.067). After controlling for human capital characteristics the effect of establishment size on wages is reduced by approximately 30% but it is still sizeable (0.047) and statistically significant, (see column [2] of Table 3). Furthermore, the rest of the estimated coefficients have the expected sign and they are all statistically significant at the 1% level. --Insert Table 3 about here--

2

Supporting evidence of the validity of this common practice for estimating size-wage elasticities is provided by Albaek et al. (1998). Furthermore, due to limitations of the SES02 database on supplyside variables, we were not able to test for possible non-random allocation of workers across firms of different size (Idson and Feaster, 1990; Main and Reilly, 1992; Brunello and Colussi, 1998).

7

IV Testing hypotheses of the establishment size-wage premium To investigate the hypothesis that the workforce in larger establishments is more skilled, we estimate the baseline specification (Column [2] of Table 3), including in Z the following measures of workforce skill: the average years of potential experience of all workers in a establishment, the percentage of workers within a establishment that have at least a university degree and the percentage of workers within a firm that are scientists, engineers or technical workers. The estimated positive and significant coefficients of mean experience, percent degree and percent skilled shown in Column [3] of Table 4 indicate that indeed more skilled workers tend to work more productively in larger establishments and thus, they receive higher wages. Notice however that there is only a slight reduction of 2.13 percent in the estimated coefficient of the establishment size variable. This finding contradicts the results reported by Troske (1999) and Barth and Dale-Olsen (2007), i.e., matching better skilled workers in larger establishments cannot be accounted for as a satisfactory explanation of the establishment size-wage question. Nevertheless, better matching of workers is an important part of the wage structure. In the next step, we examine the hypothesis that establishments which operate in concentrated markets may follow of rent sharing policies. Thus, following Weiss (1966), Mellow (1982), Stewart (1990) and Troske (1999), we include in the baseline specification a Herfindahl index of concentration3 in order to examine whether market power is correlated with establishment size. The obtained estimation results (Column 4 of Table 4) indicate that concentration alters only slightly the effect of establishment

3

Due to data limitations of the SES02 dataset regarding information on financial statements, we calculated the Herfindahl index at a 2-digit classification of economic activity for the year 2002 using data from the Business Registry of the National Statistical Service of Greece regarding the value of sales at the 5-digit classification of economic activity. These values were merged with the SES02 codes of the 2-digit classification of economic activity.

8

size on wages. Thus, rent sharing practices do not constitute a satisfactory explanation of the observed establishment size-wage premium either. To test the hypothesis that capital and labor are complementary in production and larger establishments are more capital intensive, we include in Z the capital-labor ratio4. The results of this version are displayed in column [5] of Table 4 and clearly suggest that workers in capital-intensive sectors receive higher wages. Once again however, the explanatory power of this hypothesis is rather trivial since it accounts for a very slight reduction in the coefficient of the establishment size variable. This finding is quite similar to that in Troske (1999). The next hypothesis concerns the level of collective bargaining. It is hypothesized that employees covered by a establishment-level contractual agreement receive, ceteris paribus, higher wages than those covered by collective agreements at broader levels. The implied assumption here is that larger establishments use more extensively establishment-level agreements which are more generous. In line with the findings of Lallemand et al. 2007, the estimation results (column [6] of Table 4) reveal clearly that establishment-level bargaining cannot contribute in the explanation of the establishment size-wage gap. Indeed, the establishment size-wage elasticity remains unaffected. Nevertheless, workers who bargain at the establishment level are paid, on average, higher wages (Rycx, 2003; Plasman et al. 2007; Lallemand et al. 2007). Next we explore whether wages are affected by specific establishment operating characteristics which are also correlated with establishment size. More specifically, this time Z contains two dummy variables indicating whether the establishment is a global exporter or a multiunit firm. Similar attempts were 4

Data on gross capital stock and on the number of employees have been provided by the National Statistical Service of Greece at a 2-digit classification of economic activity.

9

undertaken by Wagner (1995) and Schank et al. (2007). Again, these operating characteristics do not seem to provide a convincing explanation of the size-wage puzzle. And this despite the fact, that employees working in globally exporting establishments appear to receive higher wages. Another well-tested hypothesis in the related literature is that larger establishments compensate their workers for ―unpleasant‖ working conditions and/or utilize incentive payments schemes, i.e., alternative forms of noncompetitive wage rewards. In order to test this hypothesis we include in Z, establishment-specific variables regarding the percentage of workers who receive (i) payments for overtime work, (ii) various forms of special payments, (iii) productivity bonuses, (iv) profit bonuses and (v) piece payments. Additionally, we include the percentage of workers who work under (vi) a limited employment contract and (vii) an apprentice or trainee contract. According to the results presented in Column [8] of Table 4, the estimated coefficient of the size variable has suffered a dramatic decrease by around 38 percent. This finding is in line with the results of Lallemand et al. (2007) for Denmark, Italy and Spain. Thus, non competitive wage rewards appear to contribute substantially in the explanation of the observed size-wage premium. This finding indicates that alternative forms of compensation are primarily used by larger establishments. Next we investigate the hypothesis that larger establishments employ more heterogeneous workforce (Davis and Haltiwanger, 1996) and thus, wages are more dispersed. Wage dispersion can also be observed if internal labor markets within each establishment are present (Lazear and Rosen, 1981). In this case, vector Z contains a measure of within-establishment wage dispersion which was calculated by subtracting the average (ln)wage of workers belonging in the first quartile -of the (ln)wage distribution within each establishment- from the average (ln)wage of those in the

