An empirical analysis of disinvestment policy in India: does ideology matter?

Abstract

Using panel data estimation for limited dependent variables and sample selection models, we identify political, industry specific, firm specific and macroeconomic variables which influenced the decision to divest central public enterprises in India between 1991–2010. We find that higher partial privatization is driven by a more right-winged coalition, lower ideological spread in the coalition and lower ideological difference between the center and the state in which the public enterprise is located. We also find that after the government selects larger, more experienced and more profitable firms, it divests the relatively less experienced and less profitable firms to a larger extent.

Keywords:

• Political economy
• privatization
• panel data
• sample selection
• ideology

AMS Subject Classifications:

• P16
• D22
• L33
• C33

Acknowledgements

We thank Luis Filipe Martins’, Sudha Narayanan, C. Veeramani, Soumyanetra Munshi and Nityananda Sarkar for comments and suggestions. We are also grateful to numerous seminar participants for their valuable feedback. All remaining errors are ours.

Notes

No potential conflict of interest was reported by the authors.

Supplemental data for this article can be accessed https://doi.org/10.1080/17487870.2018.1457961.

1 We claim that this result will hold even if the incumbent state party is more market-oriented that the one at the center.

2 Memorandum of Understanding: MoU is a mutually negotiated document that is signed annually at the start of the financial year between the management of the SOE and the administrative ministry in the government. As per the MoU, the management decides the performance targets to be accomplished and the government agrees on the support to be given during the year. At the end of the financial year, performance assessment is done where the achievements are measured against the targets. This is based on both financial and non-financial parameters with specific weights allotted to each parameter using a five-point scale varying from Poor to Excellent.

Ratna status: The government has accorded the status of Maharatna, Navratna, Miniratna–Category I and Miniratna–Category II to certain SOEs based on their performance in the last five years. These SOEs have been empowered with operational and financial autonomy to equip themselves to react proactively to market forces.

3 See Annexure (supplementary file) for details.

4 Disinvestment proceeds are available only for select years. If we had a continuous time series of all disinvestment proceeds, we could have accounted for the price of shares which is an important dimension of disinvestment.

5 Ideology scores of all parties are presented in Table 6.

6 State ideology is not the weighted measure of all parties forming a state level coalition due to unavailability of coalition data.

7 Annexure is available as an online supplementary file.

8 We also use the estimates of the stage one equation to calculate the inverse Mills ratio for each observation, and use it as an explanatory variable in the second stage equation to control for selectivity bias.

9 Results of all robustness tests have been included in an online supplementary file.

10 Average asset size of an SOE is the ratio of the total asset size in all years divided by the total number of years.

11 Proportion of firms which were selected for disinvestment and were predicted correctly.

12 Proportion of firms which were not selected for disinvestment and were predicted correctly.

13 Clustering is done on the centre level political factor to control for the grouped effect across all firms.