1

Factors behind performance-related pay: evidence from Italy

Mirella Damiani* and Andrea Ricci**

Abstract

This paper examines the main determinants of the adoption of performance-related pay (PRP) in Italy. The analysis is performed by merging cross-sectional information on the spread of PRP collected for 2005 by the ISFOL survey with balance sheet data, provided by the AIDA database for a panel of firms, sampled between 2002 and 2005. By performing the cross-sectional analysis for 2005 we find the following results. Firstly, in relation to company characteristics, the probability of adoption of PRP depends on firms’ past performance in terms of value added, but not on capital accumulation and innovation activity. Secondly, in relation to workers’ characteristics, the adoption of PRP increases with the proportion of trained workforce, and with that of employees with fixed-term contracts, whereas it decreases with the female share of the workforce. Lastly, the presence of unions at establishment level and strike activities have positive impacts on the probability of the adoption of PRP.

JEL. Classification: J33; J50 Keywords: Performance-related-pay *Department of Economics, Finance and Statistics, University of Perugia ** ISFOL, Roma

Corresponding author: Mirella Damiani, Department of Economics, Finance and Statistics, University of Perugia, Via Pascoli, 20, 06123 Perugia, tel.+39-0755855424; email: mirella.damiani@gmail.com

2

1. Introduction In the mid-1990s, the Italian economy began to reveal sizeable slowing in labour productivity growth, on a scale without equal in other European economies. In the period 1995-2005, the product per employee increased by only 0.6%, as against the much higher mean value for the 13-member European Union, which reached 2.3%. A significant part of this phenomenon, the greatest in quantitative terms, was due to Total Factor Productivity (TFP), which even reached negative variation rates (-0.46%), as opposed to a mean positive figure of 0.42% in the 13-member EU (van Ark et al. 2008). An indicator like TFP is known to capture several variables including, among others, organisational and institutional changes which affect the efficiency of combined production inputs. Growth accounting thus reveals that the problem underlying the structural crisis of the Italian economy is not only one of limited accumulation of physical and human capital. Some of the stagnation is due to an organisational design which penalises the innovative capabilities of Italian firms. This is truly paradoxical in a country which had begun a process of large-scale reform of the industrial relations system since the economic and financial crisis in the early 1990s. These reforms were aimed at providing more space to wage compensation schemes linked to efficiency gains, and thus promoting reorganisation and innovation of productive processes. This unexpected result makes Italy an interesting case study in understanding the potentials and limitations of reforms of incentive systems. The main aim of the present work is to ascertain, more than ten years after the 1993 agreement, which types of firms and workers are covered by the new forms of labour relations, and to identify the main factors leading to the adoption of incentive company bonuses. Although there have been some studies on the Italian economy which have explored decisions to adopt performance-related contracts, all available research is limited to selected sectors, northern areas, or restricted to large companies (see Biagioli and Del Boca, 2004, for a summary). No works that consider the probability of adopting these schemes have so far been carried out on a national scale. The present study is the first to provide estimates based on a nationally representative sample of manufacturing and services companies. Information was collected from the 2005 ISFOL survey on the spread of PRP, but also on unionisation and the presence of works councils at establishment level, together with business activities (training, innovation, foreign exposure, use of fixed-term contracts) and workforce composition (by gender and employment status). This wide set of cross-sectional information is merged with balance sheet data, provided by the AIDA database for a panel of firms sampled between 2002 and 2005. It reveals how the spread of PRP and other firm characteristics are related to company performance over the period 2002-2005. The paper is structured as follows. Section 2 offers a brief overview of some properties of incentive pay schemes in a multiple bargaining framework and discusses some

3

problematic issues on the basis of international evidence. Section 3 outlines the basic features of the RIL-AIDA dataset and presents the descriptive and econometric results of the determinants of the adoption of PRP schemes. Section 4 concludes. 2. Company-level wage agreements: theory and empirical evidence The Italian bargaining regulations, approved in 1993, designed a new framework of multiple-level negotiations and set in motion intensification of decentralised wage bargaining. In this institutional structure, the promotion of performance-related payments was expected to induce micro-benefits in terms of productivity growth. A general claim of economic literature on two-tier wage regimes is that company wage bargaining operates in addition to central or sectoral collective agreements to enhance the degree of flexibility of the compensation package. Under this broad assumption, various classes of models have been hypothesised: efficiency wage systems, distribution of rents, risk-sharing devices. Our primary aim is to detect which of the above models has had greater influence in the Italian experience, and thus to evaluate whether the 1993 Agreement, which was intended to amplify the incentive effects of wage setting, has been implemented.

2.1 Theory

Company agreements and rent- sharing In the literature on two-tier wage regimes, a first group of models develops the rent-sharing approach and emphasises conditions of the market power of workers and employees. Some representative studies are offered by Moene (1988) and Holden (1988), who examine the crucial role of conflictual labour relations and assume that firms may find rent-sharing agreements more convenient at local level in order to reduce conflictual relations and to minimise the negative effects of strikes and work-to-rule practices.1 The threat of conflictual labour relations is made credible by the existence of hiring and firing costs which, in turn, inhibit the moderating influence of competitive pressures on wage setting. By paying above-tariff wages, the firm reduces recruitment and turnover problems, as insider-outsider models predict. Wage pressures are more powerful in the case of trade union activity. When local wages may be obtained as the outcome of Nash bargaining between a single firm and its employees' representatives, a common prediction is that wage solution is a function of the parties’ bargaining power.

1This literature shows that in case of disagreement, firms may suffer from a slow-down in work intensity, with a reduction in revenues. The threat of non-co-operative behaviour by workers may induce the payment of an extra wage, above the tariff wage previously determined at central or sectoral level.

4

More controversial are choices of indicators which reflect this power. For workers, two natural indicators are usually selected: union density, and coverage of collectively bargained wages. However, employee bargaining power may be exerted not only inside local collective bargaining but also outside firm-level negotiations, by merely threatening employers to demand a collective negotiation over wage issues (Corneo and Lucifora, 1997, p. 266). Additional information on workers’ bargaining influence, such as hours of strike, complements the mere information of the presence of unions at firm-level agreements and help to capture the monopolistic power of employees in wage-setting2. Wage premiums are also explained by the market power of firms, which achieve above-average profits and distribute part of their rents under the pressure of wage demands (Nickell and Wadhwani, 1990), thus showing that product and labour market imperfections interact and produce magnified effects. Lastly, after unions and employees, a third factor comes to the fore when rent-seeking governs wage-setting schemes: the managerial personnel3. Recent literature has shown that connections between executives and their subordinates are crucial in the design of incentive systems, suggesting focus on “the behaviour of individuals at two tiers of the firm hierarchy - managers and workers” (Bandiera et al., 2008, p. 2). It can be shown that, when managers are paid fixed wages, they do not offer optimal incentive mechanisms, but simply favour workers to whom they are socially connected, irrespective of those workers’ effort or ability. Other studies have shown that bargaining of generous long-term wage contracts may be an ingredient of managers’ optimal conduct, aimed at maximising their private benefits rather than shareholder value, as confirmed by empirical evidence for the US (Bertrand and Mullainathan, 2003). More generally, in an environment of collusive corporate alliances, between unions and firms, or between managers and their employees, the importance of the ‘rent extraction’ hypothesis must be detected and compared with the predictions of optimal contract theory, which is the second class of models we examine below. Company agreements and incentive effects

A second group of models in the two-tier wage-setting literature relies on efficiency wage theories. A single firm, as suggested by Rødseth (1993), may find it convenient to

2 Recently, the positive role of unions on productivity, in establishments where human resource practices are adopted, has been found by Black and Lynch (2001). 3Recent studies on Italian managers test the importance of two models: the performance model, adopted by multinationals and non-family firms, and the fidelity model, featuring family-controlled firms: in contrast with the first model, the second one does not rely on promotion and reward systems based on performances, but on the quality of managers’ relationships with the firms’ family owners, which is the cause of worse outcomes and of no generational shifts towards the better model (see Bandiera et al., 2008).

