The Employment Effects of Privatizing Public Utilities in OECD Countries

Abstract

This article examines whether the privatization of network-based utilities in developed countries leads to a retrenchment of the workforce. The panel regressions reveal, first, privatization does indeed lead to a reduction in the number of employees in the sectors concerned. Second, it is not typically the new investors themselves who implement the reduction. The downsizing takes place while the state is still the unique shareholder. Third, even though left-wing parties also implement privatization or at least not hamper, the results show that the displacement of workers is lower when left-wing parties dominate the cabinet.

Key words:

• Privatization
• employment
• public utilities

Notes

1. These are Australia, Austria, Belgium, Canada, Denmark, France, Finland, Germany, Greece, Japan, Italy, Ireland, The Netherlands, Norway, New Zealand, Portugal, Spain, Sweden, Switzerland and the United Kingdom.

2. The influence of union power on the effect of privatization cannot be directly tested since no sector-specific data on union density are available.

3. The countries are Australia, Austria, Belgium, Canada, Denmark, France, Finland, Germany, Greece, Japan, Italy, Ireland, The Netherlands, Norway, New Zealand, Portugal, Spain, Sweden, Switzerland, the United Kingdom and the USA.

4. The pre-period of all companies with the exception of British Telecom (BT) falls into the period of observation. I estimated the models without BT. The results do not change.

5. Selecting a seven-year period (three years before and after privatization plus the event year) is the standard approach in the economic literature (Megginson et al., 1994). Robustness tests using longer and shorter time spans in the panel regressions do not provide different results.

6. As mentioned, only overall data are available, and not sector-specific international comparative data.

7. That means including dummies for formal privatization at time t, at t + 1, t + 2 until the material privatization takes place.

8. Even though the government gives autonomy to the contracted management, the government or the responsible ministry often defines the management contract and supervises the finances of the company. Representatives of the government are furthermore in all organs of the company.

9. It might be argued that the results are driven by technological differences between the sectors or technological progress within one sector. I therefore applied several robustness checks. First, I added sector dummies controlling for systematic differences in the employment dynamics between both sectors. Second, I estimated the models including an interaction effect (sector dummy*privatization) checking whether the effect of privatization is different between the sectors. Third, I added time dummies and alternatively a linear trend variable to control whether technological progress that affects all companies in one sector drives the results. The results remain stable across all robustness checks.