10

fourth quartile. Again, the within-establishment wage dispersion appears to account for almost 38 percent of the observed size-wage premium (column [9] in Table 4), indicating that wages are indeed more dispersed in larger establishments (Krueger and Summers, 1988; Davis and Haltiwanger, 1996). The explanatory power of this hypothesis is in line with our previous finding which showed that compensation in larger establishments is also based on incentive pay mechanisms (i.e., productivity bonuses) and thus wage dispersion, compensating wage differentials, efficiency wage theories and tournament mechanisms could be in effect. To investigate the hypothesis proposed by Lucas (1978) and tested by Troske (1999), i.e., that larger establishments are managed by better-skilled managers who in turn hire better-skilled workers, we have to limit our sample only to non-managerial employees5. In other words, we test the hypothesis that non-managerial workers in larger establishments receive a wage premium because they are working under betterskilled managers. Thus, we include in Z the following measures of managerial skill: (i) the average experience of all managers within a establishment and (ii) the percentage of managers within each establishment who hold at least one university degree. The results are shown in columns [10a] and [10b] of Table 5. As in Troske (1999), it seems that better-skilled managers do hire better-skilled workers but this only marginally explains the observed premium received by non-managerial employees. Next we examine the hypothesis that larger establishments pay higher wages in an effort to reduce monitoring costs. Following Reilly (1995) and Troske (1999) we include in the baseline specification a variable indicating the percentage of supervisory workers within each establishment. Consequently, our sample is now 5

As in Troske (1999), non-managerial workers are those who are not classified as managers or as other professionals even if they hold a supervisory position within the establishment.

11

restricted to only nonsupervisory workers6. The estimation results presented at columns [11a] and [11b] of Table 5, indicate that indeed nonsupervisory workers enjoy a sizeable wage premium but the within-establishment monitoring intensity does not appear to contribute in the explanation of the observed size-wage premium. In other words, monitoring intensity does appear to exercise a significant effect on the wage structure but it is rather uncorrelated with establishment size, a finding practically identical with that in Reilly (1995). Finally, we investigate whether sector-specific union density7 can contribute in the explanation of the observed establishment size-wage premium. This hypothesis arises from the study of Lallemand et.al. (2007) who report -using data from five European countries- a significant correlation between the size-wage premium and the union density. The results of this exercise appear in columns 12a and 12b of Table 5 and indicate that sector-specific union density does not play a role in the formulation of size related wage premia. --Insert Table 4 about here---Insert Table 5 about here--

V Analysis by sector of economic activity The analysis of the previous section is based on the assumption that the employer-size wage premium is sector-indifferent. That is, the manufacturing and services sectors are practically homogeneous as far as the size-wage premium is concerned. However, it is common knowledge that significant differences characterize the production

6

As non supervisory workers are classified those who are not managers, professionals and do not hold a supervisory position. 7 Due to lack of data regarding union status in the SES02 database, the variable for union density derives from sectoral data [from the 30th Congress of the Greek General Confederation of Labor (GSEE), 2001] on the total number of unionized members. The union density is the ratio of unionized workers to the total number of employees in each sector.

12

processes of the two sectors in question (Idson and Oi, 1999). In this section, we repeat the analysis conducted so far separately for the manufacturing and the services sectors. The estimation results for manufacturing are presented in Tables 6, 7 and 8. We observe that the estimated coefficient of the establishment size variable (baseline specification with human capital and other control variables included), is around 7.7 percent (Table 6) and it is substantially larger than the corresponding estimate for the total economy (4.7 percent) reported in Table 3. Testing of the alternative explanations considered here indicates that the only variables that seem to affect the size-wage premium are those representing working conditions/ incentive payment schemes and the within-establishment wage dispersion. In particular, when the set of working conditions/incentive payments were included in the baseline specification, the size-wage premium has been reduced by around 34 percent. Similarly, the inclusion of the within-establishment wage dispersion reduced the premium by around 18 percent. However, the magnitude of the reduction is smaller than the one reported (total economy) in Table 3 (Columns 8 and 9) which was around 38 percent in both cases. --Insert Table 6 about here---Insert Table 7 about here---Insert Table 8 about here— Turning now to the services sector, the estimation results are presented in Tables 9, 10 and 11. In the baseline specification, the estimated size-wage premium is around to 4.3 percent (Table 9) and it is lower than the corresponding estimates for both the total economy (4.7 percent) and the manufacturing sector (7.7 percent). With regard to the alternative explanations of the estimated size-wage premium, the results appear to be different than those obtained so far. Again, working conditions/incentive