5

pay extra wage increases to its workers in order to extract more effort from them. Local arrangements between workers and the single firm may increase the ‘pie’ to be shared. Many alternative models belong to this group, but all of them focus on productivity outcomes, rather than on simple distributive effects. The major differences are due to the complementary factors introduced in any single model which enriches the basic Solow efficiency wage scheme. For instance, Muysken and van Veen (1996) and Garino and Martin (2000) formalised Nash bargaining on company payments between a local union and the firm, and found that wage drift is an increasing function of the local union’s bargaining power, the firm’s monopoly power in the product market, and the responsiveness of effort function to wage bonuses. Both studies show that efficiency wages and local wage bargaining are mutually reinforcing in determining higher wages. Turnover costs or union power interact with asymmetric information and efficiency wages. Marti (1997) shows that shirking effects on wages are greater if wages affect turnover costs. A firm which hires new workers must expend effort in searching and training costs, but making extra payments with respect to the reference wage is capable of deterring shirking, attracting and retaining workers. In short, selection, motivation and retention - the three major issues of the incentive wage literature - are embedded in this variety of studies on two-tier wage regimes. A final issue of compensation incentive devices concerns benefits and costs of collective variable pay systems with respect to individual ones, as offered by the literature on profit sharing (see Uvalic, 1991).These collective incentive schemes are more likely to be offered when total output is the outcome of the efforts of many agents, but individual contributions cannot easily be identified. In addition, when jobs do not require simple routine tasks, but co-operation between employees, group incentive schemes are more suitable. A profit-sharing contract thus allows efficiency gains to be achieved. Collective bonuses are not exempt from potential drawbacks: the efficiency effects of group payments, such as profit-sharing schemes, are menaced by a well-known free-riding problem, which arises when there are N workers, and any single employee only get 1/N of the surplus he can generate from his individual efforts (while bearing the whole amount of the disutility of his own efforts). The weak link between individual behaviour and firm performance needs some remedies. One solution is a policy to promote team culture, such as all-employee ownership plans, as empirically analysed by Blasi and Kruse (1996) and Kruse et al. (2003). Alternatively, horizontal mutual monitoring and peer pressure among workers may be attractive. Peer pressure may be generated by a group norm, i.e., a norm for effort, so that each individual deviating from the norm is penalised by the group (Kandel and Lazear, 1992). These models, which predict opposite effects on productivity gains, need to be further explored on empirical grounds.

6

Company agreements and risk-sharing

Decentralised wage-setting rules envisage a flexible compensation system which links workers’ pay to enterprise performance, to obtain a risk-sharing device and increase wage flexibility. That is, profit-sharing is not only an incentive mechanism. It is also a sharing tool which shifts the burden of adverse shocks from profit losses to workers’ pay contractions; hence, it intrinsically implies the reversible nature of wage premiums and lack of hysteresis on compensation patterns (OECD, 1995). As well known, there is a basic trade-off between risk and incentives, and a crucial constraint is that incentive provisions offered to risk-averse employees impose some compensation in terms of higher wages. In addition, the risk-sharing issues lead us to re-examine some interactions between wage and employment strategies4. Wage flexibility and the discretionality of firms in changing wages is conducive to higher employment stability; on the contrary, the limited freedom of firms in downward wage adjustment leads to employment reductions or the greater use of temporary contracts. This argument should lead us to ascertain whether these payment arrangements, which can increase money wage flexibility over the business cycle, reduce the (otherwise) extensive use of fixed-term contracts. This issue will be explored for the Italian case, in which significant changes in terms of liberalisation of temporary contracts have been adopted in the last decade. As noted above, all the mentioned models explain the wage drift phenomenon according to alternative assumptions, and the majority predicts extra-wage payments, highly responsive to product and labour market imperfections. Some other evidence on practices of decentralised level of negotiations over wage issues offers terms of comparison to better evaluate our findings.

4 Many other critical aspects have been raised by the literature on incentives, overviewed in Prendergast (1999). For instance, some benefits of incentives are conditioned by potential drawbacks arising in multi-tasks and multi-agents environments (Holmstrom and Milgrom, 1994). When workers are involved in various duties, incentives may induce effort in one task, but sub-optimal strains in some other occupations. In addition, quite often many contracts are non linear, i.e., offer discrete premiums when some performance threshold is reached, even if continuous and linear contracts should be preferable. Another critical issue is objective measurability, highlighted by Baker, Gibbons and Murphy (1994): malfunctioning of incentive contracts are due to “distortionary performance measures”, due to unavailability of output objective measures. In such circumstances, input or subjective performance measurements, sanctioned in implicit contracts, not enforced by courts by the reputation of the firm itself, produce less distorted incentives.

7

2.2 Evidence: related literature International evidence Empirical studies for a number of countries evaluate the spread of performance –related compensations and test their main effects along the theoretical lines discussed above. One strand of this research is represented by profit-sharing literature, which considers collective incentives schemes. This field is quite rich, especially for those countries where wages are only set at company level, such as the US and the UK. For Continental Europe, the French experience, examined by Cahuc and Dormont (1997), is particularly important, since in France profit-sharing (intéressement) has been explicitly encouraged by fiscal incentives, and some forms of profit-sharing schemes are even compulsory, at least for medium-large firms (50 employees or more). Many other economies (more than 20 countries) are explored in the literature on employee financial participation and a summary of major findings shows a large variety of results, partly because of differences in measures and specifications (see survey by Kruse, 1993 and, ten years later, that of Pérotin and Robinson, 2003). Without entering into detail, one major result is the highest efficacy of these practices as a productivity driver, especially in the US, where efficiency gains range from 3 to 32%, but positive effects are also found in Japan and some European economies5. Among single country experiences, it is convenient to consider those contexts, such as Italy, which belong to the same cluster of countries featuring a predominant sectoral bargaining level, but which also reveal recent trends towards decentralisation (Du Caju et al. 2008). A good case in point is that of the Netherlands. A recent study by Gielen, Kerkhofs and van Ours (2006), over the period 1995-2001 which manifested bargaining decentralisation, explores the main determinants of adoption of PRP schemes and their related consequences on productivity. The authors find that the spread of PRP is not a pervasive and homogenous phenomenon: wage flexibility is significantly affected by the size of companies and the sector to which they belong.6 The study also estimates that productivity gains (around 9%) are caused by incentives as well as by selective worker

5 In the UK, France and Japan the maximum estimated values remain below 10% (8% in the UK, 2-9% in France, and 9% in Japan). For France, Cahuc and Dormont (1997) estimated a productivity impact of around 2%, a very modest effect in a country where wage decentralisation has been explicitly fostered by state legislation (Eurofound, European Foundation for the Improvement of Living and Working Conditions, 2007, p.17). 6On this last issue, Gielen, Kerkhofs and van Ours (2006) suggest that output measurability makes it easier to design incentive schemes which link compensation to firm performance (thus explaining the higher incidence of variable schemes in construction (55%) with respect to the health care sector) (10%).

8

sorting: a confirmation that both channels may be important in boosting productivity gains, as originally found by Lazear (2000).7 Another interesting case-study is Germany where the introduction of contingent compensation, in the last few years, has made individual wages more conditional upon market mechanisms, although this trend is embedded within the institutional framework of central collective agreements. German sectoral contracts, the main bargaining levels, through the introduction of opening clauses, have made explicit allowance for local company bargaining; these second-level agreements were first applied to working time issues, but recently also to wages and salaries. German company management and works councils have thus gained new discretionality to conclude firm-level pacts which deviate from industry-level collective agreements (Schnabel, 1999).8 Note, however, that the spread of wage company arrangements remains modest, as documented by the 2007 WSI Works Council Survey. But what is even more remarkable is that the transfer of authority to single firms has determined income cuts in most cases (53%): reduction in wage drift, retrenchment of bonuses and pay below collective agreement standards. In sum, in Germany wage agreements at local level have taken the form of reduction and downward flexibility, among medium-sized firms, of special bonuses which remain valid only for large companies (e.g., Opel, Ford, Bayer), as shown by Hassel (1999, pp. 499–500). On the contrary, as mentioned above, in Italy the 1993 structural reform undertaken was intended to promote company-level arrangements and to encourage wage increases linked to profit and productivity gains obtained at firm level. Before passing to the empirical estimates for the Italian case, some information clarifies the general attitudes of the country. The first official national documentation on the financial participation of employees, after the 1993 Agreement, was carried out in 1997 by the Italian Statistical Institute (ISTAT). The inquiry, aimed at verifying the actual implementation of company-level arrangements, showed the limited diffusion of firm agreements which, in the biennial 1995-1996, involved only around 9.9% of companies with at least 10 employees and 38.8% of total employment. In addition, the research shed some light on sectoral and regional imbalances, documenting the diverging patterns between the industrial and services sectors and between North and South Italian areas: the majority of employees,

7 Note that Lazear (2000), who referred to a single US firm, showed that piece rates caused an increase in productivity of about 40%, half of which was due to sorting effects (i.e., the inflow of high productivity employees). 8 As known, in this economy, wages are mainly set by industry-level collective bargaining and, under the law (Works Constitution Act) works councils can sign plant-level agreements only when sectoral contracts make explicit allowance for local bargaining through opening clauses, recently applied not only to working time, but also to wages and salaries (Schnabel, 1999).

