Chapter 7 Employment protection This chapter heavily borrows from courses and slides by Tito Boeri, Professor of Economics at Bocconi University, Milan, Italy
Protecting jobs Losing a job is always a bad experience Workers lose specific human capital (future expected wages are lower) They feel bad/depressed Thus jobs must be secured, true? Who is fired? Industry Construction Tertiary Together End fixed-term contract Voluntary job quit Firing for economic reasons 41.5 26.4 56 53.1 19.4 39.8 20.2 20.6 5.9 1.8 1.1 1.9 Other motives 8.7 13.1 5.1 5.9 Exits from employment by motives in France, 2002 (in % of total exits)
Who is unemployed? 3,5 3 2,5 2 1,5 1 0,5 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Canada USA Spain France Germany Italy Japan UK Male ratio of unemployment rate by age group to unemployment rate in 8 OECD countries, 1997 So what? Employment protection (EP) is legitimate as long as it can reduce career risks BUT: those who benefit from EP are not highly exposed to unemployment the main victims of unemployment, the youth, are not covered by EP
Questions Does EP affect employment level? Does EP affect employment composition? What are the determinants of EP? Outline of the chapter 1. Measuring EP 2. An important theoretical point 3. Impact on the labour market 4. Impact of the labour market
1. Measuring Employment Protection Purpose For theworker: reduce income fluctuations protecting against un-insurable labour market risk stimulate longer-lasting employment relationships, skill formation For thefirm: buildup of loyalty, trust and co-operation less resistance to new technologies (workers do not feel threatened)
Socially: Purpose (cont.) more incentives to train and invest in human capital, higher productivity reduce monopsonistic power of firms What are they? Regulations concerning: individual dismissals definitions of juste cause for dismissals regulations on collective redundancies statutory severance pay minimum notice period procedural obligations to be fulfilled before the dismissals
Moreover also hiring restrictions Insofar as they increase the costs of replacements, e.g., Restrictions to the use of temporary work (e.g. Agency Work) Restrictions to the use of fixed-term contracts Rules for disadvantaged groups, etc. International Comparisons International comparisons are difficult because EP has several dimensions Countries are ranked for each component of employment protection Then, economists compute a weighted average of rankings Weights are based on a priori grounds concerning the relative importance of each component of EP
Strictness of EP by the end 1990s 1 US 8 Denmark 15 Norway 2 UK 9 Finland 16 Germany 3 New-Zealand 10 Holland 17 France 4 Canada 11 Japan 18 Spain 5 Irland 12 Austria 19 Italy 6 Australia 13 Belgium 20 Greece 7 Switzerland 14 Sweden 21 Portugal OCDE Shortcomings of these Measures The weight attributed to each component is arbitrary Interactions among the different components
2. An important theoretical point Effects of EPL: Theory Overall EPL as tax on the firm - it can be modelled as such. Thus EPL=T. Less EPL=lower costs for the firm 2 components: Transfers (TR) from the employer to the worker and Deadweight Costs (C) going to bureaucrats, lawyers, Govt... T=TR+C
Why is it important to disentangle TR from C? Because part of TR can be negotiated, and hence incorporated (discounted) ex-ante in wage contracts This is not the case of deadweight costs, C, associated with employment protection, such as legal and procedural costs, jurisprudence, etc. Neutrality of TR (Lazear, 1990) If: competitive labour market (no unions) flexible wages (no wage floors) risk-neutral agents (u(w)=w, individuals interested only in average wages over the period)) Then TR has no effects on employment and wages
Example of 2 period contracts Suppose jobs last 2 periods and have marginal productivities MP1 and MP2 respectively Without EPL, in competition W2 MP2 W1 MP1 1 r 1 r Introducing TR at 2 (i.e., W2= W2 +TR), this can be offset by lowering entry wage by an amount K such that the following condition is satisfied W 1 K W2 1 TR r W 1 W2 1 r or K TR 1 r Graphically W + TR W W - TR/ (1+r) 1 2 period
Interpretation The worker buys a bond in period 1. The price is K, which is transfered to the firm In period 2, the worker receives interest payments, which are transfered from the firm. The worker receives K (1+r) Intuition Mandated transfer from the employer to the worker can be undone by a voluntary transfer of the same size from the worker to the employer This works only if the employer succeeds in extracting a payment from the worker when the contract begins (the worker must be willing to pay the fee on signing the contract)
What about C? C is an administrative cost. At such, it reduces the economic surplus created by employment relationships. - An increase in C is equivalent to a fall in value added Following the reduction in value added, the wage may go down. But workers won t accept a fall which exactly balances the cost C. It follows firms profitability decreases with administrative costs Consequences EP has powerful impacts on employment relationships given administrative costs cannot be fully imputed to workers the labour market can be far from perfect competition: wage floors, mobility costs, unions