13

payments or within-establishment wage dispersion exert the greatest negative influence (37 percent and 46 percent, respectively). However, additional factors appear to play a role. For example, the effect of size has been significantly reduced by almost 14 percent when we controlled for the workforce skill, by 28 percent for market power, by 20 percent for managerial skill and by 26 percent for union density. Apparently, the wage structure and the relationship between employer size and employee wages are different in the two sectors. --Insert Table 9 about here---Insert Table 10 about here---Insert Table 11 about here--

VI Conclusions The scope of this paper was to investigate the establishment-size wage premium in Greece. For estimation purposes we employed a) the Greek Structure of Earnings Survey of 2002, a matched employer-employee dataset which contains a rich set of characteristics for both workers and their employers and b) a Mincer-type wage specification. Overall, our results suggest the existence of a significant employer-size wage premium in the Greek economy as well. The magnitude of this premium (0.047) is in line with those reported in the pertinent international literature and several traditional hypotheses have been tested in order to explain it. Furthermore, in the present study we have also tested directly two additional explanations, i.e., withinestablishment wage dispersion and union density. Our results indicate that the estimated establishment-size wage premium can be primarily explained primarily by factors of the neoclassical paradigm, i.e., human capital, compensating wage differentials, efficiency wages and internal labor markets. 14

Our analysis shows further that the magnitude of the premium varies between manufacturing and services. In fact, the magnitude of the premium is higher in the manufacturing sector, indicating that establishments in this sector enjoy economics of scale which eventually lead to higher wages, ceteris paribus. Nevertheless, a sizable part of the wage premium still remains unaccountable. Future research regarding the relationship between wages and establishment size in the Greek economy should address the issue of self selection, i.e., the choice of workers regarding the size of their potential employer is not random. It should also address the determinants of observed establishment sizes, given the high concentration of Greek establishments in small sizes and the productivity gap between small and large establishments.

15

References Abowd J.M., Kramarz F. and Margolis D. N. (1999), High Wage Workers and High Wage Firms, Econometrica, Vol. 67, No. 2, pp. 251-333. Albaek, K., Arai, M., Asplund, R., Barth, E., Madsen, E. (1998), Measuring Wage Effects of Plant Size, Labour Economics, Vol. 5, pp. 425-448. Barron, J.M, Black, D.A and Loewenstein, M.A. (1987), Employer Size: The Implications for Search, Training, Capital Investment, Starting Wages and Wage Growth, Journal of Labour Economics, Vol. 5, No. 1, pp. 76-89. Barth, E. and Dale-Olsen, H. (2005), Employer Size or Skill-Group Size Effect on Wages?, IZA Discussion Paper No. 1888, Institute for the Study of Labor (IZA). Bayard, K. and Troske, K.R. (1999), Examining the Employer-Size Wage Premium in the Manufacturing, Retail Trade, and Service Industries Using Employer-Employee Matched Data, The American Economic Review, Vol. 89, No. 2, pp. 99-103. Blanchflower, D. (1986), Wages and Concentration in British Manufacturing, Applied Economics, 18, pp. 1025-1038. Brown, C. and Medoff, J. (1989), The Employer Size-Wage Effect, The Journal of Political Economy, Vol.97, No. 5, pp. 1027-1059. Brunello, G. and Colussi, A. (1998), The Employer Size-Wage Effect: Evidence from Italy, Labour Economics, Vol. 5, pp. 217-230. Burtless, G. (2002), The Greek labour market, in R.C. Bryant, N.C. Garganas and G.S. Tavlas ,eds., Greece’s economic performance and prospects, Bank of Greece & The Brookings Institution, Athens. Davis, J.D., Haltiwanger, J., Katz, L.F. and Topel, R. (1991), Wage Dispersion Between and Within U.S. Manufacturing Plants, 1963-86, Brookings Papers on Economic Activity. Microeconomics, pp. 115-200. Davis, J.D. and Haltiwanger, J. (1996), Employer Size and the Wage Structure in U.S. Manufacturing, Annales D’ Economie et de Statistique, No. 41/42, pp. 324-367. Eaton, C. and White W.D. (1983), The Economy of High Wages: An Agency Problem, Economica, Vol. 50, No. 198, pp. 175-181. Gerlach, K. and Schmidt, E.M. (1990), Firm Size and Wages, Labour, Vol. 4, No. 2, pp. 27-49. Hamermesh, D.S., "Commentary" In John J. Siegfried (ed.), The Economics of Firm Size, Market Structure and Social Performance (Washington, D.C.: Federal Trade Commission, 1980). Hamermesh, D.S., Labor Demand (Princeton, NJ: Princeton University Press, 1993). 16