9

73.4%, were occupied in the industrial sector, and the majority of firms involved in new practices were located in the North9. Other weaknesses were signalled by the Bank of Italy’s investigations (the Survey on Investment in Manufacturing). In 1999, for instance, from a representative sample of about 1,000 firms with more than 50 employees, only 50% (about 44.9) of manufacturing companies adopted bonuses related to company results, whereas 18% of them maintained more traditional fixed bonuses. The 2001 Bank of Italy investigation, covering only firms with 20-49 employees, showed that PRP were adopted by 13% of firms. For 2001-2002, as well as the following period 2003-2007, Bank of Italy estimated that these schemes persisted in being limited, representing nearly 4% of compensation packages (Casadio, 2008). An additional source of information for the metalworking sector is the annual survey carried out by the national employers' association of the Federmeccanica and examined by Origo (2009). This database, which offers information for waves from 1989 to 1997, shows a strong variability along economic cycles and a strong fluctuation of PRP, ranging from 2.2% of firms involved in downturns (1992), to 26.4% in upturns (2006). Lastly, a long-term perspective is obtained from a third official source of information, represented by the CNEL archive, spanning a period of four years from 1998 to 2006. In November 2007, CNEL issued the results of its report covering almost 3,000 company-level agreements and related to two groups of firms: the first with more than 1000 employees, and the second within the range of 100-999 employees. The most important finding was the impressive trend, from 1998 to 2006, of falling implementation of company agreements. This tendency was generalised but more marked for the second group of smaller firms (100-999 workers). Even in a sector which is still a leader in terms of participatory practices, such as the metalworking industry, the incidence of company PRP among large companies fell from 51% in 2000 to 37% in 2004; smaller companies in the same sector recorded a larger decrease (from 50% to 22%). As seen above, previous studies have shown that two of the main proposals of the 1993 Agreement - ample coverage of company arrangements and the spread of PRP- have been only partially implemented, and not contemporaneously for the same firms. What is left to be done is not only an analysis of the whole economy covering small companies and all sectors, to obtain a more precise picture, but also deeper study of the factors which favoured or hindered adoption of PRP schemes.

9 The number of firms involved in company agreements was more limited in the South with only 4.3%, whereas the average values for North West and North East were, respectively, 10.7% and 11. 3% (ISTAT, 1999).

10

3. Determinants of adoption of flexible wage and the RIL-AIDA dataset

3.1 Data

Our analysis is based on a nationally representative sample of manufacturing and non-manufacturing firms, obtained by merging information on balance sheet data, provided by the AIDA database for a panel of firms sampled between 2002 and 2005, with cross-sectional information on flexible compensation and other workplace practices collected by the ISFOL Employer Survey for 2005. In particular, the AIDA panel of firms gives information on sales, added value per employee, wage bills and various types of investments, besides standard accounting variables on an annual basis. The establishments covered by the ISFOL survey for 2005 collect information about personnel organisation, recruitment strategies, position of employees, training investments, the presence of unions and the adoption of PRP schemes. The RIL survey refers to the universe of firms operating in the non-agricultural private sector, sampling both partnership and limited companies, for a total sample of 21,728 firms. In order to link information concerning the workers’ characteristics to indicators of firm performance and fixed capital, a sub-sample of the RIL dataset was merged with balance sheet from the Bureau Van Dijk AIDA archive, available for a period of nine years (2002-2005). This dataset collects balance sheets for limited companies with turnover higher than 100,0000 Euros for 2004 (the turnover threshold was previously 500,0000 Euros). Thus, the merged sample RIL-AIDA can exploit both cross-sectional information about employees’ participation and other workplace practices, for 2005, and the longitudinal structure of accounting data for the period 2002-2005. According to the characteristics of the RIL-AIDA dataset, the representativeness of the merged sample is reduced to partnership and limited companies. Also, we exclude from the AIDA-RIL merged sample firms with fewer than two employees. This filter is applied to identify firms characterised by a minimum level of organisational structure and internal labour market, and allows us to avoid all phenomena connected with self-employment, which was not of direct interest in our case. All observations for which full information is not available or which record outlier values are discarded; hence, our dataset, examined in the following sections, consist of 1957 firms10.

3.2 Basic features of performance-related payments: descriptive statistics

10 Concerning balance sheets data, we excluded observations exhibiting missing values; in addition, we operated trimming procedure to eliminate extreme values for value added and physical capital per employee (observation falling out the first and the last 0.5 percentile).

11

In this section we show the descriptive statistics of the RIL-AIDA merged sample for the year 2005. Our analysis, however, is not merely cross sectional as we investigate also some balance sheet variables that represent proxies of the firms’ the past performance over the period 2002-2005. To begin with, Table 1 reports the incidence of PRP by macro-region, firm size and sectors (re-aggregated from the 2-digit ATECO classification). The figures are calculated by applying sampling weights of the RIL survey. The results shown in Table 1 demonstrate that from our database the spread of company wage agreements is modest, on average in 9.68% (440) of firms. Note, however, that our sample includes all the industries of the market economy, with the only exclusion of the primary sector, and covers also small firms, those with less than 10 employees. It may plausibly be assumed that this inclusion changes average values downwards and makes our results not easily comparable with those obtained from other sources, restricted by size, or sectors. A second result is the high variability of PRP diffusion along firm size, sector and macro-region. Distribution by size class reveals that PRP schemes are mainly adopted by large companies (more than 99 employees), with percentages around 47%, but totally absent in small firms. In addition, what is remarkable is that medium-sized firms are characterised by a modest extent of company wage contracts, which are found in only 11% of firms with 15-99 employees. This result marks a notable difference with other country experiences, where medium-sized firms are more involved in PRP agreements (see, for instance, the German experience, where 36% of firms with 20-49 employees and 31% of those with 50-99 employees had some form of performance-related pay schemes in 2007(Eurofound, 2008, Tab.1). Other important findings concern sectoral diffusion. From our database, the highest values are recorded in mining sectors and energy supply (26%), followed by manufacturing (17%). Conversely, the low incidence of PRP agreements in services, shown by the ISFOL survey is particularly meaningful and is a critical aspect for a labour intensive sector, where promotion of organisational and motivational innovations considerably enhances total factor productivity growth. Our database shows that this promotion has been modest: PRP contracts are adopted in a small number of services firms, and the average value falls when Transport and Communication (which shows percentages around 11%) is excluded. PRP are also limited to Northern Italian regions (11% in North West and 14% in North-East), whereas 7% and 6% of Central and Southern companies, respectively, sign PRP contracts, a territorial dispersion not too distant from that obtained from ISTAT survey over the 1995-1996 (ISTAT, 1999). However, sector and regional differentials are highly dependent on the size of the company. Table 2 shows PRP diffusion by industry and macro region associated to

12

different size classes. The results obtained show the critical role of dimensional aspects: large firms exhibit high incidence of PRP in all regions, not only in Northern regions (53%), but also in the Centre (46%) and South (23%). Again, considering the sectors, we obtain highest values for the class of companies with more than 99 employees for mining and energy (73%) and manufacturing (61%). Other main characteristics, which distinguish establishments with and without company wage contracts, are listed in Table 3 and will be included in our estimates. First, we consider a group of variables related to firm performances (productivity) and strategies (capital accumulation, innovation, internationalisation). It is easy to note that firms with PRP bargaining had better performance over the period 2002-2005 than firms without them. In particular, the adoption of these contracts in 2005 was positively correlated with the average values of the (log) added value per employee and the (log) fixed capital per employee, calculated over the period 2002-2005, as well as with firm size. Firms involved in process and product innovation adopt PRP more frequently (70% and 67%, respectively) compared with others (48% and 32%). In addition, the high proportion of firms involved in export and adopt financial participation schemes was in line with international evidence found in a wide range of countries. Exporters have been shown to pay higher wages than non-exporting firms, and these premiums were estimated at around 7% for the Italian economy by Mayer and Ottaviano (2007)11. Our database provides additional findings and shows that 75% of firms engaged in international trade adopt PRP, whereas this percentage falls to 41% among non- exporters. Instead, concerning workforce characteristics, from our database we do not observe striking differences between firms with and without PRP contracts. Quite unexpectedly, workforce composition in terms of job qualifications and tasks performed is not clearly correlated with flexible pay policies, since the incidence of managerial, supervisory, white-collar and blue-collar workers is the same in both groups of firms. However, firms with trained employees adopt more frequently PRP (28%) with respect to others (17%) and make less use of fixed-term contracts (7% versus 10%). Lastly, the female component appears to be more ‘segmented’ in firms without performance-related contracts (25% versus 36%). Concerning industrial relations, we observe the higher presence of unions in PRP firms (91% instead of 13%), and higher incidence of strike activities. The penultimate rows of Table 2 shows the difference associated with labour market tightness indicator. Lower local unemployment rates, which capture the role of external

11As mentioned in the Appendix to their study, Mayer and Ottaviano for the Italian economy “cover the following periods: 1989-91, 1992-94, 1995-97, 1998-00 and 2001-03. The sample is selected with a stratified design on location, industrial activity and size for all firms with less than 500 employees and more than 11. All firms with more than 500 employees are included in each wave” (p. 66).

forces and outside options for employees, explain the higher percentages of company bonuses. Conversely, these premiums do not increase with vacancy rates, suggesting that company bonuses are not used as recruitment devices when firms open vacancies and encounter difficulties in matching unemployed workers. To summarise, some fundamental differences emerge from the ISFOL database between firms adopting PRP contracts and others. The first group consists of much larger firms, which, over the period 2002-2005, achieved higher average added value and higher capital per worker. These firms are also involved in process and product innovations of higher intensity. Concerning the workforce, the main differences are the higher proportions of men and trained employees, as beneficiaries of company wage contracts. Lastly, the intensity of their collective action through the presence of unions is much higher.

