Unemployment Week 8 1 Textbook reading Borjas Chapter

Unemployment Week 8 1

Textbook reading • Borjas, Chapter 12. • Ehrenberg & Smith, Chapter 14. • Cahuc & Zylberberg, Chapters 3, 8 (9). 2

Additional readings • “Two-Tier Labour Markets in the Great Recession: France Versus Spain. ” Samuel Bentolila, Pierre Cahuc, Juan J. Dolado, Thomas Le Barbanchon (Economic Journal 2012) 3

Motivation • Labor market equilibrium determines the level and distribution of wages and employment. – Across regions, sectors, demographic characteristics, etc. • In the basic competitive model of the LM, there is no unemployment in equilibrium. • In practice, however, unemployment is quantitatively important as well as persistent! • We would like to understand the determinants of its level as well as trends. 4

Unemployment in the basic competitive model • The only wage for unemployment to emerge in the basic competitive model is: • If wages are above their equilibrium level, • And for some reason they cannot adjust down. • If no restrictions to wage adjustment, it should be temporary as wages fall to the new equilibrium level. 5

Outline 1. Facts 2. Types of unemployment 3. Frictional unemployment (search). 4. Structural unemployment 5. Cyclical unemployment. 6. International comparisons. 6

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Unemployment rate, Spain 1980 -2012 25 23 21 19 17 15 13 11 9 7 5 7

Unemployment rate by province, Spain 2012 8

25 International unemployment rates, 2003 -2012 20 15 10 5 0 2003 2004 Alemania 2005 España 2006 2007 Estados Unidos 2008 2009 Francia Grecia 2010 Italia 2011 Reino Unido 2012 9

Unemployment rate by characteristics (Spain, 1 st q. 2013) 10

2. Types of unemployment • Frictional • Seasonal • Structural • Cyclical 11

The steady-state unemployment rate • The UR observed in the long run as a result of flows of workers in and out of the market. • The steady state UR depends on the transition probabilities. • Lower UR if: – Jobs are more stable – Unemployment spells are shorter. 12

3. Frictional unemployment (search) • There would be U even if D=S, because of “search costs”. – Frictional unemployment. • Diff. firms make diff. wage offers to the same worker. – Incentives to “shop around”. – It takes time! – Unemployment spell. 13

The wage offer distribution • Assume only the unemployed search. • The wage offer distribution is known. • Optimal search length? – Costs and benefits of searching longer. 14

The reservation (or “asking”) wage • An optimal “stopping rule”. • Sequential search with a threshold. • The link between the reservation wage and the unemployment spell. • The reservation wage is determined by comparing the costs and benefits from additional search. 15

Determinants of the reservation wage • “Impatience” (individual discount rate). • Non-labor income. – Including unemployment benefits. 16

The basic job search model • Continuous time. • The job-seeker knows the distribution of possible wages, H(. ). • Successive wage offers are independent draws from this distribution. • Utility increases with income. – There is no leisure. 17

Setup • Over a short interval dt, instantaneous utility is w·dt. • Workers lose their jobs at a rate q·dt. • Real instantaneous interest rate is r. • Discounted value of $1 at t, available at t+dt is 1/(1+r·dt). – The “discount factor” over interval dt. 18

Expected utility for an employed worker • Discounted expected utility of an employed person receiving wage w: Ve = • Rearranging, r. Ve = w + q(Vu – Ve) • Also, Ve(w) – Vu = 19

The optimal search strategy • Assume the worker only meets a single employer on any date. • The employer offers wage w. – Constant over the duration of the job. • The optimal search strategy is: – If no offer on date t, keep searching. – If wage offer w, accept if Ve(w)>Vu, otherwise keep searching. 20

Implications • Note that Vu does not depend on w. • While Ve(w) is increasing in w (linearly). • Thus, the strategy is equivalent to the “stopping rule”: accept wage w iff it is higher than r. Vu. – The “reservation wage”. 21

Expected utility for a job-seeker • Arrival rate of job offers: l – A function of the state of the labor market, worker characteristics, search effort. – For now, constant and exogenous. • Search costs: c>0 • Earnings if unemployed: b>0. • Net instantaneous income from looking for work: b-c 22

Expected utility if receiving an offer: • Vl = Expected utility if no offer: Vu Expected utility in stationary state: 23

The reservation wage as a function of the parameters of the model • It can be shown that: – There is only one optimal value, and – It maximizes the intertemporal utility of a job-seeker. • The “hazard rate” (exit rate from unemployment). • The average duration of unemployment. – An increasing function of the reservation wage. 24

Comparative statics • The reservation wage and the duration of unemployment with respect to: b, c, l, r, q. 25

Extensions of the basic model • Non-participation. • Eligibility for unemployment insurance. • On-the-job search. • Endogenous search intensity (effort). • Non-stationary unemployment benefits. 26

The Equilibrium Search Model • The basic job search model focuses on the jobseeker and takes the distribution of (potential) wages as given. • Now we will consider how wages are determined, by thinking about the behavior of firms. • H(. ) can be “endogenized”. 27

The Diamond critique • If employers can set wages, and workers and firms are identical, in equilibrium the distribution of wages would be concentrated at w. R. • This changes if workers are heterogeneous. • Or if workers can search “on the job”. – Firms may offer high wages in order to achieve low quit rates and attract many workers. 28

Implications of the equilibrium search model • The wage of a worker rises (only) when she switches jobs. • Wages are positively correlated with size of the firm. • Wages rise as workers gain experience. • An unemployed job-seeker accepts all offers. • The wage offer distribution is entirely determined by parameters (q, y, b-c, lu, le). • But: the fit with empirical wage distributions is not great. 29

