In a recent journal article George Tsebelis and Dominic J. Nardi, Jr. present statistical evidence that longer constitutions are associated with lower levels of GDP per capita. In this post they summarise their findings and speculate that this may be because detailed constitutions are more likely to prevent governments from adopting measures necessary to combat economic shocks.
When the Eurozone crisis struck in 2009, it soon became apparent that southern European countries would have to drastically restructure government spending and improve their competitiveness in order to reduce excessive levels of public debt. However, some states found themselves unable to do so, and not merely because of lack of political will. In some cases, detailed socioeconomic provisions in their constitutions limited their scope of action. For example, Greece’s 1975 Constitution prohibits the privatization of universities (art. 16(5)). In short, such constitutional provisions risk locking governments into policies that in other countries would not even be considered questions of constitutional law.
The debate over the ideal level of detail in a constitution has gone on since at least the US Constitutional Convention of 1787. Founding Father James Madison famously argued for a framework constitution that would simply delineate government authority. By contrast, Anti-Federalists feared that brevity might leave important rights unprotected, which prompted the subsequent enactment of a more detailed Bill of Rights. Thanks to the advent of statistical computing software and cross-national datasets, the comparative constitutions literature has begun to assess the impact of constitutional detail outside the U.S. In an article in the British Journal of Political Science, we seek to address one portion of this debate: does the length of a constitution have an effect on economic development?
Broadly speaking, we consider three types of constitutional provisions. First, a constitution can regulate technical matters, which might be critical to the operation of government but tend to be politically innocuous. For example, some constitutions describe the national flag, but the colours are seldom debated in political discourse. Second, constitutions can contain aspirational goals or ideological statements. Revolutionary regimes often issued new constitutions to announce major shifts in policy. Finally, constitutions contain restrictive or prescriptive statements that constrain government action. For example, the US President cannot circumvent the constitutional requirement that he seek the ‘advice and consent’ of the Senate for presidential appointments (as demonstrated in NLRB vs. Canning).
Of course, not all constitutions are equally binding. It is difficult ascertain which constitutions exhibit bindingness solely from the text. Even text that appears restrictive in theory might have a minimal effect in practice if there is no external enforcement mechanism. To circumvent this problem, we limited our sample to 32 countries that are members of the Organisation for Economic Co-operation and Development (OECD) and have written constitutions. The OECD is an exclusive organization and sets criteria for admissions. Members must commit to democracy, the rule of law, and open market economics, which the OECD verifies. However, unlike the EU, it is not limited to a single region, allowing for some geographic diversity. Although economic development in the OECD is not representative of the entire world, there is still considerable variation in GDP per capita.
We first used data from the Comparative Constitutions Project (CCP) to compare the length (measured by the number of words), detail (measured by number of words divided by number of topics), rigidity of amendment procedures, and frequency of amendments for all OECD constitutions in 2006. This led to a puzzle: more detailed constitutions tended to be more difficult to amend, but were also amended more frequently! Because more rigid amendment provisions set higher voting thresholds against passage, proposed amendments that succeed must be the result of an overwhelming desire to change the status quo. Moreover, because amending a constitution is costly, few politicians would invest the resources necessary merely to change an innocuous provision. As such, the results suggest that more detailed constitutions are amended more frequently despite being more difficult to amend because they are more likely to contain restrictive provisions that impede the government’s ability to pass necessary or desired legislation.
Next, we ran a statistical regression to test the effect of constitutional detail on the economy of each OECD country during the late 2000s. Even controlling for standard socioeconomic variables, such as investment, education, and natural resources, we found that detailed or longer constitutions are associated with lower levels of GDP per capita (PPP). We believe this correlation suggests that detailed constitutions are more likely to prevent governments from adopting measures necessary to combat economic shocks. By contrast, amending constitutions (measured by the number of amendments) seems to increase GDP per capita, suggesting that countries that can more readily enact amendments can mitigate the negative effects of longer constitutions.
This of course begs the question: why do some countries choose to enact more detailed constitutions? We found that constitutional length is highly correlated with measures of corruption from both the World Governance Indicators and Transparency International. Longer constitutions tended to appear in countries that suffered from greater corruption. Because our data is limited to a single year, we cannot determine if corruption causes longer constitutions, or vice versa. We could imagine two different causal pathways. First, knowing that they cannot trust political elites to adopt sound economic policies, drafters might adopt longer constitutions in order to combat corruption. Alternatively, corrupt special interest groups could convince constitutional drafters to protect their interests from future challenges. In this case, longer constitutions would stem from elite capture and rent seeking.
Although we certainly appreciate and acknowledge the important benefits that constitutional pre-commitments provide, such as property protection, our research suggests that there are economic costs associated with longer, more detailed constitutions. Of course, the next logical question is which parts of longer constitutions are most detrimental. We suspect that, all else equal, longer provisions related to socio-economic issues (e.g., natural resources, right to work) will have a greater effect than longer provisions related to technical or innocuous matters. Unfortunately, no currently available datasets contain information about the length of specific topics within constitutions. We hope that such data can be collected to allow us to better understand why longer constitutions seem to have a negative impact on economic development.
To read the full article, ‘A Long Constitution is a (Positively) bad Constitution: Evidence from OECD Countries’ in the British Journal of Political Science, please click here.
About the authors
George Tsebelis is the Anatol Rapoport Collegiate Professor of Political Science at the University of Michigan.
Dominic J. Nardi, Jr. is a PhD candidate at the University of Michigan Political Science Department.