Idson, T.L. and Feaster, D.J. (1990), A Selectivity Model of Employer-Size Wage Differentials, Journal of Labour Economics, Vol. 8, No. 1, pp. 99-122. Idson, T.L. and Oi, W.Y. (1999), Workers Are More Productive in Large Firms, The American Economic Review, Vol. 89, No. 2, pp. 104-108. Kremer, M. and Maskin, E. (1996), Wage Inewuality and Segregation by Skill, NBER Working Paper Series, Working Paper No. 5718, National Bureau of Economic Research. Krueger, A.B. and Summers, L.H. (1988), Efficiency Wages and the Inter-Industry Wage Structure, Econometrica, Vol. 56, No. 2, pp. 259-293. Kumar, K.B, Rajan, R.G., and Zingales, L. (2001), What Determines Firm Size?, CRSP Working Paper No. 496; and USC Finance & Business Econ. Working Paper No. 01-1. Lallemand, T., Plasman, R. and Rycx, F. (2007), The Establishment-Size Wage Premium: Evidence from European Countries, Empirica, Vol. 34, pp. 427-451. Lallemand, T., Plasman, R. and Rycx, F. (2005), Why do Large Firms Pay Higher Wages? Evidence from Matched Worker-Firm Data, International Journal of Manpower, Vol. 26, No. 7/8, pp. 705-723. Lazear, E.P. and Rosen, S. (1981), Rank-Order Tournaments as Optimum Labour Contracts, Journal of Political Economy, Vol. 89, No. 5, pp. 841-864. Lucas, R.E, Jr. (1978), On the Size Distribution of Business Firms, The Bell Journal of Economics, Vol. 9, No. 2, pp. 508-523. Main, B.G.M. and Reilly, B. (1993), The Employer Size-Wage Gap: Evidence from Britain, Economica, Vol. 60, No. 238, pp. 125-142. Mellow, W. (1982), Employer Size and Wages, The Review of Economics and Statistics, Vol. 64, No. 3, pp. 495-501. Mincer, J. (1974). Schooling, Experience and Earnings. New York: National Bureau of Economic Research. Nicolitsas, D. (2007), Growth, Jobs and Structural Reforms in Greece, in R. Tilly, P.J. J. Welfens and M. Heise, eds., 50 Years of EU Economic Dynamics Integration, Financial Markets and Innovations, Springer Berlin Heidelberg. Plasman, R., Rusinek, M. and Rycx, F. (2007), Wages and the Bargaining Regime under Multi-level Bargaining: Belgium, Denmark and Spain, European Journal of Industrial Relations, Vol. 13, No. 2, pp. 161-180. Reilly, K.T. (1995), Human Capital and Information: The Employer Size-Wage Effect, Journal of Human Resources, Vol. 30, No. 1, pp. 1-18. 17

Rycx, F. (2003), Industry Wage Differentials and the Bargaining Regime in a Corporatist Country, International Journal of Manpower, Vol. 24, pp. 347-366. Schank, T., Schnabel, C. and Wagner, J. (2007), Do Exporters Really Pay Higher Wages? First Evidence from German Linked Employer-Employee Data, Journal of International Economics, Vol. 72, pp. 52-74. Schmidt, C.M. and Zimmermann, K.F. (1991), Work Characteristics, Firm Size and Wages, The Review of Economics and Statistics, Vol. 73, No. 4, pp. 705-710. Stewart, M.B. (1990), Union Wage Differentials, Product Market Influences and the Division of Rents, The Economic Journal, Vol. 100, pp. 1122-1137. Troske, K.R. (1999), Evidence on the Employer Size-Wage Premium from WorkerEstablishment Matched Data, The Review of Economics and Statistics, Vol. 81, No. 1, pp. 15-26. Wagner, J. (1995), Exports, Firm Size and Firm Dynamics, Small Business Economics, Vol. 7, pp. 29-39. Weiss, L.W. (1966), Concentration and Labor Earnings, American Economic Review, Vol. 56, No. 1/2, pp. 96-117. White, H. (1980). A Heteroscedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroscedasticity, Econometrica Vol. 48, pp. 817–838. Winter-Ebmer R. and Zweimüller J. (1999), Firm-Size Wage Differentials in Switzerland: Evidence from Job-Changers, The American Economic Review, Vol. 89, No. 2, pp. 89-93.