4. Structural unemployment • How can structural unemployment arise in a competitive market? • Demand shifts may affect some sectors differentially. • The skills of workers laid off in the declining sectors may not be easily transferred to the growing sectors. – Sector-specific skills. – It may take time to acquire them. – Long-term unemployment. 30

The “sectoral shifts” hypothesis • Long-term unemployment due to a structural imbalance between: – The skills of unemployed workers, and – The skills that employers are looking for. • There is disagreement about how much sectoral shifts currently contribute to unemployment. • Empirical analysis: – Dispersion in economic growth across industries. – Some evidence that sectoral shifts are important. 31

The recent debate in the US • http: //krugman. blogs. nytimes. com/2012/06/0 8/the-structural-obsession/ • http: //krugman. blogs. nytimes. com/2012/09/0 2/mishmash-not/ • http: //www. thijsvanrens. com/SU/AMU_intro duction_201205. pdf 32

5. Cyclical unemployment • Imbalance between the number of workers looking for jobs and the number of jobs available. – Maybe because the economy has moved into a recession. – Shrinking consumer demand. • Excess supply of workers and “sticky” wages. – The market does not clear. 33

Policy prescriptions • Stimulate aggregate demand. – To reestablish market equilibrium at the sticky wage. 34

Are wages really rigid downward? • Nominal wages may be rigid, but real wages can fall if prices are rising. – If nominal wages increase less than inflation. – This is quite common during recessions. – But it may not be enough to offset the fall in demand. 35

In the Spanish news today (May 31, 2013) • “El Banco de España plantea suprimir el salario mínimo para algunos trabajadores” (“The Bank of Spain suggests eliminating the minimum wage for some workers”) 36

“Los salarios también caen en el sector privado” (El País) 37

Questions • In a recession, • Why do firms prefer to reduce employment than wages? • Why are workers not more willing to take wage cuts to save their jobs? 38

Sources of rigid wages and involuntary unemployment • Unions • Minimum wages • Efficiency wages • Implicit contracts • (…) 39

i) Efficiency wages • Firms find it expensive to monitor the worker’s output. • They may want to pay “above-market” wages to “buy” the worker’s cooperation. • This generates (involuntary) unemployment. • The efficiency wage is the profit-maximizing wage. – No pressures on the firm to lower the wage. 40

The “no-shirking” supply curve • Monitoring is expensive. – The firm may want to offer a (high) wage that encourages workers not to shirk. • How much should the firm pay? • The cost of shirking. – You can get caught and get fired. – If unemployment is high, this is bad. • The (upward-sloping) no-shirking supply curve – Gives the number of non-shirking workers the firm can hire at each wage. – It never touches the (inelastic) supply curve. – The difference is the number of unemployed. • “Some unemployment is needed to keep the workers in line” (!) 41

Efficiency wages and unemployment • No market pressures forcing the efficiency wage down toward the competitive wage. • Workers do not shirk in this labor market. • There is involuntary unemployment. – “Unproductive” for the worker (not for the firm!). • Implication: wages will be sticky over the business cycle. 42

The impact of an economic contraction • Output demand falls. – The labor demand curve shifts. • The competitive wage drops. • The same shock reduces the efficiency wage. – But the efficiency wage is less responsive. • The unemployment rate rises. 43

The evidence for efficiency wages • Within-country, wages tend to be high in regions where unemployment is low. – The “wage curve”. – Downward sloping. • The standard demand-supply framework would predict the opposite! – Efficiency wages are a possible explanation. – Firms in regions with high unemployment need not offer high wages. 44

ii) Implicit contracts • In long-term contracts, firms and workers can bargain over wages and layoff probabilities. – The bargain can specify wage and hours for different aggregate economic conditions. – Perhaps unwritten and unspoken (“implicit”). • Two extreme types of contracts: – “Fixed-employment” – “Fixed-wage” 45

The business cycle • In the fixed-employment contract, the firm adjusts by varying the wage. – High wage during expansions, wage cuts during recessions. – The worker’s income fluctuates a lot over the cycle. • In the fixed-wage contract, the firm adjusts hours. – Fewer hours during recessions. – Lower income, but higher leisure and potential unemployment compensation. – Workers’ “real income” would fluctuate less. 46

Do workers prefer fixed-wage contracts? • Maybe. – Some evidence. – So their incomes are relatively stable, even if their hours are not. • Why? – Perhaps workers are risk-averse. – Diminishing marginal utility of income. – They prefer to avoid low incomes during a recession. – In exchange for not-so-high income during booms. • Fixed-wage contracts offer “insurance” against wage declines during recessions. – This would generate wages that are sticky over the business cycle and increases in unemployment during a recession. 47

Voluntary unemployment? • Workers are better off with fixed-wage contracts. – They “accept” layoffs in return for a more stable consumption path. • Some evidence suggesting that implicit contracts may play a role. 48

6. International comparisons • Sources of different unemployment levels and trends across countries? ? • Some factors to take into account: – The magnitude of the recession (demand shifts). – Unemployment insurance system. – Employment protection regulations. – Payroll taxes. – Wage rigidity (unionization, etc). 49

25 International unemployment rates, 2003 -2012 20 15 10 5 0 2003 2004 Alemania 2005 España 2006 2007 Estados Unidos 2008 2009 Francia Grecia 2010 Italia 2011 Reino Unido 2012 50

The case of Spain • Size of construction boom (and bust). • Temporary contracts and regulations. • Austerity and lack of growth. 51