3.3. Probability of adoption of PRP contracts: logit estimates

We estimated the role of various determinants on the probability of the adoption of PRP contracts by applying a standard logistic regression model. Formally, this is equivalent to assuming that a latent variable, y*, indicating the benefits the employer receives from adopting a PRP scheme, depends on a vector of establishment and worker characteristics as follows:

(1) i=1,.. ; t=2002,...2005 iiitii WZXy εδβα +⋅+⋅+⋅= ,*

where Xi,t is a vector of variables that indicates firms’ past performance collected by AIDA dataset over the period 2002-2005, Zi is a vector of other firms’ characteristics collected by RIL survey for 2005 and Wi is a vector of workplace and workforce characteristics for 2005. The parameters βα , and δ are the vector of coefficients to be

estimated while the iε represent the i.i.d error terms draw from a standardized logistic

regression. According to this model, firms adopt a PRP (yi=1) when , whereas

firms do not to introduce PRP (yi=0) when .

0>∗iy

0≤∗iy

In particular, as proxy variables for the firms’ past performance we include in the equation (1) the average added value per employee, the average fixed capital per employee over the period 2002-2005. In order to verify whether firms with better performance, greater investments in physical capital and higher labour costs are more or less likely to introduce PRP compensation schemes12. The other firms characteristics included in the equation (1) are collected by RIL survey for 2005 and consist of the

13

12 As said above, we ran the regression analysis by selecting firms which had been present in the panel for at least 3 years over the period 2002-2005, in order to limit sample selection problems.

14

number of employees (four size classes, omitting the category of fewer than 5 employees), two dummy variables, equal to one if establishments innovate their products and/or processes, a dummy variables if some sales are directed towards foreign market and twelve 2-digit industry controls. As worker characteristics we consider the percentage of employees with fixed-terms contracts; the percentage of women, a dummy variable equal to one, if any part of the plant is unionised; the percentage distribution of employment by five occupational categories (managerial workers omitted), and the percentage of trained employees. Finally, after excluding firms exhibiting missing values in these variables and operating usual trimming procedure to eliminate extreme values form variables (excluding observations falling out the first and the last 0.5 percentile), we and up with a merged sample counting 1957 observations which has been used for estimation. In particular we estimate two different specification of equation (1). In the first specification, the productive past performance is measured with the average value added per employee. In the second, we use the sample distribution of the average value added per employee over the period 2002-2005 to define three dummy variables: the variable “low” indicates whether the past performance of the firm is minor of the 25th percentile of the average value added per employee distribution, the variable “middle” indicates whether the past performance is between the 25th and 75th percentile and the variable “high” whether the past performance is over 75th percentile. The main results of the logit maximum likelihood regression are shown in Table 3, where the estimates of the marginal effects are calculated taking into account sectoral clusters which potentially affect the robustness of standard errors. Discussion of our results may be organised as answers to two main questions: the eligibility issue, which reveals which firms and employees sign PRP agreements and motivation, which means identifying company strategies applied to PRP schemes. The first answer we obtain is that company wage component is positively conditioned by the performance indicator, as confirmed by the positive statistical significance of the coefficient associated with productivity growth. Thus it emerges that concession agreements, which include flexible clauses, added to base wage component, are mainly adopted in those firms with higher ‘ability to pay’. A more detailed analysis confirms these results and offers information about the heterogeneity of firms. We have included two dummy variables for distinct groups of companies, those with high productivity growth (values higher than the 0.75 quantile)

and those characterised by medium productivity growth (values between 050-0.75 quantile). The estimated regression coefficients can be interpreted as the partial derivative of the probability of the adoption of PRP with respect to the productivity of firms of the omitted class (firms with productivity values lower than the 0.25 quantile); in other terms, we obtain the marginal change in the probability of the adoption of PRP due to labour productivity of the best and medium performers, with respect to low

15

performers (the omitted category). The results show whether the effect of a particular regressor is positive or negative, and how large this effect is compared to the benchmark group. The findings we obtain show that firms with the highest efficiency growth are significantly more oriented to adopt incentive schemes, since the performance indicator has a greater impact. Note, however, that even if the effect of productivity on PRP is positive and the efficiency-wage relation has the expected sign, some reverse causality problems cannot be ruled out, but this will be a matter for future research. We also obtain that the probability of PRP premiums is not correlated with company size, premiums not being more frequent in large firms (i.e. firms with more than 99 employees). Note that, in the efficiency wage literature, there are two different effects concerning dimensional aspects. On one hand, asymmetric information and monitoring costs increase with firm size and explain the positive correlation with firm size. On the other hand, opportunistic behaviour and free-riding arguments are more frequent in large firms, since horizontal mutual monitoring and peer pressure are more difficult, thus generating an opposite, negative correlation. On empirical grounds, international evidence is not conclusive, since the role of firm size is contradictory13. The probability of PRP is correlated with company bargaining. The model we have estimated introduces as proxies for workers’ bargaining power the union presence and labour disputes, which are expected to enhance the probability of concessionary bargaining. From our results we obtain significant effects for both variables which confirm that in unionised establishments PRP may be added to minimum wages, set by sectoral agreements. They also suggest that wage drift emerges as outcome of conflictual labour relations, which force employers to give extra payments over sectoral settlements. In addition, labour market tightness, measured by local unemployment rates, which capture the role of reservation wages and outside options, induces wage demands. PRP agreements are more likely to be adopted in firms where there are more managerial positions, confirming the importance of agency problems. The higher the proportion of managers within firms (the omitted category in our regression), the higher the propensity to introduce incentive contracts is. Here again the Italian findings seem at odds with other countries’ experiences. For instance, the German evidence shows that employees’ financial participation is more probable for blue-collar jobs (Heywood and Jirjahn 2002), whereas higher involvement by white-collar employees has been found in the UK (Robinson and Wilson, 2001) and Finland (Arranz-Aperte and Heshmati, 2003). These disparities do not appear in Italian enterprises, where neither blue-collars, nor white-collars have eligible conditions in terms of PRP contracts.

13For some countries, such as the US, Germany and Canada, some studies find that employees’ financial participation is positively related to size, others do not find any statistically significant relationship. Further, for the UK, both positive and negative relationships between firm size have been found, whereas Japan is the only country where profit-sharing schemes are definitely higher in smaller firms (Pérotin and Robinson, 2003).

16

In addition, the main beneficiaries of PRP agreements are men and the percentage of women has a negative effect on the probability of adoption of PRP contracts. This finding is far from being an obvious and unavoidable result. A case in point is France, where distributive contents of firm-level agreements, set independently of union influence, benefit women particularly (Fakhfakh and FitzRoy, 2004). Analogous benefits are obtained by the female component in a country such as Germany, where collective bargaining, as in Italy, plays a great role: in the economy, the wage premium obtained by works councils, around 11%, is also higher for women than for men (Addison et al. 2006). In Italy, conversely, the negative correlation between the percentage of female workers and the spread of integrative bargaining schemes has some points in common with the traditional literature of economic discrimination against women in the labour market (Addabbo et al. 2001; Naticchioni and Ricci, 2009). The second answer relates to strategies applied to the adoption of PRP. As seen in Section 2, the theoretical and empirical literature has traditionally indicated three main reasons behind the adoption of PRP. These are taken as the basis for identifying key factors which account for the adoption of performance-related contracts in the Italian economy. The first reason, offered by agency theories, emphasises that, in conditions of imperfect competition, company agreements allow redistribution of rents and may favour cooperative behaviour between management and employees. This explanation may count in the case of Italy, since unionisation, strikes, the incidence of managers and of the male component, which are also proxies for employees’ bargaining power, enhance the probability of adoption of company bonuses. The second reason is that PRP schemes are risk-sharing instruments, which would arise from the outset of risk situations for firms (erratic sales, volatility of profits, high level of indebtedness). This reason has been important in the past Italian experience, especially from the second half of the 1980s: firms offering financial participation, often more labour intensive, invested more than others and implemented innovative plans which needed cooperative relationships to deal with those organizational changes (Biagioli and Curatolo, 1999). This is not case of the last few years. From our estimates, the limited influence of product and process innovation and the slightly higher capital accumulation, made by firms adopting PRP agreements in the foregoing years, reveal that risky situations are not conducive to wage flexibility. This finding seems to mark an impasse in those trends towards implementation of strategies devoted to enhancing organisational flexibility, which emerged in Italy from similar studies for the 1980s. The third important reason offered by the employee financial participation literature is that compensations linked to firm performance reveal a commitment device to select, retain and motivate workers (Kruse, 1993).