18

Tables

Table 1. The distribution of firms by employment size in Greece, 2002 Size class Number of Percent Percent (size ≥10 ) firms 0-4 844.917 96.09 5-9 17.713 2.01 10-19 8.588 0.98 51.5 20-29 2.908 0.33 17.4 30-49 2.335 0.27 14.0 50-99 1.534 0.17 9.2 ≥100 1.323 0.15 7.9 Total 879.318 100.00 100.00 Source: Business Register 2002, National Statistical Service of Greece (NSSG, www.statistics.gr)

Table 2. Descriptive statistics for the full sample and by employer size Full sample [10-19] [20-49] [50-99] [100-249] [250-499] [500-999] [≥1000] Gross hourly wage (€) 8.045 6.081 6.898 7.727 8.187 8.733 9.048 10.299 Standard deviation of gross hourly wage 5.085 3.451 4.198 4.715 5.167 5.167 5.399 6.035 Employee’s age 36.734 35.283 36.496 36.533 36.776 37.974 37.148 37.530 Years of schooling 11.595 11.008 11.293 11.643 11.696 11.451 11.570 12.403 Tenure in establishment (yrs) 6.605 4.204 5.020 5.661 6.056 8.049 7.251 10.539 Female (0/1) 0.402 0.405 0.400 0.386 0.399 0.380 0.395 0.430 Foreign (0/1) 0.057 0.121 0.087 0.049 0.037 0.037 0.022 0.017 Blue collar worker (0/1, ISCO 71-93) 0.402 0.436 0.456 0.402 0.416 0.468 0.428 0.257 Paid overtime (0/1) 0.159 0.029 0.067 0.117 0.145 0.250 0.262 0.318 Shift work, night/weekend work (0/1) 0.142 0.061 0.080 0.115 0.181 0.184 0.238 0.200 Productivity bonus (0/1) 0.143 0.034 0.062 0.108 0.141 0.164 0.111 0.357 Profit bonus (0/1) 0.022 0.003 0.002 0.001 0.007 0.005 0.016 0.103 Piece payments (0/1) 0.066 0.009 0.021 0.025 0.047 0.074 0.027 0.225 Unlimited employment contract (0/1) 0.907 0.911 0.928 0.892 0.893 0.903 0.910 0.904 Limited employment contract (0/1) 0.089 0.087 0.069 0.101 0.097 0.088 0.089 0.095 Apprentice/trainee contract (0/1) 0.004 0.002 0.002 0.007 0.010 0.008 0.001 0.002 Supervisory worker (0/1) 0.122 0.067 0.102 0.120 0.129 0.125 0.133 0.180 Multi-unit establishment (0/1) 0.560 0.224 0.322 0.447 0.575 0.732 0.904 0.931 Establishment level bargaining (0/1) 0.096 0.100 0.092 0.097 0.122 0.113 0.070 0.079 Global exporter (0/1) 0.153 0.079 0.117 0.137 0.228 0.245 0.302 0.084 Mining and quarrying (0/1) 0.007 0.004 0.017 0.004 0.007 0.021 0.000 0.000 Manufacturing (0/1) 0.337 0.279 0.335 0.349 0.391 0.504 0.455 0.192 Electricity, gas, and water supply (0/1) 0.001 0.001 0.001 0.001 0.003 0.000 0.000 0.000 Construction (0/1) 0.056 0.055 0.060 0.073 0.047 0.051 0.138 0.010 Wholesale and retail trade; repair etc (0/1) 0.263 0.346 0.281 0.250 0.254 0.184 0.165 0.274 Hotels and restaurants (0/1) 0.116 0.189 0.156 0.113 0.128 0.080 0.110 0.025 Transport, storage and communication (0/1) 0.079 0.054 0.059 0.085 0.045 0.048 0.018 0.190 Financial intermediation (0/1) 0.061 0.002 0.006 0.014 0.018 0.013 0.029 0.284 Real estate, renting and business activities (0/1) 0.080 0.071 0.084 0.113 0.107 0.099 0.083 0.025 Number of employees 40393 5626 7753 5069 5990 3896 2908 9151 Number of establishments 2667 536 857 434 557 171 64 48 Source: Structure of Earnings Survey-2002 (SES02), National Statistical Service of Greece (NSSG). The SES02 includes establishment with more than 9 employees. All numbers have been weighted by sampling weights.

