Should state-owned firms change CEOs before privatization? Some evidence from the telecommunications industry

Abstract

Should state-owned enterprises change chief executive officer (CEO) before privatizing? We test competing views on this question by complementing a recently released database with newly collected data. We are able to cover 77 telecommunications privatizations, which account for nearly 80% of the sector in terms of value. We find that CEO replacement will improve performance in the telecommunications industry before privatization as measured by penetration, operating efficiency and profitability. Chief executive officer change before privatization does appear to have real consequences in firm performance before privatization.

Acknowledgements

We are grateful to Florencio Lopez-de-Silanes and Julio Rotemberg for comments and suggestions. The usual disclaimer applies. The findings and interpretations of the authors do not necessarily represent the views of the Inter-American Development Bank or its executive directors.

Notes

¹ They prepared a questionnaire addressed to the CEO in three areas. The first covered pre-privatization firm characteristics, and asked about sales, assets, profits, liabilities and management changes. The second area covered pre-privatization characteristics, with some emphasis on labour issues, such as the presence of unions and number of strikes. The third focused on the privatization process, such as privatization prices, the transaction methods used and shares sold.

² Since CEO replacement often fails to coincide with annual reporting years the replacement year is excluded from the analysis. When including the sample increases by three observations but results do not change. On the other hand, according to our sample, CEO change before privatization varies widely from region to region. While it has been common in Africa, Latin America and developed countries, it has been much less used in Asia and Transition Economies.

³ Subscription-based Bloomberg, Economatica and Worldscope deliver international financial information in a standardized format that facilitates comparisons between companies, countries, regions and industries.

⁴ As the privatizations occur at different points in time, we are also able to control for time trends in order to identify the effect of CEO change more convincingly. In fact, results hold when introducing time dummies and comparable data of telecommunication firms that were not privatized.

⁵ In some cases we have multiple share offerings for a given firm (see sebsequent text). Besides controlling for the size of block sold, we also include a dummy for whether the sale corresponds to an initial offering or not. There is no difference in the results. Moreover, including a dummy to account for transfer of controlling group is not statistically significant. The countries included in this research are: Albania, Belgium, Belize, Israel, United Kingdom (2), Bulgaria, Venezuela, El Salvador, Cape Verde, China, Chile, Germany (2), Bolivia, Brazil (3), Peru (2), Argentina (2), Estonia (2), France (2), Ghana, Guyana, Croatia, Indonesia, El Salvador, Panama, Jordan, Korea, Latvia, Lithuania, Japan (4), Indonesia, Portugal (3), Puerto Rico, Qatar, Netherlands (2), St Kitts (2), Czech Republic, Singapore (2), Senegal, Finland (2), Sri Lanka, Russia, Switzerland, Denmark, Ireland, Italy, Yugoslavia, Jamaica, Austria, Poland (2), Spain, Guatemala, Sweden, South Africa, Australia (2), Canada, Trinidad and Tobago and Uganda (2). Number of privatizations within country are shown in parenthesis.

⁶ While there is some evidence that firm performance improves after CEO change it is relatively limited and do not explicitly assess the signalling view (e.g., Megginson et al ., 1994; Cragg and Dyck 1999).

⁷ Future research should focus on the fact CEOs may not be randomly matched with firms as CEO selection may be linked with firm characteristics. Ideally one would like to have data on failed privatizations in order to better assess whether CEO affects the probability of privatization or the terms of the privatization conditional on an economic environment. Data limitation does not allow us to provide a clear answer to these issues.