Evaluating alternative measures of institutional protection of private property and their relative ability to predict economic development.

Author:Bennett, Daniel L.
 
FREE EXCERPT
  1. Introduction

    While scholars appear to have reached a consensus that North (1990) was correct in his assertion that a society's institutions define the incentive structure that economic actors face and are therefore key determinants of economic performance, there is less of a consensus regarding which institutions are most important for the development process. Some authors argue that democratic political institutions drive economic development (e.g., Acemoglu et al. 2014; Persson and Tabellini 2003; Rodrik 2000), while others stress that economic and legal institutions are essential in the development process (e.g., Barro 1996; Gwartney, Holcombe, and Lawson 2006).

    A multitude of institutional measures are now at researchers' disposal and have been used in empirical research on institutions and economic performance. Although there is some debate regarding the specific institutions that matter most for economic development, there is considerable agreement that the protection of private property is necessary for sustained, long-run economic development.

    Secure private property rights incentivize the efficient use of economic resources by enabling economic actors to engage in rational economic calculus. When property rights are well-defined and secure, economic actors can internalize the full economic costs and reap the full economic benefits or losses of their decisions. Actors can engage in profit and loss accounting and market prices are a reliable indicator of the relative value of a resource's alternative uses. The market process directs resources away from unprofitable activities and into more profitable and hence productive ventures, promoting economic growth.

    Without secure property rights, prices are distorted and fail to convey accurate information through markets in a timely manner to signal the relative profitability of alternative uses of a given resource. In addition, insecure property rights increase the attractiveness of short-term investments relative to longer-term and potentially more productive investments because the higher risk of expropriation reduces the latter's expected returns. Insecure property rights therefore lead to less productive employment of resources and less innovation, stymieing economic development.

    Several studies show that institutions that protect private property rights are a fundamental cause of economic development (Acemoglu, Johnson, and Robinson 2001; Bennett et al. 2016; Faria and Montesino 2009; Gwartney, Holcombe, and Lawson 2006; Hall and Jones 1999; Knack and Keefer 1995; Rodrick, Subramanian, and Trebbi 2004). Although a growing body of literature supports the notion that institutional protection of private property rights is vital for sustainable economic development, researchers continue to use alternative measures for this variable. A growing body of literature also exists that examines the methodology of institutional measurement (Berggren, Bergh, and Bjornskov 2012; Feld and Voigt 2003; Langbein and Knack 2010; Munck and Verkuilen 2002; Voigt 2012, 2013). This paper assesses the methodological strengths and weaknesses of five commonly used measures of property rights, as well as their relative ability to predict the level of economic development.

    The empirical results suggest that the International Country Risk Guide risk of expropriation, Fraser Institute and Heritage Foundation property rights indices, and World Governance Indicator rule of law index measures are all robustly correlated with the level of economic development, and the estimated effects are quantitatively similar across measures. Meanwhile, the Polity IV executive constraints measure has a much smaller and less robust impact on economic development. The methodological analysis suggests that the executive constraints measure may, to a degree, reflect limitations on the political executive's ability to engage in predatory extraction of private property, but it does not reflect similar constraints placed on other governmental bodies or private actors. It is therefore, at best, an incomplete and weak proxy for the protection of private property rights from public predation. The other four measures are based on a strong theoretical concept of property rights, potentially explaining their robust correlation with economic development. This paper also assesses several other methodological features of the measures, including the aggregation method, data objectivity, whether the indicators measure de facto or de jure institutions, the public availability of the data, and the indicators' transparency and replicability.

    Section 2 recaps the institutional measurement methodology literature, and section 3 analyzes the five measures of institutional protections of private property. Section 4 describes additional data used in this study. Section 5 presents the empirical results, and section 6 concludes.

  2. Institutional Measurement and Methodology

    Social scientists, policymakers, and development agencies are increasingly concerned with the importance of institutions for economic development and other socioeconomic concerns, as evidenced by the surge of research estimating the quantitative causal impact of institutions on a variety of outcomes. Munck and Verkuilen (2002, p. 5) indicate that while "this is a welcome development ... quantitative researchers have paid sparse attention to the quality of data ... that they analyze." Voigt (2013, p. 2) adds that "if one wants to show that institutions matter ... one needs a reliable way to identify and measure them." A growing body of literature examines institutional measurement methodology. Five issues this literature has raised are relevant to the current study.

    First, the literature stresses the importance of linking theory to the measurement of institutions. Voigt (2013, p. 9), for instance indicates that the "attempt to measure institutions needs to be driven by an underlying theory." Munck and Verkuilen (2002, p. 7) suggest that the first step in constructing an institutional measure is to identify "attributes that are constitutive of the concept under consideration ... which amounts to a specification of the meaning of the concept." Langbein and Knack (2010, p. 351) add that indicators of an abstract concept such as an institution should "systemically and reliably relate to the concept." The degree to which an institutional indicator measures the concept it is intended to measure reflects the indicator's validity (Munck and Verkuilen 2002).

    Second, the aggregation of variables into an institutional indicator is also a concern. Munck and Verkuilen (2002) outline a methodology for aggregating institutional indicators that follows from the conceptualization step described previously. A highly abstract institutional concept should be broken down into subsequently lower levels of abstraction that delineate certain attributes and components of the metaconcept. The components are measurable and can be aggregated to measure the higher order attributes. The attributes can, in turn, be aggregated to generate a measure of the highly abstract concept. (1)

    The authors stress the importance of selecting multiple components drawn from multiple sources to reduce the potential for biases associated with a single variable. The aggregation method should, to the extent possible, be driven by theory concerning the relationship between the various components and attributes of an institutional concept, and should be done in such a way as to prevent the loss of useful information. Voigt (2013, p. 2), on the other hand, argues that "measures of institutions should refer to specific institutions because aggregate measures ... are too broad and fuzzy to contain meaningful information." Studies by Berggren, Bergh, and Bj0rnskov (2012) and Langbein and Knack (2010), respectively, examine the International Country Risk Guide and World Governance Indicators data, two common composite institutional indices.

    Third is the issue of objective versus subjective variables in the construction of institutional indicators. Objective variables are based on underlying objective data, while subjective variables are based largely on survey data that capture subjective evaluations of survey respondents that may be systematically biased by the respondents' perception of how the institutional indicator they are being asked to evaluate affects some other observable measure of the prevailing economic, political, and/or social conditions. Although institutional indicators derived from objective data are generally preferable, data availability is often limited to subjective measures, and subjective indicators may provide useful information that is difficult to ascertain from objective measures (Langbein and Knack 2010; Voigt 2013).

    Fourth is the issue of de facto versus de jure institutional measures. Voigt (2013) indicates that effort should be made to measure both the legal and practical specifications of an institution because there is interest in studying the effects of both. Feld and Voigt (2003), for instance, provide evidence that economic growth is affected by de facto but not de jure judicial independence. The interaction of de facto and de jure institutions may also be important. For example, Justesen (2014), Justesen and Kurrild-Klitgaard (2013), and Voigt and Gutmann (2013) provide evidence that the interaction of de facto legal and de jure political institutions promotes growth.

    Lastly, this literature suggests that the data used to derive an institutional measure should be made publicly available and the methodology transparent for replicability. In addition, these two conditions permit researchers to derive customizable institutional indicators to allow for testing of the robustness of results (Munck and Verkuilen 2002; Langbein and Knack 2010).

  3. Measures of Institutional Protection of Private Property Rights

    The effective protection of private property requires a set of...

To continue reading

FREE SIGN UP