ASSESSING STATE CAPACITY LIBERTARIANISM.

• Author: Murphy, Ryan H.
• Position: Free economic institutions, maximal growth rate - Report

Tyler Cowen (2020), in a controversial and widely discussed blog post, has argued that free economic institutions must be accompanied by state capacity to achieve maximal growth rates. He calls this "State Capacity Libertarianism," which echoes positions he has posed previously (Cowen 2007, 2018). Besley and Persson (2011) can be perhaps seen as a direct predecessor. Criticisms immediately emerged, with Henderson (2020) arguing that Cowen's specific proposals are in direct conflict with libertarianism, and with minor caveats, free economic institutions are already able to achieve the goals Cowen hopes to achieve with state capacity. Geloso and Salter (forthcoming) argue that the lack of examples of wealthy countries with weak states is due to survivorship bias, and they apply their argument to criticize Cowen (Geloso and Salter 2020). Caplan (2018), while not directly addressing State Capacity Libertarianism, argues that there is little reason to believe that the effects of state capacity are the result of strong states themselves, rather than the social and cultural factors that allowed a strong state to emerge in the first place. The purpose of this article is to put data to the question of the individual effects of state capacity and free economic institutions on economic performance, and the potential interaction between the two.

State capacity can be decomposed, primarily, into two broad sets of concepts. The first is fiscal capacity, which is the ability of a state to raise revenue should it need to; additionally, states higher in fiscal capacity are able to raise revenue in a more efficient manner. The second is to play a market-supporting role. At minimum, this is to mean legal capacity, which in libertarian tenns would mean for the country to be capable of running a night watchman state efficiently and effectively. But "supporting markets" could encapsulate enforcing certain kinds of regulations, the provision of public goods, or fixing externalities. The emphasis is not how far the state goes in tenns of actually performing any of these tasks, but how well it is able to accomplish these tasks, should it attempt to. Savoia and Sen (2015), Johnson and Koyama (2017), Berwick and Christia (2018), and Piano (2019) offer reviews of the literature from alternative perspectives.

This article's approach to assess State Capacity Libertarianism is to test the effects of economic freedom and state capacity, and their interaction, on economic growth using a series of regressions employing panel data. While panel analysis which tests the effects of institutions on economic performance is very common, the literature on state capacity tends to have a more historical flavor. In contrast, we make use of two contemporary measures of strong state institutions which have a limited time element, in addition to two less conventional measures, one of which originates with Murphy (2019) and the other of which is original to this paper. The use of a control for output per capita and country-level fixed effects accounts for some of the immediate concerns of State Capacity Libertarianism's critics: that the effect of state capacity is the result of backwards causality, or that cultural variables underlie state capacity. We do not find strong support for State Capacity Libertarianism, but we also do not find that state capacity plays no role in economic development. When considering our findings as a whole, we conclude that, as a first approximation, state capacity and economic freedom are substitutes for one another.

Before commencing with our analysis, we will first put state capacity in a broader context to help motivate our project. The state capacity perspective is actually a rather natural outgrowth of the seminal work done by North and Weingast (1989) and North (1990) which highlights the importance of the Glorious Revolution as the turning point in the modern history of Britain, and therefore, the origins of world economic growth. But while the work of North and his co-authors point most strongly to the Glorious Revolution leading to much stronger protection of property rights, the most tangible outcome of the Glorious Revolution may have been to allow the state to raise more tax revenue (Cox 2012). That is to say, the Glorious Revolution created much higher fiscal capacity.

Recognizing this, subsequent literature has found the two aspects of state capacity to be strongly intertwined (Besley and Persson 2009; Karaman and Pamuk 2013). A broad conclusion of the literature is that an external threat, such as the need to fight wars, causes states to invest in fiscal capacity, the investment of which allows states to gain other capabilities (Besley and Persson 2008, 2009, 2010; Dincecco and Prado 2012; Karaman and Pamuk 2013; Gennaioli and Voth 2015), and these institutions ultimately lead to economic growth (Acemoglu et al. 2016; Dincecco and Katz 2016). While institutions protecting property rights may appear to be what is wholly driving these positive results, provocatively, Irigoin and Grafe (2012) have gone as far as arguing that development in Spain was held back due to too little fiscal capacity relative to legal capacity. And one of the most important recent works in political economy, Acemoglu and Robinson's (2019) The Narrotv Corridor, in effect argues that economic development requires state capacity to exist in some form of balance with the power of civil society, so that they can complement one another rather than one overwhelming the other.

