24
deteriorate. Unemployed workers will then presumably be more willing to accept “bad” jobs,
including temporary jobs. This adjustment, however, is best seen as a response to more
fundamental forces that caused the decline in labour demand in the first place.
5.2 Changes in Labour Demand
If legislative changes or supply-side shifts are ruled out as main explanations, what remains is
the possibility that employers have become increasingly prone to hire workers on fixed-term
contracts. We consider four types of demand-side explanations: (i) shifts in labour demand
towards industries that are intensive in the use of fixed-term contracts; (ii) a general decline in
the profitability of hiring labour; (iii) a rise in incentives to use fixed-term contracts to screen job
applicants; and (iv) increased volatility in technology or product demand conditions.
Consider first the sectoral shift hypothesis. We have explored the effects of sectoral shifts by
a simple shift-share analysis pertaining to 48 industries covering the whole economy, using data
for 1990 and 2000 from the labour force surveys. This exercise suggests that sectoral shifts
explain virtually nothing of the rise in the overall share of temporary work. If the sectoral
employment figures of 2000 were applied to the sector-specific temporary work shares of 1990,
the implied total share of temporary work would be 10.4 percent, to be compared to actual shares
of 10.2 percent in 1990 and 15.2 percent in 2000. Thus, sectoral shifts cannot be an explanation.
The second hypothesis concerns how firms choose between temporary and permanent
contracts.14 Consider a firm that takes wages as given and can choose between two types of jobs
(or contracts), labeled temporary and permanent. Temporary jobs carry no firing costs but entail
high separation rates. Permanent jobs, on the other hand, involve high firing costs but also low
separation rates. Temporary jobs thus force the firm to engage frequently in search for new
workers, a cost that has to be weighed against the advantage of low firing costs. Permanent jobs
14 This discussion is inspired by a paper by Wasmer (1999), where the standard matching model of equilibrium
unemployment is extended to incorporate both temporary and permanent jobs.