Table 3. Individual (ln)wage regressions for the full sample controlling for human capital (ages 18-64)

Constant Establishment size (ln) Age Age2/100 Years of schooling Tenure Tenure2/100 Female Foreign F-test (region) F-test (occupation) F-test (industry) R-squared Observations

Establishment-specific variables [1] 1.410 (.011)*** .067 (.002) *** 62.41*** 734.67*** 90.34*** .305 40064

Individual and establishment variables [2] .286 (.037)*** .047 (.002)*** .041 (.002)*** -.037 (.002)*** .025 (.001)*** .016 (.001)*** -.006 (.003)*** -.163 (.005)*** -.066 (.012)*** 76.64*** 250.37*** 54.71*** .469 40064

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES02 database. Source: Structure of Earnings Survey, 2002 (SES02).

Table 4. Individual (ln)wage regressions for the full sample, controlling for establishment-specific characteristics (ages 18-64)

Constant Establishment size (ln) Mean experience Percent degree Percent skilled Herfindahl index K/L ratio Establishment-level bargaining Multiunit establishment (0/1) Global exporter (0/1) Paid overtime work Receive special payments Productivity bonus Profit bonus Piece payments Limited employment contract Apprentice/trainee contract Within-establishment wage dispersion R-squared Observations

Workforce Skill

Market Power

K/L ratio

Bargaining level

[3] .239 (.037)*** .046 (.002)*** .004 (.001)*** .467 (.022)*** .070 (.012)*** .486 40064

[4] .268 (.037)*** .045 (.002)*** .186 (.018)*** .472 40064

[5] .282 (.036)*** .046 (.002)*** .149 (.031)*** .470 40064

[6] .280 (.037)*** .047 (.002)*** .090 (.008)*** .472 40064

Market operating characteristics [7] .288 (.037)*** .047 (.002)*** -.016 (.006)*** .071 (.007)** .472 40,64

Working conditions [8] .353 (.037)*** .029 (.002)*** .099 (.013)*** .044 (.010)*** .170 (.010)*** -.181 (.018)*** .031 (.010)*** .161 (.014)*** -.034 (.042) .485 40064

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Structure of Earnings Survey, (SES02).

Withinestablishment wage dispersion [9] .162 (.037)*** .029 (.002)*** .341 (.010)*** .513 40064

Table 5. Individual (ln)wage regressions on non-managerial, nonsupervisory workers and sectors with union activity (ages 18-64)

Constant Establishment size (ln) Mean managerial experience Managers with university degree(s) Percent supervisory Union density R-squared Observations

Non-Managerial [10] Only size Size plus managerial skill [a] [b] .405 (.037)*** .414 (.037)*** .048 (.002)*** .047 (.002)*** .008 (.002)*** .515 (.080)*** .392 .397 35126

Non-Supervisory [11] Only size Size plus percent supervisory [a] [b] .483 (.038)*** .478 (.038)*** .045 (.002)*** .043 (.002)*** .217 (.034)** .369 .371 32886

All workers [12] Only size Size plus sectoral union density [a] [b] .045 (.002)*** .477

.044 (.002)*** .193 (.019)*** .478 32083

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Greek Structure of Earnings Survey, (SES02).

Table 6. Individual (ln)wage regressions for manufacturing controlling for human capital (ages 18-64)

Constant Establishment size (ln) Age Age2/100 Years of schooling Tenure Tenure2/100 Female Foreign F-test (region) F-test (occupation) F-test (industry) R-squared Observations

Only establishment-specific variables [1] 1.208 (.017) .107 (.004)*** 40.51 257.24 .335 14059

Individual and establishment variables [2] .384 (.069)*** .077 (.003)*** .033 (.003)*** -.029 (.004)*** .019 (.002)*** .016 (.002)*** -.014 (.005)*** -.211 (.008)*** -.089 (.023) 35.96 129.14 .494 14059

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES02 database. Source: Greek Structure of Earnings Survey, 2002 (SES02).

Table 7. Individual (ln)wage regressions for manufacturing, controlling for establishment-specific characteristics (ages 18-64)

Constant Establishment size (ln) Mean experience Percent degree Percent skilled Herfindahl index K/L ratio Establishment-level bargaining Multiunit establishment (0/1) Global exporter (0/1) Paid overtime work Receive special payments Productivity bonus Profit bonus Piece payments Limited employment contract Apprentice/trainee contract Within-establishment wage dispersion R-squared Observations

Workforce Skill

Market Power

K/L ratio

Establishmentlevel bargaining

[3] .352 (.072)*** .073 (.003)*** .003 (.001)*** .344 (.034)*** .035 (.017)** .501 14059

[4] .384 (.066)*** .074 (.003)*** .156 (.029)*** .498 14059

[5] .358 (.068)*** .072 (.003)*** .556 (.112)*** .496 14059

[6] .393 (.069)*** .074 (.003)*** .105 (.012)*** .498 14059

Market operating characteristics [7] .382 (.069)*** .078 (.004)*** -.017 (.009)* .008 (.010) .494 14059

Working conditions [8] .433 (.069)*** .051 (.004)*** .105 (.019)*** .059 (.018)*** .205 (.015)*** .013 (.056) .058 (.016)*** .194 (.031)*** -.016 (.085) .513 14059

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Greek Structure of Earnings Survey, 2002 (SES02).

