رفع ابهام در سیاست های آزادسازی و خصوصی : آیا سیاسی تجارت کردن وجود دارد؟

We empirically investigate the political determinants of liberalization and privatization policies in six network industries of 30 OECD countries (1975–2007). We unbundle liberalization and privatization reforms and study their simultaneous determination in a two-equation model. Unlike previous studies, we account for cross-effects between the two pro-market measures. Our findings unveil that both right-wing and left-wing governments implement liberalizations and privatizations, showing a common trend under the so-called neo-liberalism wave. However, although the privatization rate is higher than liberalization in right-wing environments, the opposite occurs under left-wing governments. We argue that ideological cleavages still affect pro-market reforms, particularly the combination of privatization and liberalization policies. We conclude that different deregulation patterns should be expected under governments characterized by different political ideologies. Our results shed new light on the literature investigating the political-economic rationale underpinning pro-market choices.

In the last three decades, most OECD countries have experienced a deep and extensive wave of pro-market reforms in sectors once dominated by State-controlled monopolies, such as network industries (Armstrong and Sappington, 2006, Guriev and Megginson, 2007 and Roland, 2008). Policies aimed at removing restrictions to entry into previously heavily regulated markets (e.g., electronic communications, transportation, energy, postal services) and promoting government withdrawal from corporate ownership have been implemented throughout the world, and especially in OECD countries (Conway and Nicoletti, 2006), stimulated by globalization and by the diffusion of pro-market reforms. In addition to the analysis of economic determinants (Vickers and Yarrow, 1991, Levy and Spiller, 1996, Newbery, 1997, Newbery, 2002 and Armstrong and Sappington, 2006), a large group of scholars have investigated the role of institutional and political determinants of market-oriented policy in network industries, following the “political economics” approach (Persson and Tabellini, 2000, Persson, 2002, Besley and Case, 2003 and Besley et al., 2010). Within this framework, empirical studies commonly consider privatizations and liberalizations as a whole, under the comprehensive umbrella of pro-market policies (e.g., Pitlik, 2007, Bortolotti and Pinotti, 2008 and Potrafke, 2010). However, disentangling privatization and liberalization initiatives suggests that the extent and implementation of the two policies in OECD network industries followed quite a heterogeneous pattern. Some countries have actually focused more on decreasing barriers to entry (i.e., liberalization policies) in previously State-owned monopolies, and some others have been much more concerned with the reduction of State ownership in incumbent firms (i.e., privatization policies).1 Thus far, the empirical literature has failed to acknowledge country-level variations in which liberalizations and privatizations have been differently combined in network industries. Consequently, we still lack a full understanding of how so many countries adopting similar reforms, simultaneously and in identical sectors, actually differ in their policy “bundling”, intensity and implementation of, respectively, liberalizations and privatizations. The purpose of this article is to investigate precisely the determinants behind such cross-government variations in pro-market policy adoptions. Different political and economic configurations, partisan politics, and interest concentrations may play roles in this respect. Here, all else being equal, we focus our analysis on the role of partisan orientation in the government in shaping the different liberalization–privatization patterns we observed in OECD network industries over the last thirty years. A traditional view still supports the idea that governments dominated by right-wing parties tend to espouse liberal economic programs, including deregulation, privatization and liberalization, and, generally, the introduction of competition (Boix, 1997 and Rodine-Hardy, 2013). Conversely, left-wing governments typically are assumed to foster stronger State intervention, defending “national champions” as massive employers and heavily regulated markets as a secure manner with which to control price and wage dynamics (e.g., Appel, 2000). According to this view, the enduring wave of pro-competitive initiatives should be exclusively attributed to right-wing executives. Recent empirical research shows, however, that after the nineties, the so-called “second-wave neoliberalism” also flooded left-wing governments, boosted by a process of policy diffusion and imitation in the global arena (Dobbin et al., 2007 and Stiglitz, 2008). Accordingly, we should expect a convergent pattern between right-wing and left-wing governments, with the former fully embracing the entire range of pro-market policies and the latter progressing toward a similar path. Surprisingly, available data, when properly parsed, suggest quite a different story. In Fig. 1, we report the intensity of liberalization and privatization policies adopted by right-wing and left-wing governments in OECD network industries, measured as the yearly variation of, respectively, the OECD’s 2009 indicators of entry barriers and public ownership over the last three decades. Data show that right-wing governments pushed for privatization policies more intensively than left-oriented ones, whereas left-wing governments favored liberalization over privatization.Disentangling pro-market policies reveals that, within the common neo-liberalism wave, a partisan trade-off between privatization and liberalization characterizes OECD countries’ governance of network industries. Thus, on the one side, the alleged primacy of right-wing governments in advocating opening markets to competition – by both privatization and liberalization policies – is to be challenged. On the other, data suggest that traditional ideological biases have not been completely absorbed by converging policy diffusion processes in the globalization era. Rather, alternative pro-market paradigms, based on various combinations of privatization and liberalization intensity, appear to have emerged in OECD network industries, driven by governments’ partisan orientation. If confirmed by rigorous econometric inquiry, this result would overturn the main conclusions reached thus far by the existing empirical literature, forcing reconsideration of the political and economic rationale behind these two alternative pro-market paradigms as they have evolved and been reinforced over the last decades. In this paper, we investigate whether such observed patterns are robust enough for a rigorous econometric analysis. We employ the largest available database on pro-competitive policies (OECD, 2009). This database covers 30 OECD countries observed from 