Reforms, labour market functioning and productivity dynamics: a sectoral analysis for Italy

Transcription

1 Reforms, labour market functioning and productivity dynamics: a sectoral analysis for Italy Cecilia Jona-Lasinio (Istat and Luiss) Giovanna Vallanti (Luiss)

2 Motivations In the last two decades Italy registered notable improvements in the functioning of labour market. Such improvements in labour market performance seem to have been accompanied with a deterioration in terms of productivity and competitiveness Are the reforms in labour market responsible for both a better functioning labour market on the one side and a poor performance in terms of productivity and competitiveness on the other?

3 Theoretical considerations Labour market institutions and productivity: Employment protection o More stringent EPL is an obstacle to the reallocation of workers and jobs across firms and sectors, reduce the propensity to undertake risky activities, reduce workers effort; o On the other hand EPL may promote specific investment and result in more learning by doing with a positive effect on productivity o Partial EPL reform via the introduction of temporary contracts has an ambiguous effect on productivity depending on the reasons why firms employ temporary workers rather than permanent workers

4 Theoretical considerations Labour market institutions and productivity: Wage Bargaining setting o More centralized/coordinated wage-bargaining systems reduce the responsiveness of industry and firm-level wages to sectoral price and productivity developments (Holmlund and Zetterberg, 1991; Teulings and Hartog, 1998) with a negative effect on allocative efficiency o In decentralized wage-bargaining regimes, incentives to innovate and adopt new technologies depend crucially on workers' bargaining power. The risk of hold-up can be partly mitigated when bargaining occurs at the national level (or at the industry level but with economy-wide co-ordination) and sets the general frame of the wage schedule. (Teulings and Hartog, 1998).

5 Empirical Evidence Number of papers which try to assess the effect of labour market institutions on productivity mainly based on crosscountries regressions on aggregate outcomes (identification problem). Few exemptions: Scarpetta and Tressel (WB, 2004), Autor et al (EJ, 2007), Bassanini et al (EP, 2009), Cingano et al (EP, 2010).

6 Objective Estimate the effects of the labour market reforms which have occurred in Italy over the last thirty years on two main components of aggregate productivity: An intra-sectoral component : measuring the productivity growth within industries A rellocative component : measuring the contribution to aggregate productivity of the shift in employment between industries

7 Institutional settings and reforms Reduction in the wage indexation and introduction of part-time and training contracts (Contratti di formazione e lavoro) New procedures for collective firing and limits to the use of Cassa Integrazione e Guadagni, abolishment of indexation (1992), Giugni agreement (1993) introduces two-tiers wage bargaining (centralized and decentrilezed at firm level) The training contracts (Contratti di formazione e lavoro) are extended to a wider range of situations and the collaboration contracts are introduced (no firing and hiring costs) 1998 Pacchetto Treu: new atypical contracts such as job-sharing and staff leasing, use of temporary worker agency 2000 Reform of TWA extends the use of TWA and removes restrictions concerning unskilled workers 2001 Fixed term contracts are extended to regular employees 2003 Legge Biagi provides a common framework to atypical contracts and extends further the use of TWA

8 Institutional settings and reforms We identify three periods of major reforms EPL regular contracts EPL temporary contracts share of temporary contracts centralization Index coordination index o EPL index: measure the restrictions on the use of temporary employment by firms, with respect to the type of work for which these contracts are allowed and their duration (scale: 1-6); OECD (2008) o Centralization index: The dominant level(s) at which wage bargaining takes place (scale: 1-3); OECD (2004) o Coordination index: coordination of wage bargaining (scale1-5). 4=mixed industry and economy-wide bargaining (key unions and employers associations set pattern for the entire economy); 2=mixed industry- and firm level bargaining, with weak enforceability of industry agreements; ICTWSS database, Visser (2007)

9 Institutional settings and reforms Labour market effects of the reforms: More numerical flexibility with a steadily increase in the number of temporary forms of employment Wage moderation (decentralized wage bargained was very limited, however a change in the industrial relations environment led to period of institutional wage moderation)