17

The selection objective, as mentioned above, does not find clear support from our estimates, since vacancy rates do not enter with a positive sign as determinants of adoption of PRP agreements. Retention is another strategic tool of incentive plans. In the case of Italy, where temporary contracts are not only used as a buffer stock, but also to fill in for permanent workers on leave, firms offer PRP to encourage employees to immediately acquire specific human capital, as assumed by Booth, Francesconi and Frank (2002) and confirmed from our estimates. We also find the accumulation of human capital, as proxied by the proportion of trained employees, has a positive impact, although it is only slightly significant, confirming that hold-up theories about human capital investments may be at work in our sample. The use of PRP contracts may act as a commitment device for workers whose specific skills are enhanced by training investments sponsored by firms (Acemoglu, 1997). Qualified workers are expected to obtain enterprise bonuses and these premiums are probably more important when firms intend to reduce the turnover costs of training new employees with firm-specific skills. In any case, the low statistical significance of this variable (trained employees) is in line with the modest probability of adoption of wage incentives for innovation activities in productive processes and in production quality, as already noted. All these findings are in contrast with the vast empirical literature, in which retention and innovation strategies are found among the main drivers of incentive plans (Damiani, 2006). The third aspect of productivity enhancement is motivation. In Italy, since the mid-1980s, firms with PRP schemes offered efficiency wages to improve production quality and productivity (Del Boca and Cupaiuolo, 1998), to implement restructuring plans and remove their previous efficiency and profitability gap (Biagioli and Curatolo, 1999; Amisano and Del Boca (2004). For the last few years, however, our results show some reversal trends and are coherent with findings obtained from other studies, which prove rent sharing, modest sorting effects (Origo, 2009, for the metalworking industry) and low elasticity of wages to efficiency gains (see Cristini and Leoni, 2007, for the Northern Italy). From our database weakness of incentive devices is confirmed from the anomalous sectoral diffusion14. Sectors which show highest propensities to sign PRP agreements are mining and energy supply and manufacturing. Conversely, belonging to service sector does not positively affect the probability of adoption on new wage practices. It is remarkable result since service sector is the key industry were productivity increases, obtained through good human resource practices, should be more necessary, especially in a country, as Italy, where productivity growth lags behind

14 Table 4 does not show sectoral dummies for reasons of space, but these results are available upon request.

18

that of the US and which is, over the period 1995-2004, the only country-sectoral case of negative productivity in the EU-13 (Inklaar et al., 2008)15. To conclude, we find that for the whole economy company compensations have failed to keep pace with (decreasing) productivity gains, but wage moderation and favourable periods for profits has not determined the basis for investments efforts (Tronti, 2007), nor for promotion of company PRP arrangements, as confirmed by our estimates. In short, the thesis of “weakness” and “incomplete” implementation of the Italian bargaining rules originated by the 1993 reform” (Tronti, 2007) is fully confirmed from our database.

4. Conclusions

Our research, based on the ISFOL database, reveals two main results. The first is the limited spread of PRP agreements, which are still not a common practice, since they involve only a minor proportion (around 10%) of Italian firms. The second concerns the existence of ample differentials by region, sector and firm size. These disparities, already shown by the survey carried out by the national statistics office for the mid-1990s (ISTAT, 1999), are still persistent and reveal heterogeneity in company strategies and in their utilisation of wage flexibility. A crucial aspect is the moderate diffusion of PRP contracts in the most labour-intensive sector, i.e., services. The tertiary sector, where productivity differentials recorded between countries are even more pronounced (van Ark et al. 2008), is the typical sector where motivational and organisational devices should be implemented by under-performing countries to fill the gap with the best achievers. Even in countries where the incidence of company wage agreements is quite limited, such as Germany, greater diffusion of PRP has been recorded in the private tertiary sector. This is not the case in Italy, where these negotiations are more familiar practices in manufacturing, but not common in services companies. The overall portrait which describes the typical profile of firms adopting variable compensation systems is ambiguous: they are more successful in terms of added value levels and growth and more commonly present in international markets. Their higher propensity for international trade (with respect to companies only working within national borders) confirms that heterogeneous firms, under pressure of international competition, adopt different wage policies and these findings complement information on the wage premiums of Italian exporters. However, the typical PRP firm, in 2005, is not significantly more active in terms of outlays in investments and not greatly involved in offering wage bonuses to solve recruiting problems and fill vacancies. Lastly, its attitude

15Poor results in the services sector contribute towards explaining Italy’s disappointing productivity growth: over the period 1995-2004, private services performance (-0.1%) was the only country-sectoral case of negative productivity growth in EU13.

19

towards signing company wage agreements is not exempt from offering grants to the more protected (male and unionised) workforce component. This ambiguous profile is a stimulus to ascertain the actual motivations behind the adoption of company wage agreements. The reasons for implementing PRP contracts, as tested with our estimates, may be classed in three different groups: risk-sharing, productivity improvement, and rent division. Past experience, especially from the second half of the 1980s onwards, appears in a number of studies, as overviewed by Biagioli and Del Boca (2004). It was clear that the driving forces behind PRP agreements were strategic choices aimed at promoting organisational changes and obtaining employees’ acceptance of such changes. In the last few years, the Italian scenario has been characterised by some reversal trends. One is the issue of dimensional size. In 2005, after two decades of implementation of firm compensation systems, which, as already noted, had been in operation since the mid-1980s, small and even medium-sized companies do not appear to have completely caught up with large companies in adopting PRP schemes. However, firm large size is a factor which does not increase the probability of adoption of PRP contracts. In this scenario, it is also remarkable that high vacancy rates do not induce the payment of wage incentives, and matching and recruitment efforts, far from being a stimulus to offering wage bonuses, end up by absorbing financial resources and distracting firms from extra-payments. A second weakness of incentive motivation is related to investment and innovation strategies. Undertaking investment projects and implementation of product and process innovation do not significantly promote employee financial participation: in 2005 Italian firms did not show the same attitude of previous decades, when they implemented restructuring processes and signed wage agreements to elicit worker cooperation and commitment. In the last few years, Italian companies are still probably involved in aimed at less conflictual relations, although the climate of industrial relations features declining weights of membership and unionised employees. Firms’ efforts are now oriented to pursuing a different strategy - basically to obtain employment flexibility and relocate a portion of its workforce from permanent to temporary jobs. These considerations may explain why Italian companies, as shown by our estimates, are likely to sign wage agreements when the fraction of fixed-terms contracts is higher, i.e., when a confrontational climate should threatens these occupational changes. In addition, the positive influence of the managerial and male personnel component are factors which increase the probability of offering company wage bonuses, confirming that the spread of PRP contracts is related to market power and rent division. Firm-specific agreements thus offer opportunities for local rent sharing, which are more probably adopted in unionised firms, under the threat of conflictual labour relations and for firms offering higher labour compensations. Thus, our study compared with other recent countries’ experience, with similar wage setting institutions, shows that the Italian

20

case is not totally at variance with comparable economies. For instance, similar trends, as those found in Belgium and Germany by Rusinek and Rycx (2008) and Gürtzgen (2005), demonstrate that firm-specific agreements generate opportunities for local rent-sharing, thus corroborating an implicit scepticism on the functioning of firm specific agreements. However, in Italy, the limited implementation of PRP reveals a scenario of asymmetric bargaining of firm rents, with a declining weight of workers’ power and of limited promotion of company PRP arrangements, as confirmed by our estimates. The related consequences on the disappointing Italian efficiency growth open the key question which needs to be further explored: what is left to be done is to ascertain the role of specific factors that have limited exploitation of potential advantages of company level arrangements in terms of productivity gains. But this is a matter for further research. Data appendix The RIL (Rilevazione Imprese e Lavoro) survey refers to the universe of firms operating in the non-agricultural private sector, sampling both partnerships and limited companies. The RIL sample, stratified by firm size, industry, geographical area and legal form, totalled 21,728 firms in 2005. The useful insights of the RIL survey with respect to other dataset is that it provides complete information about firms’ characteristics, labour force composition and workplace practices. In particular this dataset provides information about employee composition, type of contract, gender, and share of trained workers. Useful insights on industrial relations and workplace practices are provided with variables reporting firms’ unionisation, existence and modes of firm-level bargaining, etc. In order to link information on employees composition and workplace characteristics to indicators of firm performance, assets and labour costs, a sub-sample of the RIL dataset was merged with balance sheet data from the Bureau Van Dijk AIDA archive, available for a four year period (2002-2005). This dataset collects balance sheets for limited companies with turnover over than 100000 euros from 2004 on. This turnover threshold was 500,000 euros for the years 2002 and 2003. After the merge, the merge was reduced to 1957 firms, composed by a panel section including balance sheet data for the period 2002-2005 and a cross-section referring to 2005, containing information on employee composition and workplace practices. According to the typical features of the RIL and AIDA datasets, the representativeness of this merged sub-sample was restricted to limited companies. The RIL-AIDA merged sample also shows a shift in the distribution of firm size with respect to the original RIL dataset, due to the turnover criterion used to include firms in the AIDA archive. In particular, using the AIDA-RIL dataset rather than full RIL dataset, the share of firms

21

with fewer than 10 employees decreased from 46.1% to 27.9% of the total, while the share of firms with more than 50 employees increased from 11.6% to 29.4%.