1975 to 2007. In addition to being the longest time span for which rigorous data are available, this period also includes the entire liberalization and privatization waves observed in OECD network industries. We used information on sectoral privatization and liberalization concerning six sectors (passenger air transport, telecommunications, electricity, gas, post, and rail) and thus utilized three sources of exogenous variation (country, time, and industry). We then estimated two equations (one explaining privatization interventions and the other explaining liberalization interventions) using Seemingly Unrelated Regression (SUR) to account for the presence of unobservable factors responsible for the simultaneous determination of privatization and liberalization. Our econometric analysis identifies the existence of a causal relation underpinning the graphical correlation between governments’ ideology and pro-market policies displayed in Fig. 1, with right-wing parties favoring privatization over liberalization and left-wing parties favoring liberalization over privatization. These effects are shown to be robust in a number of different empirical specifications. In addition to the econometric findings, our study contributes to existing empirical political and economic literature on pro-market reforms in three manners. First, we approach liberalization and privatization policies as two distinct components of a general pro-market reform process. This is a significant innovation because standard empirical literature tends to treat the two policies as two aligned and substitutable issues under the comprehensive umbrella of pro-market reforms (e.g., Potrafke, 2010). The distinction between liberalization and privatization policies has often been confused in the economic literature, and the two concepts have often been overlapped and conflated in a generic notion of “pro-competitive deregulation”. It is not only a matter of definition; it is also a matter of content and of alternative political and economic rationales behind the governmental choice of a policy mix. In our study, by disentangling privatization and liberalization, we emphasize that the two policies may have different economic and political motivations and consequences. Whereas liberalization should properly denote the abatement of legal provisions impeding the free entry of new competitors into a market and does not directly involve corporate ownership changes, privatization is a process of formerly Stare-owned firms going private and does not imply legal free entry into the market (without liberalization, privatization simply determines a shift from a public to a private monopoly). Second, we acknowledge that the effects of liberalization are not independent of the level of privatization in the market and vice versa. On the one hand, entry barrier abatement before privatization reduces the monopolistic rents of the incumbent and lowers the price at which the State will be able to sell its shares of a company; on the other, the State’s withdrawing from ownership before liberalization increases the returns that the State can obtain from privatization. This is suggested by many studies on pro-market policy effect (De Fraja, 1991, De Fraja, 1994, Roland, 1994, Wallsten, 2001, Wallsten, 2002, Li and Xu, 2002 and Armstrong and Sappington, 2006); however, it has been thus far ignored by the extant empirical political economic literature. Third, we employed the largest database available on market reforms in network industries and systematically studied both liberalization and privatization processes from the beginning to the end whereas other empirical inquires have examined shorter periods and smaller groups of countries. Our choice to focus on network industries is motivated by the fact that these sectors have long been characterized by the presence of natural monopoly segments, network externalities, and firms having non-economic objectives (e.g., universal service obligations). Given these peculiar characteristics, network industries have typically been burdened with legal restrictions to entry, widespread public ownership, and extensive cross-subsidies to a larger extent than other sectors. Only since the end of the 1970s have technological advances, the evolution of governance and regulatory techniques, and an increasing international exposure made liberalization and privatization possible in these sectors, thus allowing governments to reveal in practice their pro-market preferences. OECD network industries are therefore an appropriate and interesting case for exploring the reform aptitudes of governments. Moreover, the intense reform processes in network industries have stimulated economists to collect quantitative information of a higher quality than for other sectors, thus making rigorous data on these industries available for econometric analysis. Our paper fits easily into various veins of political economic literature. As for the political determinants of pro-market reforms, our study relates to a number of empirical papers that have focused alternatively on privatization or liberalization policies in isolation, most often suggesting that both policies are a prerogative of right-wing governments (Duso, 2002, Pitlik, 2007, Bortolotti and Pinotti, 2008, Arin and Ulubasoglu, 2009, Duso and Seldeslachts, 2010, Potrafke, 2010, Belloc and Nicita, 2011 and Belloc and Nicita, 2012). More generally, our paper relates to the literature on partisanship and macroeconomic policy-making (Alesina and Tabellini, 1988, Alesina and Rosenthal, 1995, Roemer, 2001, Biais and Perotti, 2002 and Osterloh, 2012) and on policy diffusion (Simmons and Elkins, 2004, Clifton et al., 2006 and Rodine-Hardy, 2013). Our study also coincides with the literature on the economic determinants of pro-market reforms (Bertero and Rondi, 2002, Berg et al., 2012 and Bortolotti et al., 2013), to the extent that it shows how free competition spurs progressive State-ownership reductions and vice versa in regulated environments. Finally, it is worth emphasizing what we do not do in this paper. Our empirical analysis provides statistically significant evidence on governments’ policy strategies (i.e., whether they favor liberalization over privatization or vice versa) but cannot tell us anything regarding the governments’ political strategies (i.e., why they favor liberalization over privatization or vice versa). However, although assessing political strategies of governments requires a dedicated interdisciplinary analysis that goes well beyond a pure econometric study, we indicate how our findings can stimulate future research on this issue in the concluding section. The remainder of the paper is organized as follows: in Section 2, we present our estimation analysis of liberalization and privatization reforms in OECD network industries; in Section 3, we check the robustness of our estimation results against variables such as governments’ stability, policy diffusion and outlier values; in Section 4, we draw our conclusions.