Withinestablishment wage dispersion [9] .293 (.068)*** .063 (.004)*** .203 (.020)*** .508 14059

Table 8. Individual (ln)wage regressions for manufacturing on non-managerial, nonsupervisory workers and sectors with union activity (ages 18-64)

Constant Establishment size (ln) Mean managerial experience Managers with university degree(s) Percent supervisory Union density R-squared Observations

Non-Managerial [10] Only size Size plus managerial skill [a] [b] .489 (.071) *** .495 (.070) *** .074 (.003) *** .072 (.003) *** .010 (.004) ** .353 (.127) *** .426 .428 12689

Non-Supervisory [11] Only size Size plus percent supervisory [a] [b] .514 (.072) *** .512 (.073) *** .072 (.004) *** .071 (.004) *** .119 (.052) ** .411 .413 12021

All workers [12] Only size Size plus sectoral union density [a] [b] .191 (.088)*** .155 (.090)*** .074 (.004) *** .075 (.004) *** .319 (.090)*** .506 .507 8861

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Greek Structure of Earnings Survey, 2002 (SES).

Union density has been calculated by using data by sector of economic activity [from the 30th Congress of the Greek General Confederation of Labour (GSEE), 2001] on the total number of unionized members to the total number of employees in each sector.

Table 9. Individual (ln)wage regressions for services controlling for human capital (ages 18-64)

Constant Establishment size (ln) Age Age2/100 Years of schooling Tenure Tenure2/100 Female Foreign F-test (region) F-test (occupation) F-test (industry) R-squared Observations

Only establishment-specific variables [1] 1.380 (0.13)*** .068 (.002)*** 38.44 602.81 .278 25691

Individual and establishment variables [2] .237 (.044)*** .043 (.002)*** .044 (.002)*** -.040 (.003)*** .027 (.013)*** .014 (.001)*** .007 (.004) -.137 (.006)*** -.038 (.014)** 56.26 189.13 .458 25691

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the 2002 SES database. Source: Greek Structure of Earnings Survey, 2002 (SES02).

Table 10. Individual (ln)wage regressions for services, controlling for establishment-specific characteristics (ages 18-64)

Constant Establishment size (ln) Mean experience Percent degree Percent skilled Herfindahl index K/L ratio Establishment-level bargaining Multiunit establishment (0/1) Global exporter (0/1) Paid overtime work Receive special payments Productivity bonus Profit bonus Piece payments Limited employment contract Apprentice/trainee contract Within-establishment wage dispersion R-squared Observations

Workforce Skill

Market Power

K/L ratio

Establishmentlevel bargaining

[3] .212 (.044)*** .037 (.002)*** .006 (.001)*** .473 (.027)*** .102 (.018)*** .482 25691

[4] .241 (.044)*** .031 (.002)*** .267 (.014)*** .470 25691

[5] .231 (.044)*** .042 (.002)*** .064 (.024)*** .459 25691

[6] .224 (.045)*** .044 (.002)*** .060 (.010)*** .460 25691

Market operating characteristics [7] .228 (.044)*** .047 (.002)*** -.027 (.007)*** .094 (.009)*** .462 25691

Working conditions [8] .302 (.045)*** .027 (.002)*** .074 (.020)*** .024 (.011)*** .138 (.014)*** -.084 (.017)*** .035 (.013)*** .187 (.015)*** -.091 (.045)** .472 25691

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Greek Structure of Earnings Survey, 2002 (SES02).

Withinestablishment wage dispersion [9] .108 (.043)*** .023 (.002)*** .402 (.012)*** .520 25691

Table 11. Individual (ln)wage regressions for services on non-managerial, nonsupervisory workers and sectors with union activity (ages 18-64)

Constant Establishment size (ln) Mean managerial experience Managers with university degree(s) Percent supervisory Union density R-squared Observations

Non-Managerial [10] Only size Size plus managerial skill [a] [b] .362 (.045)*** .576 (.046)*** .044 (.002)*** .035 (.002)*** .008 (.003)*** .503 (.098)*** .369 .394 22140

Non-Supervisory [11] Only size Size plus percent supervisory [a] [b] .484 (.045)*** .478 (.044)*** .039(.002)** * .037 (.002)*** .353 (.045)*** .343 .347 20595

All workers [12] Only size Size plus sectoral union density [a] [b] .281 (.046)*** .257 (.045)*** .046 (.002)*** .034 (.002)*** .463 .474 22908

Robust standard errors in parentheses. All regressions include regional, industry and occupational dummies. Asterisks ***, ** and * indicate statistical significance at 1%, 5% and 10%, respectively. All results are weighted using the weight variable included in the SES database. All specifications include the baseline explanatory variables presented on column 2 of Table 3. Source: Greek Structure of Earnings Survey, 2002 (SES).