References

Acemoglu D. (1997), “Training and Innovation in an Imperfect Labour Market”, Review of Economic Studies, vol. 64, n. 3, pp. 445-464.

Acocella N., and Leoni R. (eds.) (2007), Social Pacts, Employment and Growth, Physica- Verlag, Heidelberg.

Addabbo T., Addis E., Biagioli M., Caniglia P., Waldmann R. and Spanò A. (2001), I differenziali salariali per sesso in Italia, Comitato Nazionale di Parità e Pari Opportunità (Ministero del lavoro), Roma, Poligrafico dello Stato.

Addison J. T., Schnabel C. and Wagner J. (2004), “The Course of Research into the Economic Consequences of German Works Councils”, British Journal of Industrial Relations, vol. 42, n. 2, pp. 255–281.

Addison J. T., Schnabel C., Schank T. and Wagner J. (2006), “ Works Councils in the Production Process, Journal of Applied Social Science Studies, vol. 126, n.2, pp.251-283.

Amisano, G., and Del Boca A. (2004), “Profit related pay in Italy: a microeconometric analysis”, International Journal of Manpower, vol. 25, n.5, pp. 463–478.

Arranz-Aperte and Heshmati A. (2003), Determinants of Profit Sharing, IZA Discussion Paper, 776.

Baker G.P., Gibbons R. and Murphy K. (1994), “Subjective Performance Measures in Optimal Incentive Contracts”, Quarterly Journal of Economics, vol. 109, n. 4, pp. 1125–56.

Bandiera O., L. Guiso, A. Prat and Sadun R. (2008), Italian Managers: Fidelity or Performance?, London School of Economics, mimeo.

Bank of Italy (2008), Rapporto sulle tendenze nel sistema produttivo italiano, Rome.

Bertrand M. and Mullainathan S. (2003), “Enjoying a Quite life”, Journal of Political Economy, vol. 111, n. 5, pp. 1043-1075.

22

Biagioli M. and Curatolo S. (1999), “Microeconomic determinants and effects of financial participation agreements: An empirical analysis of the large Italian firms of the engineering sector in the eighties and early nineties”, Economic Analysis, vol. 2, n. 2, pp 99 - 130.

Biagioli, M., and Del Boca A. (2004), "The institutional setting of the Italian experience of employee financial participation", in Pendleton, A., Perotin, V. (Eds), Profit Sharing in Europe, Elgar, London.

Black S.E. and Lynch L. M. (2001), “How To Compete: The Impact Of Workplace Practices And Information Technology On Productivity”, Review of Economics and Statistics, vol.83, n. 3, pp. 434–445

Blasi J. and Kruse D.L. (1996), “Employee stock ownership and corporate performance among public companies”, Industrial & Labor Relations Review, vol. 50, n. 1, pp. 60-80.

Booth A. L. , Francesconi M. and Frank J. (2002), “Temporary jobs: Stepping Stones or Dead Ends”, The Economic Journal, vol. 112, n. 6, pp. F189- F213.

Cahuc P. and Dormont B. (1997),”Profit Sharing: Does it Increase Productivity and Employment? A Theoretical Model and Empirical Evidence on French Micro Data”, Labour Economics, vol. 4, n. 3, pp. 293-319.

Casadio P. (2008),“Contrattazione aziendale integrativa e differenziali salariali territoriali: informazioni dall’Indagine della Banca d’Italia”, lavoro preparatorio per la Commissione Interistituzionale di Indagine sul Lavoro istituita nel 2007 dai Presidenti del Senato della Repubblica, della Camera dei Deputati e del CNEL.

CNEL (2007), “Lineamenti della contrattazione aziendale nel periodo 1998-2006”, Commissione dell’Informazione, Rome, November, 2007.

Corneo G. and Lucifora C. (1997), “Wage Formation under union Threat effects: Theory and empirical evidence”, Labour Economics, vol. 4, n. 3, pp. 265-292.

Cristini A. and Leoni R. (2007), “The ’93 July Agreement in Italy: Bargaining Power, Efficiency Wages or Both?,”, in Acocella N., and Leoni R. (eds.) Social Pacts, Employment and Growth, Physica- Verlag, Heidelberg, pp. 97-119.

Damiani M. (2000), “Profit sharing in multiple bargaining systems”, Economic Analysis, vol.3, n. 2, pp. 113 - 135.

Damiani M. (2006), Impresa e Corporate Governance, Roma, Carocci.

23

Del Boca, A. and Cupaiuolo E. (1998), “Why Do Firms Introduce Financial Participation? Evidence from A Sample Of Small And Medium Italian Manufacturing Companies”, Economic Analysis, vol. 1, n. 3, pp. 221-237.

Du Caju et al. (2008), Institutional Features of Wage Bargaining in23 European Countries, the US and Japan, IZA DiscussionPaper, n. 3867.

Eurofound, European Foundation for the Improvement of Living and Working Conditions, EIROnline, European industrial relations observatory on-line, various issues.

Fakhfakh F. and FitzRoy F.R. (2004), “Basic wages and firm characteristics: Rent-sharing in French Manufacturing”, Labour: Review of Labour Economics and Industrial Relations, vol. 18, n. 4, pp. 615-631.

Freeman, R.B. and Lazear E.P. (1995), “An Economic Analysis of Works Councils”, in Works Councils: Consultation, Representation, and Cooperation in Industrial Relation, ogers J. and Streeck W. (eds.), pp. 27–50. Chicago, University of Chicago Press.

Garino, G. and Martin C. (2000), “Efficiency Wages and Firm-Union Bargaining”, Economics Letters, vol. 69, n. 2, pp. 181-185.

Gielen A. C, Kerkhofs M.J.M. and van Ours J.C. “(2006), Performance Related Pay and Labor Productivity, IZA Discussion Paper, 2211.

Griliches Z. and Mairesse J. (1998), "Production Functions: The Search for Identification", in Econometrics and Economic Theory in the Twentieth Century, Cambridge, Cambridge University Press, Chapter 6.

Gürtzgen N. (2005), Rent-Sharing And Collective Bargaining Coverage– Evidence From Linked Employer-Employee Data, Centre for European Economic Research Discussion Paper n. 05-90

Hassel, A. (1999), “The erosion of the German system of industrial relations”, British Journal of Industrial Relations, vol. 37, n. 3, 483-505. Heywood, J.S. and Jirjahn U. (2002),“Payment systems, gender and industrial relations in Germany”, Industrial and Labor Relations Review, vol. 56, pp 44 – 64.

Holden S. (1988), “Wage Drift and the Relevance of Centralised Wage Setting”, Scandinavian Journal of Economics, vol. 100, pp. 711-731

Holmstrom B. and Milgrom P. (1991), “Multitask Principal-Agent Analyses: IncentiContracts, Asset Ownership, and Job Design”, Journal of Law, Economics, a

24

Organization, vol. 7, n. 1, pp. 24-52

Inklaar R., Timmer M.P. and van Ark B. (2008), “Market services productivity across Europe and the US”, Economic Policy, vol. 53, pp. 139-194

ISTAT (1999), I Principali Risultati della Rilevazione sulla Flessibilità nel Mercato del Lavoro, Roma.

Kandel E. and Lazear E.P. (1992), “Peer Pressure and Partnership”, Journal of Political Economy, 100, pp. 801-817.

Kruse D.L. (1993), Profit-Sharing, Does it Make a Difference?, Upjohn Institute for Employment Research, Kalamazoo, MI.

Kruse D.L et al. (2003), Motivating Employee-Owners In ESOP Firms: Human Resource Policies and Company Performance, NBER Working Paper, n. 10177,

Lazear E.P: (2000), “Performance Pay and Productivity”, American Economic Review, 90, n. 5, pp. 1346-1361

Marti C. (1997), “Efficiency Wages. Combining the shirking and turnover costs models”, Economic Letters, vol. 57, n. 3, pp. 327-330.

Mayer T. and Ottaviano G. (2007),“The Happy Few: The internationalisation of European firms New facts based on firm-level evidence”, mimeo.

Moene K.O. (1988), “Union's Threats and Wage Determination “, The Economic Journal, vol. 98, n.391, pp. 471-483

Muysken, J.and van Veen, T. (1996), “Efficiency Wages and Local Wage Bargaining”, Scandinavian Journal of Economics, vol. 98, n.1, pp. 119-127.

Naticchioni P. and Ricci A. (2009) " Differenziali salariali di genere: un'analisi dinamica delle caratteristiche e della loro remunerazione nel mercato", Rivista di politica sociale, forthcoming.

Nickell J. and Wadhwani S. (1990), “Insider Forces and Wage Determination”, The Economic Journal, vol. 100, n. 401, pp. 496-509.

OECD (1995), Employment Outlook, Paris.