10 Measurement and empirical strategy Decomposition of aggregate labour productivity growth (Foster, Haltiwanger, and Krizan, 2001) LP ω 1 1( 1 1) ( ) t jt LPjt ωjt LPjt LPt ωjt LPjt LPt = + + LPt 1 j LPt 1 j LPt 1 j LPt 1 within component reallocative component interaction effect (residual) where: ω jt is the employment share of sector j at time t LP t is the economy-wide productivity at time t LP jt is the productivity of sector j at time t

11 Measurement and empirical strategy Sectoral labor shares Rif 1 (yearly average rates of growth) Agr Industry Construction Trade, transports and hothel Financial interm real estate other services

12 Measurement and empirical strategy Sectoral labor shares Rif 2 & Rif 3 (yearly average rates of growth) Agr Industry Construction Trade, transports and hothel Financial interm real estate other services

13 Decomposition labour productivity growth within reall int total within reall int total within reall int total within reall int total Agricolture and fishing Industry Construction Trade, hotels and restaurants, transport and comunication Financial intermediation and real estate, renting and business activities Real estate, renting and business activities Other services Total

14 Measurement and empirical strategy Step 1: assess the effect of reforms on the within component (sector labour productivity) Rajan and Zingales (1998) methodology: the amount of regulation changes over time and is given for all sectors but its impact could be different according to some sectoral characteristics which influences the frequency of labour adjustments Y = Zδ + ( Z sec ) γ + X λ + µ + ε (1) jt t t j jt j jt where: Y jt = productivity of sector j Z t = labour market institution indexes (EPL, CE and CO indices) sec j = sector characteristics (manufacturing, intrinsic JR, human capital endowment) X it = sector-level control variables

15 Measurement and empirical strategy Sectoral intrinsic need for JR: job turnover and excess job turnover calculated at sectoral level using data from an economy with a deregulated labour market (UK) γ gives the change in the difference in productivity between high and low reallocation industries (10th and 90th percentile for example) originated by a change in the institutional variable. In our sample sectors at the 10th and 90th percentile of intrinsic reallocation are Transport, storage and communication and Manufacture of electrical and optical equipment respectively

16 Measurement and empirical strategy Step 2: assess the effect of reforms on the reallocation effect Following Brown and Earle (2004, 2006) methodology, we can express the reallocative effect as a covariance, namely: ωjt ( LPjt 1 LPt 1) = ncov( ωjt, LPjt 1 LPt 1) The effect may also be computed as β from the following OLS regression: LP LP = + + ω α β jt 1 t 1 jt jt nvar( LPjt 1 LPt 1) υ

17 Measurement and empirical strategy LP LP = + + ω α β jt 1 t 1 jt jt nvar( LPjt 1 LPt 1) υ β is the responsiveness of the sector size to its relative performance The responsiveness is measured in terms of its contribution to the aggregate productivity growth We express β as a function of institutional indices and other sector specific variables in order to capture the impact of changes in labour market institutions on the extent of productivity- enhancing reallocation as follow ω = βd + δ( D sec ) + ( Z D sec ) γ + µ + µ + ε (2) jt jt jt j t jt j t j jt LPjt 1 LPt 1 where: Djt = nvar( LP LP ) jt 1 t 1

18 Results (1)

19 Results (2)

20 Results (2) The marginal effect of reducing the EPL index by 1 unit ranges from -2% for the sector at 10th percentile of the distribution of flexibility need (Transport, storage and communication) to -5% for the sector at the 90th percentile (Manufacture of electrical and optical equipment)

21 Results (3)

22 Results (4)

23 Results (5)

24 Results (6)

25 Results (7)

26 Conclusions Changes in EPL on fixed term workers and the increase in the share of temporary jobs have had a negative impact on both the level of productivity and the growth rate; Such negative effects are less important in sectors which employed a larger share of high skilled workers and are larger in high reallocation sectors; The magnitude of the effects is not negligible, and lies around 5.6% of the difference in labour productivity of high relative to low reallocative sectors Reforms seem to have negatively affected the reallocative capacity of the economy, by reducing the reallocative contribution to aggregate growth of high reallocative sectors.