Although it is well documented how political and economic globalization processes have sustained the increase in pro-market reforms in OECD countries and particularly in network industries, a detailed analysis of the manner in which such reforms have been designed, articulated and combined at the country level has received much less attention. Disentangling privatization and liberalization policies in OECD network industries reveals a significant country-level variation in the manner in which governments have combined the two policies. However, because existent empirical political economy research failed to outline this pervasive feature of pro-market policy implementation, we lack understanding of why OECD countries have chosen, over the same period, different options within the identical pro-market policies paradigm. In this paper we argued that, even under the common trend of “neoliberalism wave”, the political ideology of governing parties has influenced the pro-market initiatives implemented by OECD countries over the last thirty years. We have unbundled liberalization and privatization policies and investigated how they have been adopted by executives of different political affiliations. Our econometric findings reveal that right-wing governments privatize to a greater extent and liberalize to a lesser extent than left-wing governments; thus, different countries have taken different pro-market paths according to the political orientation of the government in office. We also showed that this empirical result is robust to the presence of cross-effects between liberalization and privatization, to sectoral path-dependency in policy adoption, and to policy diffusion across European countries. Our results suggest a much more complex dynamic surrounding the structure of the pro-competitive reform process in network industries than is commonly proposed by empirical political-economic literature. In particular, the evidence provided in this paper reveals that political ideology influences the composition of the policy combinations chosen by governments rather than the reforms’ aggregate level as is generally argued. What can be learned from our study is that even within a common shift toward pro-market reforms at the global level (the so-called neo-liberalism waves of the ‘80s and ‘90s), ideology still affects policy design and that the relation between policy and politics cannot be reduced to the anachronistic view of left-wing governments hindering competition and right-oriented parties promoting market development. Although contributing to current political economy literature with counter-intuitive and rigorous econometric evidence, our empirical inquiry may also influence three different areas of future research. First, our results suggest the importance of investigating the economic effects of different policy sequencing. Several economists have argued that gradualism in policy adoption is crucial to the success of a pro-market reform process (e.g., Dewatripont and Roland, 1992 and Roland, 1994): privatizing without first granting free entry hampers the emergence of effective competition in the market and an efficient corporate restructuring of incumbents. If governments choose different liberalization–privatization paths, then it is interesting to measure the economic outcomes of the various policy mixes and to understand whether an optimal sequencing of reforms does exist. Second, our analysis may also encourage deeper econometric work on the effect of economic policies on political equilibriums and indirectly on subsequent economic outcomes. Causality factors may indeed run both ways, i.e., from politics to economic policy and vice versa. In the econometric model presented in this paper, we used lagged ideology and governmental characteristics variables to circumvent endogeneity issues. Nevertheless, to investigate the effect of pro-market reforms on executives’ re-election and composition is an area of research that deserves further exploration. Third, we believe that more effort should be expended in collecting qualitative data on the reforms’ progress. Available measures of the liberalization and privatization initiatives implemented in OECD countries allow a quantitative evaluation of pro-market processes (see OECD, 2009) but do not say anything regarding the quality of the reforms. How has the process of the removal of entry barriers been combined with regulatory infrastructures? How have privatization programs been conducted with respect to the method of sale and the valuation procedure? Which systems of both corporate and industrial governance have emerged from the liberalization–privatization wave? The production of new qualitative data would be of great help in answering these and other questions that await a conclusive analysis.