Origo F. (2009), “Flexible pay, firm performance and the role of unions. New evidence from Italy”, Labour Economics, vol.16, n.1, pp. 64-78

25

Pérotin V., Robinson A. (2003), Employee Participation in Profit and Ownership: a Review of the Issues and Evidence, Working paper, Social Affairs Series, SOCI 109 EN, Luxembourg Parliament, Directorate- General for Research.

Prendergast C. (1999), “The Provision of Incentives in Firms”, Journal of Economic Literature, vol. 37, n. 3, pp. 7-63.

Robinson A. and Wilson N. (2001), “Employee participation, ownership and productivity: An empirical re-appraisal,” Leeds University Business School.

Rødseth, A., (1993), “Efficiency wages and Local versus Central Bargaining”, Oxford Economic Papers, 1993, vol. 45, n. 3, pp. 470-481.

Rusinek and Rycx (2008) Rent-Sharing Under Different Bargaining Regimes: Evidence from Linked Employer-Employee Data, IZA Working Paper N. 3406

Schnabel C. (1999), “Collective Bargaining under Stress: Decentralisation and Opening Clauses in Germany”, Institut der deutschen Wirtschaft Köln and Ruhr University Bochum.

Tronti L. (2007), The July Protocol and Economic Growth: The Chance Missed, in Acocella and Leoni (eds.), pp.217-267.

Uvalic M. (2001), The Pepper Report: Promotion of Employee Participation in Profits and Enterprise Results in the Member States of the European Community, Supplement to Social Europe, 3.

van Ark B., O’Mahony M., and Timmer M.P., (2008) The productivity gap between Europe and the United States: trends and causes, Journal of Economic Perspectives,vol. 22, n. 1, pp.25-44.

26

Table 1 Descriptive statistics: diffusion of PRP by macro-region, firm size and sector

mean Std. D. macro-region North-West 0,11 0,31 North-East 0,14 0,35 Centre 0,07 0,26 South 0,06 0,23 firm size < 5 employees 0,00 0,02 6-14 employees 0,02 0,15 15 -99 employees 0,11 0,31 >=100 employees 0,47 0,50 sector mining, quarrying; electricity gas and water supply

0,26 0,44

manufacturing 0,17 0,38 construction 0,02 0,13 trade, hotels and restaurants 0,03 0,18 transport and communication 0,11 0,32 financial intermediation 0,03 0,19 other business services 0,05 0,22 education, health and other public services

0,10 0,31

Total 0,10 0,30

27

Table 2 Descriptive statistics by sector , firm-size and macro-region size sector 2 – 5 6 – 14 15-99 >=100

mining, quarrying; electricity gas and water supply 0,08 0,18 0,27 0,73

manufacturing 0,00 0,02 0,17 0,61 construction 0,00 0,01 0,02 0,10 trade, hotels and restaurants 0,00 0,03 0,04 0,30 transport and communication 0,00 0,00 0,19 0,39 financial intermediation 0,00 0,10 0,02 0,00 other business services 0,00 0,02 0,01 0,28

education, health and other public services 0,00 0,06 0,12 0,19

macro-region North-West 0,00 0,00 0,13 0,53 North-East 0,00 0,07 0,17 0,53 Centre 0,00 0,05 0,06 0,46 South 0,00 0,02 0,08 0,23

28

Table 3: Firms with and without PRP

no PRP PRP whole sample Variable Mean Std. D. Mean Std. D. Mean Std. D. (a) (b) (a) (b) (a) (b) av.(log) value added 6,14 0,46 6,24 0,41 6,15 0,46 av. (log) fixed capital 5,04 1,36 5,67 1,24 5,10 1,36 % female 0,36 0,25 0,25 0,20 0,35 0,25 % managers and supervisors 0,07 0,12 0,07 0,08 0,07 0,11 % white collars 0,46 0,31 0,46 0,26 0,46 0,31 % blue collars 0,47 0,31 0,47 0,27 0,47 0,31 % trained 0,17 0,28 0,28 0,33 0,18 0,29 % fixed term 0,10 0,16 0,07 0,11 0,10 0,15 foreign 0,41 0,49 0,75 0,43 0,45 0,50 union 0,14 0,34 0,91 0,28 0,21 0,41 strike 0,06 0,24 0,61 0,49 0,12 0,32 prod. innov. 0,32 0,47 0,67 0,47 0,35 0,48 proc. innov. 0,48 0,50 0,70 0,46 0,51 0,50 local unempl. rate ( 2004) 6,63 4,25 5,34 2,76 6,50 4,15 vacancy rate 0,09 0,22 0,01 0,03 0,08 0,21 Obs. 440 1517 1957

29

Table 4: Probability of adoption PRP – 2005 dy/dx Std. Err. dy/dx Std. Err. avegare (log) value added per employee, 2002-2005

0,04 * 0,02

middle performance 0,03 ** 0,02 high performance 0,05 ** 0,03 avegare (log) fixed capital per per employee, 2002-2005

0,00 0,01 0,00 0,01

% females -0,10 * 0,03 -0,10 * 0,03 % white collars -0,08 0,08 -0,08 0,08 % blue collars -0,09 0,08 -0,09 0,08 % trained workers 0,08 * 0,02 0,08 * 0,02 % fixed term workers 0,10 *** 0,06 0,10 *** 0,06 foreign 0,03 ** 0,01 0,03 * 0,01 product innovation 0,00 0,01 0,00 0,01 process innovation 0,00 0,01 0,00 0,01 union 0,278 * 0,03 0,28 * 0,03 strike 0,066 * 0,02 0,07 * 0,02 unemployment rate (2004) -0,01 ** 0,00 -0,01 ** 0,00 vacancy rate -0,22 *** 0,14 -0,21 0,14 North-West 0,02 0,03 0,02 0,03 North-East 0,08 0,06 0,08 0,06 Centre 0,00 0,03 0,00 0,03 5-14 employees 0,046 0,09 0,04 0,09 15-99 employees 0,058 0,07 0,05 0,07 >=99 employees 0,161 0,12 0,16 0,12 Industry controls (Yes) (Yes) N. Obs. 1957

* Omitted variables: managers and supervisors, South, firms less than 5; * significant at 1%, ** significant at 5%, *** significant at 10%

QUADERNI DEL DIPARTIMENTO DI ECONOMIA, FINANZA E STATISTICA

Università degli Studi di Perugia

1 Gennaio 2005

Giuseppe CALZONI Valentina BACCHETTINI

Il concetto di competitività tra approccio classico e teorie evolutive. Caratteristiche e aspetti della sua determinazione

2 Marzo 2005 Fabrizio LUCIANI Marilena MIRONIUC

Ambiental policies in Romania. Tendencies and perspectives

3 Aprile 2005 Mirella DAMIANI Costi di agenzia e diritti di proprietà: una premessa al problema del governo societario

4 Aprile 2005 Mirella DAMIANI Proprietà, accesso e controllo: nuovi sviluppi nella teoria dell’impresa ed implicazioni di corporate governance

5 Aprile 2005 Marcello SIGNORELLI Employment and policies in Europe: a regional perspective

6 Maggio 2005 Cristiano PERUGINI Paolo POLINORI Marcello SIGNORELLI

An empirical analysis of employment and growth dynamics in the italian and polish regions

7 Maggio 2005 Cristiano PERUGINI Marcello SIGNORELLI

Employment differences, convergences and similarities in italian provinces

8 Maggio 2005 Marcello SIGNORELLI Growth and employment: comparative performance, convergences and co-movements

9 Maggio 2005 Flavio ANGELINI Stefano HERZEL

Implied volatilities of caps: a gaussian approach

10 Giugno 2005 Slawomir BUKOWSKI EMU – Fiscal challenges: conclusions for the new EU members

11 Giugno 2005 Luca PIERONI Matteo RICCIARELLI

Modelling dynamic storage function in commodity markets: theory and evidence

12 Giugno 2005 Luca PIERONI Fabrizio POMPEI

Innovations and labour market institutions: an empirical analysis of the Italian case in the middle 90’s

13 Giugno 2005 David ARISTEI Luca PIERONI

Estimating the role of government expenditure in long-run consumption

14 Giugno 2005 Luca PIERONI Fabrizio POMPEI

Investimenti diretti esteri e innovazione in Umbria

15 Giugno 2005 Carlo Andrea BOLLINO Paolo POLINORI

Il valore aggiunto su scala comunale: la Regione Umbria 2001-2003

16 Giugno 2005 Carlo Andrea BOLLINO Paolo POLINORI

Gli incentivi agli investimenti: un’analisi dell’efficienza industriale su scala geografica regionale e sub regionale

I

17 Giugno 2005 Antonella FINIZIA Riccardo MAGNANI Federico PERALI Paolo POLINORI Cristina SALVIONI

Construction and simulation of the general economic equilibrium model Meg-Ismea for the italian economy

18 Agosto 2005 Elżbieta KOMOSA Problems of financing small and medium-sized enterprises. Selected methods of financing innovative ventures

19 Settembre 2005 Barbara MROCZKOWSKA Regional policy of supporting small and medium-sized businesses

20 Ottobre 2005 Luca SCRUCCA Clustering multivariate spatial data based on local measures of spatial autocorrelation

21 Febbraio 2006 Marco BOCCACCIO Crisi del welfare e nuove proposte: il caso dell’unconditional basic income

22 Settembre 2006 Mirko ABBRITTI Andrea BOITANI Mirella DAMIANI

Unemployment, inflation and monetary policy in a dynamic New Keynesian model with hiring costs

23 Settembre 2006 Luca SCRUCCA Subset selection in dimension reduction methods

24 Ottobre 2006 Sławomir I. BUKOWSKI The Maastricht convergence criteria and economic growth in the EMU

25 Ottobre 2006 Jan L. BEDNARCZYK The concept of neutral inflation and its application to the EU economic growth analyses

26 Dicembre 2006 Fabrizio LUCIANI Sinossi dell’approccio teorico alle problematiche ambientali in campo agricolo e naturalistico; il progetto di ricerca nazionale F.I.S.R. – M.I.C.E.N.A.