State Capacity Libertarianism is the latest in a series of fusions of libertarianism with more conventional political ideologies, which began itself with Fusionism: the fusion of libertarianism with conservatism (see Meyer 1996). Much later, in reaction to the dominance of Fusionism within the libertarian movement, a series of academics and public intellectuals argued for "liberaltarianism," which blended libertarianism and left-liberalism (Lindsey 2006; Tomasi 2012: 162-63). Since then, there has been an attempt to rejuvenate the word "neoliberalism," which historically was largely used as a pejorative, as a means to somewhat moderate libertarianism in an ideologically ambiguous way, while emphasizing a belief in modern institutions (Bowman 2016; Murphy 2020: 161; Tait 2020). Cowen (2020) explicitly links State Capacity Libertarianism to liberaltarianism, while its relationship to "neo-neoliberalism" is readily apparent as well.

State Capacity Libertarianism, however, comes in direct conflict with certain strands of literature in political economy holding that state enforcement of property rights is redundant in the context of private institutions and norms (Stringham 2015; Williamson 2011) or is actively counterproductive (Benson 1990; Leeson 2014; Leeson and Harris 2018). The idea that free economic institutions must be paired with strong state institutions is anathema to this point of view. The aforementioned Geloso and Salter (2020) are working in this tradition in their criticism of State Capacity Libertarianism, while their formal work (forthcoming) confronts state capacity more directly than the tradition has confronted it previously.

The article is structured as follows: first, we discuss our measures of state capacity, the data sources, and empirical model; second, we report our regression results; third, we interpret the regression results while taking into account the effects of the interaction term; finally, we offer some concluding remarks.

Measures of State Capacity

How can state capacity be quantified? In this section we describe four measures of state capacity; (1) the Government Effectiveness measure from the Worldwide Governance Indicators (Kaufmann et al. 2009, 2010); (2) the average of Government Effectiveness and Control of Corruption; (3) a new index constructed using data from the Varieties of Democracy dataset; and (4) a measure of the strength and power of states from Murphy (2019). Each of these variables possesses a fair amount of cross-sectional and time series variation in a panel, which will be necessary for our empirical approach.

Three fundamental aspects of state capacity are emphasized in the literature: the ability of the state to support markets (which in terms of measurement closely overlaps with the quality of bureaucracies), the state's fiscal capacity, and to a lesser extent, the state's monopoly on violence. The first of these, intuitively, is inclusive of the enforcement of property rights. But more broadly, it should be thought of as the state's capability to do what attempts to do, whether that means enforcing property rights, building an electrical grid, or building a road. Sweden and Singapore are both strong in this, even tiiough Sweden does much more than Singapore. Fiscal capacity is the degree to which the state is able to raise revenue from those living under its control. A state will not have the capacity to compensate a professional bureaucracy or provide public goods if it lacks the capacity to raise revenue through taxation. Finally, as Olson (2002) describes, without a monopoly on coercion (and control of its territory), the threat of violence or violence itself increases the risk of expropriation and reduces the incentives to invest in productive activities (North 1990; Collier 1999).

Government Effectiveness

Government Effectiveness, like other elements of Worldwide Governance Indicators is created using a pastiche of outside sources, which documentation separates into "representative" and "nonrepresentative" sources. Those that are representative pertain primarily to the quality of the bureaucracy and the quality of the provision of public goods and originate with The Economist Intelligence Unit, the Global Competitiveness Report, the Gallup World Poll, the Institutional Profiles Database, the Political Risk Services' International Country Risk Guide, and Global Insight Rusiness Conditions and Risk...