Does electoral politics distort or even “corrupt” the privatisation process in emerging democracies? Drawing on new co-authored research using Serbian micro-data, Luca J. Uberti shows that sales of state-owned enterprises in the months leading up to an election are more likely to be affected by corruption.

In post-communist countries, privatisation has been the most emblematic feature of the transition from a planned to a market economy. Across most of East-Central Europe, the transformation of state-owned enterprises is now complete. Yet, privatisation remains an important policy objective in other transition and developing economies.

In Ukraine, where the private sector share in GDP is still below 70%, the sale of state-owned enterprise has resumed despite the war. In Egypt, large swathes of the economy are still owned and controlled by the military. The IMF and other creditors are now putting pressure on the Egyptian government to reduce the army’s economic footprint. But what are the economic benefits and risks of privatisation?

A large literature in economics has explored the effects of ownership transformation on corporate performance. The evidence from East-Central Europe shows that privatisation had a generally beneficial effect. At the micro level, privatisation helped to reduce the inefficiencies and overstaffing associated with state ownership. At the macro level, a higher private sector share of GDP is thought to have reduced opportunities for political corruption.

In countries with less-effective institutions, however, the impact of privatisation has been more mixed. In Russia, for instance, the sell-off of state property has been linked to the rise of corrupt oligarchs. While the outcome of privatisation may increase efficiency and stem corruption, the privatisation process itself may sometimes have perverse effects. The link between privatisation sales and corruption, however, remains underexplored. This lack of understanding makes it difficult to formulate policy recommendations as to how to reduce corruption risks when privatising state-controlled assets.

Privatisation and elections in emerging democracies

In a recent study, my co-authors and I identify one specific context in which privatisation sales may be unadvisable from a corruption point of view – namely, election-campaign periods in young democracies with weak institutions. This conclusion is arrived at by solving a game-theoretic model in which politicians and firm managers bargain over the decisions of a privatised firm.

In countries with weak institutions and widespread corruption, politicians can strike credible bargains with the buyers of former state-owned enterprises, and thus continue to extract rents after privatisation. In our model, the bargain between politicians and managers takes a simple form. Politicians transfer hidden subsidies to former state-owned enterprises – for example, by means of a one-off discount at the point of sale. In exchange, the managers of privatised firms continue to pursue economically costly strategies such as excess employment, which politicians value because they generate votes.

We show that, during election periods (though not in general), politicians can earn a higher pay-off by privatising state-owned enterprises than by plundering them under state ownership. This finding is underpinned by a simple intuition. During election campaigns, the political costs of bargaining with managers are lower than the political costs of creating or maintaining inefficiencies in state firms, which are typically under heightened public scrutiny.

This mechanism implies that, in pre-election periods, politicians have an incentive to increase privatisation efforts. It also implies that the state-owned enterprises privatised before elections should be more likely to receive subsides and exhibit inefficiencies (for example, excess employment) than those privatised outside of election periods.

Elections, privatisation and firm performance in Serbia

To examine the plausibility of this mechanism, we take the model’s predictions to the data. We focus on Serbia as the privatisation process here only began in earnest after the fall of former strongman president Slobodan Milošević in 2000. We can thus study the interrelationship between privatisation, corruption and elections using recent data.

We comb through sales contracts and other official sources to assemble a unique firm-level dataset covering the universe of former socially owned enterprises in Serbia. Our sources indicate the exact date when each enterprise was put up for sale. They also provide rich information on firm characteristics, the firm’s buyers and the features of the sale itself.

Our findings reveal the perverse effects of elections on the sale and performance of former state-owned enterprises in Serbia. Consistent with politicians having a preference for privatisation over state ownership before elections, we document an election cycle in privatisation activity.

On average, privatisation revenues and the number of privatisation sales more than double in the last three months before a presidential vote (see Figure 1). This result is not driven by seasonality as Serbian elections do not take place at regular intervals. Nor does it appear to be influenced by other election-related factors affecting investment decisions, such as economic stimulus and monetary policy.

Figure 1: Privatisation revenues and the number of sales around elections, relative to the long-term average

Note: For more information, see the author’s accompanying paper in the European Journal of Political Economy (co-authored with Vladan Ivanović and Drini Imami).

Next, we examine the terms of sale and post-privatisation performance of the state-owned enterprises sold in pre-election periods. The data indicate that these firms were both offered and sold at a price discount (around 17%) relative to otherwise similar state-owned enterprises sold in non-election periods. This finding is consistent with incumbent politicians transferring an implicit subsidy to the buyers of state-owned enterprises. We do not find support for alternative explanations: the firms earmarked for sale before elections were not more likely to have gone unsold in previous privatisation rounds (which would have made them less valuable to investors).

Do managers reciprocate by choosing firm strategies that benefit politicians? Our data suggest that they do. Even several years after privatisation, the state-owned enterprises sold in pre-election periods exhibit a less favourable cost structure than those sold in non-election periods. Specifically, being privatised before elections appears to raise total operating costs after privatisation by around 4%. We argue that excess costs may arise from a number of inefficient strategies chosen by corrupt managers and valued by subsidy-paying politicians – notably, excess employment.

Finally, we investigate how corrupt agreements between politicians and managers affect the overall performance of privatised firms. The results are striking. State-owned enterprises sold before elections are 14% more likely to go bankrupt than those sold in non-election periods. Conditional on surviving, they are also 35% less “profit-efficient” – they generate less profit than they should given the technology they use and the market environment they operate in.

Policy implications

This research has important policy implications, especially for international actors such as the World Bank, the IMF and the EU. In the countries of the European periphery – where institutions are weak, corruption often widespread and state ownership still common – the privatisation sales taking place before elections should be closely monitored.

Some countries have introduced controls aimed at containing pre-electoral opportunism and corruption. In Brazil, for instance, public-sector hiring is not allowed in the final months of a mayor’s term in office. In 2020, Albania took the step to ban privatisation sales in pre-election periods. Other countries should follow suit.

For more information, see the author’s accompanying paper in the European Journal of Political Economy (co-authored with Vladan Ivanović and Drini Imami).

Note: This article gives the views of the author, not the position of EUROPP – European Politics and Policy or the London School of Economics. Featured image credit: BalkansCat / Shutterstock.com