27 Dicembre 2006 Elvira LUSSANA Mediterraneo: una storia incompleta

28 Marzo 2007 Luca PIERONI Fabrizio POMPEI

Evaluating innovation and labour market relationships: the case of Italy

29 Marzo 2007 David ARISTEI Luca PIERONI

A double-hurdle approach to modelling tobacco consumption in Italy

30 Aprile 2007 David ARISTEI Federico PERALI Luca PIERONI

Cohort, age and time effects in alcohol consumption by Italian households: a double-hurdle approach

31 Luglio 2007 Roberto BASILE Productivity polarization across regions in Europe

32 Luglio 2007 Roberto BASILE Davide CASTELLANI Antonello ZANFEI

Location choices of multinational firms in Europe: the role of EU cohesion policy

33 Agosto 2007 Flavio ANGELINI Stefano HERZEL

Measuring the error of dynamic hedging: a Laplace transform approach

II

34 Agosto 2007 Stefano HERZEL Cătălin STĂRICĂ Thomas NORD

The IGARCH effect: consequences on volatility forecasting and option trading

35 Agosto 2007 Flavio ANGELINI Stefano HERZEL

Explicit formulas for the minimal variance hedging strategy in a martingale case

36 Agosto 2007 Giovanni BIGAZZI The role of agriculture in the development of the people’s Republic of China

37 Settembre 2007 Enrico MARELLI Marcello SIGNORELLI

Institutional change, regional features and aggregate performance in eight EU’s transition countries

38 Ottobre 2007 Paolo NATICCHIONI Andrea RICCI Emiliano RUSTICHELLI

Wage structure, inequality and skill-biased change: is Italy an outlier?

39 Novembre 2007 The International Study Group on Exports and Productivity

Exports and productivity. Comparable evidence for 14 countries

40 Dicembre 2007 Gaetano MARTINO Paolo POLINORI

Contracting food safety strategies in hybrid governance structures

41 Dicembre 2007 Floro Ernesto CAROLEO Francesco PASTORE

The youth experience gap: explaining differences across EU countries

42 Gennaio 2008 Melisso BOSCHI Luca PIERONI

Aluminium market and the macroeconomy

43 Febbraio 2008 Flavio ANGELINI Marco NICOLOSI

Hedging error in Lévy models with a fast Fourier Transform approach

44 Febbraio 2008 Luca PIERONI Giorgio d’AGOSTINO Marco LORUSSO

Can we declare military Keynesianism dead?

45 Febbraio 2008 Pierluigi GRASSELLI Cristina MONTESI Paola IANNONE

Mediterranean models of Welfare towards families and women

46 Marzo 2008 Mirella DAMIANI Fabrizio POMPEI

Mergers, acquisitions and technological regimes: the European experience over the period 2002-2005

47 Marzo 2008 Bruno BRACALENTE Cristiano PERUGINI

The Components of Regional Disparities in Europe

48 Marzo 2008 Cristiano PERUGINI Fabrizio POMPEI Marcello SIGNORELLI

FDI, R&D and Human Capital in Central and Eastern European Countries

49 Marzo 2008 Cristiano PERUGINI Employment and Unemployment in the Italian Provinces

50 Marzo 2008 Sławomir I. BUKOWSKI On the road to the euro zone. Currency rate stabilization: experiences of the selected EU countries

51 Aprile 2008 Bruno BRACALENTE Cristiano PERUGINI Fabrizio POMPEI

Homogeneous, Urban Heterogeneous, or both? External Economies and Regional Manufacturing Productivity in Europe

III

52 Aprile 2008 Gaetano MARTINO Cristiano PERUGINI

Income inequality within European regions: determinants and effects on growth

53 Aprile 2008 Jan L. BEDNARCZYK Controversy over the interest rate theory and policy. Classical approach to interest rate and its continuations

54 Aprile 2008 Bruno BRACALENTE Cristiano PERUGINI

Factor decomposition of cross-country income inequality with interaction effects

55 Aprile 2008 Cristiano PERUGINI Employment Intensity of Growth in Italy. A Note Using Regional Data

56 Aprile 2008 Cristiano PERUGINI Fabrizio POMPEI

Technological Change, Labour Demand and Income Distribution in European Union Countries

57 Aprile 2008 Simona BIGERNA Paolo POLINORI

L’analisi delle determinanti della domanda di trasporto pubblico nella città di Perugia

58 Maggio 2008 Simona BIGERNA Paolo POLINORI

The willingness to pay for Renewable Energy Sources (RES): the case of Italy with different survey approaches and under different EU “climate vision”. First results

59 Giugno 2008 Simona BIGERNA Paolo POLINORI

Ambiente operativo ed efficienza nel settore del Trasporto Pubblico Locale in Italia

60 Ottobre 2008 Pierluigi GRASSELLI Cristina MONTESI Roberto VIRDI

L’interpretazione dello spirito del dono

61 Novembre 2008 Antonio BOGGIA Fabrizio LUCIANI Gianluca MASSEI Luisa PAOLOTTI

L’impatto ambientale ed economico del cambiamento climatico sull’agricoltura

62 Novembre 2008 Elena STANGHELLINI Francesco Claudio STINGO Rosa CAPOBIANCO

On the estimation of a binary response model in a selected population

63 Dicembre 2008 Gianna FIGÀ-TALAMANCA Limit results for discretely observed stochastic volatility models with leverage effect

64 Maggio 2009 Mirella DAMIANI Andrea RICCI

Factors behind performance-related pay: evidence from Italy

IV

V

ISSN 1722-618X

I QUADERNI DEL DIPARTIMENTO DI ECONOMIA Università degli Studi di Perugia

1 Dicembre 2002

Luca PIERONI:

Further evidence of dynamic demand systems in three european countries

2 Dicembre 2002 Luca PIERONI Paolo POLINORI:

Il valore economico del paesaggio: un'indagine microeconomica

3 Dicembre 2002 Luca PIERONI Paolo POLINORI:

A note on internal rate of return

4 Marzo 2004 Sara BIAGINI: A new class of strategies and application to utility maximization for unbounded processes

5 Aprile 2004 Cristiano PERUGINI: La dipendenza dell'agricoltura italiana dal sostegno pubblico: un'analisi a livello regionale

6 Maggio 2004 Mirella DAMIANI: Nuova macroeconomia keynesiana e quasi razionalità

7 Maggio 2004 Mauro VISAGGIO: Dimensione e persistenza degli aggiustamenti fiscali in presenza di debito pubblico elevato

8 Maggio 2004 Mauro VISAGGIO: Does the growth stability pact provide an adequate and consistent fiscal rule?

9 Giugno 2004 Elisabetta CROCI ANGELINI Francesco FARINA:

Redistribution and labour market institutions in OECD countries

10 Giugno 2004 Marco BOCCACCIO: Tra regolamentazione settoriale e antitrust: il caso delle telecomunicazioni

11 Giugno 2004 Cristiano PERUGINI Marcello SIGNORELLI:

Labour market performance in central european countries

12 Luglio 2004 Cristiano PERUGINI Marcello SIGNORELLI:

Labour market structure in the italian provinces: a cluster analysis

13 Luglio 2004 Cristiano PERUGINI Marcello SIGNORELLI:

I flussi in entrata nei mercati del lavoro umbri: un’analisi di cluster

14 Ottobre 2004 Cristiano PERUGINI: Una valutazione a livello microeconomico del sostegno pubblico di breve periodo all’agricoltura. Il caso dell’Umbria attraverso i dati RICA-INEA

15 Novembre 2004 Gaetano MARTINO Cristiano PERUGINI

Economic inequality and rural systems: empirical evidence and interpretative attempts

16 Dicembre 2004 Federico PERALI Paolo POLINORI Cristina SALVIONI Nicola TOMMASI Marcella VERONESI

Bilancio ambientale delle imprese agricole italiane: stima dell’inquinamento effettivo