Constitutions and Social Trust: An Analysis of the US States.

Author:Ahnen, David
  1. Introduction

    Societies with anemic social capital tend to draft longer national constitutions than societies with more robust social capital and social trust (Bjornskov and Voigt 2014). The idea is that a lack of social trust--"authority relations, relations of trust and consensual allocations of rights which establish norms" (Coleman 1988)--requires political and commercial contracts to be more specific. We ask if this logic extends to subnational polities in the United States. All fifty states have constitutions, which vary substantially in length. Vermont has the shortest constitution of any state at about 8,500 words. Alabama's constitution, at more than 380,000 words, is the longest of any government worldwide. (1)

    In our analysis, we use state-level volunteer rates from 2002 through 2015 as our measure of social capital. La Porta et al. (1997), Jennings and Stoker (2004), and Brown and Uslaner (2005) all document a strong and positive relationship between volunteerism or civic engagement and social trust. We examine the relationship between social trust and constitutions in two ways: first, in verbosity or length, and second, in endurance--the amount of time between the adoption of new constitutions. A US subnational approach is different from a cross-country analysis as each state is subject to two constitutions: one unique to the state and the other binding across all states. This dual-constraint structure gives us reason to examine the relationship between constitutions and social trust at the state level to determine if the findings of Bjornskov and Voigt (2014) hold. Section 2 summarizes the existing literature and background on constitutional garrulity and places this paper within the context of the broader scholarship. Section 3 details the data and methods used in our analysis. Section 4 presents the results, and section 5 concludes. (2)

  2. Literature and Background

    The relationship between social capital and a contract's verbosity has been well established (see Knack and Keefer 1997; La Porta et al. 1997). Knack and Keefer (1997) hypothesize that social capital facilitates economic activity in places that lack formal institutions to enforce contracts and secure property rights. Thus, the return on investment for regulations and institutions that formally enforce contracts and secure property rights is higher in places where trust is low (Knack and Keefer 1997). High-trust contexts are endowed with unwritten contracts, which, to varying degrees, can supplant aspects of formal contracts (Bjornskov and Voigt 2014).

    In The Calculus of Consent: Logical Foundations of Constitutional Democracy, James Buchanan and Gordon Tullock (1962) describe the political economy of efficiently designing a constitution. While constitutional architects attempt to constrain undesired political behavior, they face high transaction costs for creating a consensus. Within this framework, Bjornskov and Voigt (2014) argue that any constraint's usefulness will partially be a function of its ability to preempt misconduct by contemporary and future political actors. Thus, the constitutional architects of polities that anticipate more misconduct will include more constraints on political behavior (Bjornskov and Voigt 2014).

    Expectations for misconduct are similar to the notions of social capital in the context of constitutional garrulity and specificity. (3) When expectations for misconduct are high and social capital is low, constitutions will be longer and more specific because costly constraints on political behavior are justified. When social capital is high, constitutions will be shorter.

    Voigt (2009), using World Values Survey data on trust from 135 countries, finds general support for the hypothesis that higher levels of social capital are correlated with shorter constitutions. However, the inclusion of Voigt's trust variable shrinks his sample to thirty countries. Thus, Voigt states that he does "not put too much trust in this result" (2009, p. 296). Bjornskov and Voigt (2014), using the same 135-country dataset and various country barometers, confirm that the relationship between social capital and constitutional length at the international level is both negative and statistically significant.

    Other scholars have examined the political and economic implications of constitutional verbosity. Montenegro (1995) finds that countries with longer constitutions have a lower GDP per capita. Similarly, Tsebelis and Nardi (2016) demonstrate that countries with longer constitutions are more corrupt and have constraints on political behavior that impede free markets, governance, and wealth creation.