Union density has been calculated by using data by sector of economic activity [from the 30th Congress of the Greek General Confederation of Labour (GSEE), 2001] on the total number of unionized members to the total number of employees in each sector.

Figures Figure 1. Correlation coefficients between sector-specific labor productivity and the sectoral distribution by firm size.

Note: The data for Gross Value Added and the total number of annual work hours by sector (1-digit) for 2002 were drawn from the Greek part of the STAN Database for Industrial Analysis (OECD). The data for the distribution of firms (1-digit) by employment size groups were drawn from the Business Registry (NSSG, 2002).

Appendix Table A1. ISIC Rev.3 (International Standard Industrial Classification of all Economic Activities) 1 digit C Mining and quarrying

D Manufacturing

E Electricity F Construction G Wholesale and retail trade; repair of motor vehicles, motorcycles and personal and household goods H Hotels and restaurants I Transport, storage and communications

J Financial intermediation

K Real estate

2 digit C 11 Extraction of crude petroleum and natural gas; service activities incidental to oil and gas extraction excluding surveying C 13 Mining of metal ores C 14 Other mining and quarrying D 15 Manufacture of food products and beverages D 16 Manufacture of tobacco products D 17 Manufacture of textiles D 18 Manufacture of wearing apparel; dressing and dyeing of fur D 19 Tanning and dressing of leather; manufacture of luggage, handbags, saddlery, harness and footwear D 20 Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and D 21 Manufacture of paper and paper products D 22 Publishing, printing and reproduction of recorded media D 23 Manufacture of coke, refined petroleum products and nuclear fuel D 24 Manufacture of chemicals and chemical products D 25 Manufacture of rubber and plastics products D 26 Manufacture of other non-metallic mineral products D 27 Manufacture of basic metals D 28 Manufacture of fabricated metal products, except machinery and equipment D 29 Manufacture of machinery and equipment n.e.c. D 30 Manufacture of office, accounting and computing machinery D 31 Manufacture of electrical machinery and apparatus n.e.c. D 32 Manufacture of radio, television and communication equipment and apparatus D 33 Manufacture of medical, precision and optical instruments, watches and clocks D 34 Manufacture of motor vehicles, trailers and semi-trailers D 35 Manufacture of other transport equipment D 36 Manufacture of furniture; manufacturing n.e.c. D 37 Recycling E 40 Electricity, gas, steam and hot water supply E 41 Collection, purification and distribution of water F 45 Construction G 50 Sale, maintenance and repair of motor vehicles and motorcycles; retail sale of automotive fuel G 51 Wholesale trade and commission trade, except of motor vehicles and motorcycles G 52 Retail trade, except of motor vehicles and motorcycles; repair of personal and household goods H 55 Hotels and restaurants I 60 Land transport; transport via pipelines I 61 Water transport I 62 Air transport I 63 Supporting and auxiliary transport activities; activities of travel agencies I 64 Post and telecommunications J 65 Financial intermediation, except insurance and pension funding J 66 Insurance and pension funding, except compulsory social security J 67 Activities auxiliary to financial intermediation K 70 Real estate activities K 71 Renting of machinery and equipment without operator and of personal and household goods K 72 Computer and related activities K 73 Research and development K 74 Other business activities

Table A2. ISCO88 (International Standard Classification of Occupations) 1 2

1 digit Legislators, senior officials and managers Professionals

3

Technicians and associate professionals

4

Clerks

5

Service workers and shop and market sales workers Skilled agricultural and fishery workers Craft and related trade workers

6 7

8

Plant and machine operators and assemblers

9

Elementary occupations

12 13 21 22 23 24 31 32 33 34 41 42 51 52 61

2 digit Corporate managers Managers of small enterprises Physical, mathematical and engineering science professionals Life science and health professionals Teaching professionals Other professionals Physical and engineering science associate professionals Life science and health associate professionals Teaching associate professionals Other associate professionals Office clerks Customer services clerks Personal and protective services workers Models, salespersons and demonstrators Skilled agricultural and fishery workers

71 72 73 74 81 82 83 91 92 93

Extraction and building trades workers Metal, machinery and related trades workers Precision, handicraft, craft printing and related trades workers Other craft and related trades workers Stationary plant and related operators Machine operators and assemblers Drivers and mobile plant operators Sales and services elementary occupations Agricultural, fishery and related laborers Laborers in mining, construction, manufacturing and transport