3. Impact on the labour market EP reduces the mobility of the workforce between employment and unemployment states Decrease in the inflow rate in unemployment direct impact of EP: fewer firings but increased risk of banqueroute Decrease in outflow rate from unemployment EP increases the expected labour cost firms become more selective fewer jobs are created
Impact on total employment is ambiguous Reminder: Employment is a stock which results from flows in and flows out firings=outflows hirings=inflows EP lowers both flows, so the impact on stock cannot be determined a priori The Effects of Employment Protection on the Labour Market: Empirical Results STOCKS FLOWS Author(s) Employment Unemployment Employment Unemployment Emerson (1988)?? - - Lazear (1990) - + Bertola (1990)?? Grubb & Wells (1993) - Garibaldi,Konings,Pissarides(1994)??? - Addison & Grosso (1996)?? Jackman,Layard,Nickell(1996)?? - - Gregg & Manning (1997)?? - Boeri (1998)?? + - Di Tella & McCulloch (1998) - + OECD (1998)??? - Kugler & StPaul (2000) + -
Summarising cross-country evidence EPL increases unemployment duration and its incidence among marginal groups of the workforce lower job creation means longer unemployment durations, which primarily affects the most exposed to unemployment EP redistributes jobs from youngsters and females to prime-aged males In Europe, young workers pay their father s job stability
Impact on productivity is ambiguous At the micro level improves the «social climate» within firms, which stimulates specific investments (+) favourizes shirking (-) At the macro level slows down the diffusion of innovations by lowering job reallocation from obsolete to modern sectors (-)
A study by Ichino and Riphahn (2004) Thesis EP favours absenteeism, which lowers productivity The authors compare absenteeism rates in three cases 1) in private vs public German sectors 4,1% vs 4,9% 2) in small vs large Italian firms 1,6% vs 2,8% 3) before vs after the probation period (tenure) in an Italian bank 0,7% vs 1,5% An estimation on aggregate panel data Correlation between growth and EP is not sufficient. One need to take into account the whole set of variables that are likely to affect growth Layard et Nickell (1999) regress labour productivity growth rate on the different labour market institutions for 20 OECD countries g 1taxes 2EP 3alloccho error They find the estimate of parameter positive 2 is significantly EP seems to increase labour productivity growth
4. Impact of the labour market EP is endogenous Hypothesis 1 Judges are biased by the state of the labour market Even if the law does not change, applied law is endogenous
A study by Ichino et al (2001) Data a huge dataset covering one major Italian bank; several establishments throughout the country; conflicts with employees and trials regional unemployment rate (North = poor, South = rich) Question Does unemployment affect judges decisions? Ichino et al (2001) At first, unemployment has a surprising effect: judgements are more favourable to employers in the North than in the South But more conflicts go to trial in the North than in the South Given the fault, the issue is more favourable to the employees in the South than in the North there is a selection bias: firms are relunctant to go to trial in the South; they only do so when the worker has done a major fault judges are biased by labour market conditions
EP is endogenous Hypothesis 2 The political support for EP is stronger when the labour market is depressed Employees benefit from EP. The need for EP rises when jobs are threatened, and when it is difficult to find another job once unemployed Empirical findings Re-read the graphs relating EP to unemployment/employment flows Does strict EP induce lower job creation, or lower job creation induce strict EP? It raises the issue of endogeneity
A study by Boeri et al (2003) Data 2002 sample of 1000 Italian males aged 16 to 80 individual characteristics Question Would you prefer «a labour market in which it is relatively easy to find a job, but it is likewise easy to lose a job», or a labour market endowed with the opposite properties (long unemployment and job durations)? Method Boeri et al (2003) Estimate of the probability to prefer the rigid market as a function of individual characteristics Results Being unemployed reduces the latter probability, while being employed tends to increase it the individual state matters Leaving in a region where unemployment is high raises the latter probability the state of the labour market matters
EP is endogenous Hence, the strictness of EP responds to the state of the labour market Consequences cycles may be amplified risks of lock-in in long-run situations where workers experience long job durations and long unemployment durations Other determinants of EP Demographic determinants The demand for EP increases with workers age (Boeri et al, 2003) Economic determinants The demand for EP decreases with workers education (id)
How to reform? Many observers believe EP should be softened in Europe Citizens must be convinced first! But, employment gains are at best rather small, which means the majority will lose while a minority will gain EP is a kind of zero sum game How to reform? There are two ways to convince the likely losers Design compensatory transfer schemes from the winners to the losers Redistribute income from the youth to the old? Two-tiers system, in which coexist different legislations for jobs created at different dates Older jobs are secured, while new jobs are not