    As noted, scholars identify volunteerism as conceptually distinct from, albeit highly related to, social capital, and using volunteerism to measure social capital has a well-established precedent in the literature. (4) Both Putnam (2000) and Hawes et al. (2013) use volunteerism in their social capital indices. Brehm and Rahn (1997) argue that social trust at the individual level is the complementary association between civic engagement and interpersonal trust. Moreover, the strong and positive relationship between volunteerism or civic engagement and trust has been well documented by La Porta et al. (1997), Jennings and Stoker (2004), and Brown and Uslaner (2005).

    At the state level, Hawes et al. (2013) examine social capital levels in each state. Jackson et al. (2015) examine social capital and economic freedom. Berkowitz and Clay (2005) find that both the age and legal origins of the state are positively and significantly correlated to constitutional length. In contrast, Hammons (1999) demonstrates that short state constitutions tend to be replaced more often by newer constitutions, while their more verbose counterparts tend to endure.

    1. Social Trust: Stability over Time

      This paper presupposes that social capital is an antecedent to constitutional choice and predicts that social trust affects constitutional verbosity. Since this paper uses social trust data from 2002 through 2015 to analyze older constitutions (the most recent is North Carolina's 1983 constitution), a question arises: Can contemporary social trust explain previous decisions about constitutional length? Ideally, state-level social capital data would be readily available from 1780, the year Massachusetts enacted its first and only constitution, to the present. However, they are not: scholars of social capital have long noted the challenge of reliably measuring social capital in the late twentieth century, much less the late eighteenth (Hawes et al. 2013; Putnam 2001).

      This paper reconciles the theoretical concern by following the approach of Bjornskov and Voigt (2014), who show that trust is relatively stable over time. However, the stability of trust is an ongoing debate that pits the "institutionalists" against the "culturalists." According to the institutionalist school, social capital is formed post hoc by the interactions between individuals and their society's institutions. Thus, institutionalists argue that trust is malleable over time and that changes in social capital are attributable to institutional changes (Dinesen 2012a, b). Conversely, the culturalists argue that social capital shapes institutions. Rooted in child psychology, their position is that because individuals acquire a basic sense of trust in childhood, social capital is stable over time since parents pass along their own levels of social trust to their children (Katz and Rotter 1969).

      Moreover, scholars across disciplines have found evidence to suggest that social capital is, in fact, stable over time. Nunn and Wantchekon (2011) argue that regional variations in the number of individuals captured in the nineteenth-century slave trade help to explain contemporary regional variations in social capital. Their findings suggest that the roots of Africa's levels of social trust are centuries deep. Uslaner (2008) argues for the heritability of social trust by illustrating that ethnic heritage is more important than culture and ethnicity for determining social trust. Bjornskov and Svendsen (2013) examine the experiences of third-generation Americans and find evidence that social capital is stable over time, and Kesler and Bloemraad (2010) find that immigration does not erode social capital. Thus, studies suggest that social capital is determined by culture and is stable enough over time that contemporary levels can reasonably function as a proxy for historical levels of social capital.

    2. Federal vs. State Power

      The constitutions of the fifty states operate under the same federal governing structure, have a common language, share a similar history, and have remarkably similar gubernatorial, legislative, and judicial institutions. However, the federalist structure provides greater latitude for state constitutions to address the local concerns of a subnational polity (Ginsburg and Posner 2010; Delledonne and Martinico 2010; Hammons 2001; Connor and Hammons 2008). Thus, a tension exists between federal and state power. The Supremacy Clause, Article six of the US Constitution, makes any valid exercise of federal laws enumerated in the Constitution superior to any state law. The Tenth Amendment to the US Constitution attempts to address this delineation of power by acknowledging, "The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people."

      Most of the thirteen colonies established their first constitutions before the Articles of Confederation in 1781 or the US Constitution in 1789. Massachusetts, with its first constitution enacted in 1780, has the oldest operating constitution in the United States. Vermont, which has the shortest constitution, has had three constitutions. The last one was established in 1793, only four years after the United States adopted its constitution. Similarly, New...

To continue reading