Share this:

Adam Cayton 80x108Just as the US has a founding document – The Constitution – the states themselves have their own constitutions. But what influences how and when these documents are altered? In new research, Adam Cayton compares the partisan makeup of state legislatures in every year since 1834 to the year in which a state’s constitution was enacted to see when they are replaced. He finds that a state’s constitution is more likely to be replaced when the constitution includes a large amount of policy content and the composition of the legislature has changed.

Governments do not last forever, and while war and revolution are the most dramatic causes of institutional death, in the US, political systems are replaced peacefully, and by the very people who work within them. Social scientists, historians, policy makers, and founders of governments devote a great deal of thought and ink to identifying the kinds of institutions that are stable over the long term, and rightly so, but to understand why some institutions last longer than others it is not enough to focus on their characteristics. We also need to understand how those characteristics influence the people who live under them to try to keep or replace the institution. Thinking about institutional lifespans from this perspective causes us to reframe it as an interesting puzzle: Why would politicians, who have succeeded under the existing rules of the game, collectively decide to change those rules? This happens more often than one would think, and a possible explanation is that the rules prevent them from using elected office to accomplish the goals that led them to seek that office in the first place. A common source of this problem is when institutions deny politicians authority over certain policies.

A good way to test this proposition is to compare the lifespans of constitutions, a nearly ubiquitous institution among large political units. It is tempting to view constitutions as high minded statements of ideology, values, or political culture, and as documents that are worthy of reverence and above the political fray. To do so, however, is to misunderstand how constitutions are used in most states, whether in the international system or in federations such as the United States. Constitutions define the rules under which politicians and interest groups compete for power and create policy. These rules by their nature give advantages to some and hinder others. It is simply unrealistic to expect political actors, who are strategic in crafting and voting on policy, structuring executive agencies, designing electoral and legislative procedures, and organizing political parties, to not have the same approach when it comes to revising constitutions.

In new research, I find that when state constitutions are more policy focused, they are more likely to be changed when the state’s legislature becomes more and more different in terms of party affiliation from how it was when the original document was drafted.

While all constitutions are political documents, some are more likely to be affected by factional conflict than others. Constitutions define the internal framework of most democratic systems, but may also include more specific provisions that directly set policy, making the policy difficult or impossible to change by normal legislative means. This is good for those who like the policy because it places it out of reach of their political opponents. For politicians who took office after the constitution was created, and who disapprove of the policy, this kind of constitutional content impedes their goals and thwarts the perceived will of the voters. If enough politicians share this frustration they may decide to invest the time and effort to replace a constitution. Political institutions that limit the policies that can be enacted and were created by people with different preferences from current policy makers face greater risk of being replaced or revised. These policy-laden, or “particularistic,” constitutions do not last as long as their counterparts that focus primarily on the framework of government because they are more likely to be replaced as politicians’ preferences change.


The 50 US states provide a unique opportunity to study this proposition. The US Constitution allows states to design their own institutions, provided only that they maintain a “republican form of government.” States occasionally use this freedom to undertake a wholesale replacement of their governmental framework. Another scholar has compiled a dataset measuring the amount of policy specific content in US state constitutions. Less straightforward is measuring change in the preferences of politicians in each state. The best feasible way to do this is to use information on the partisan composition of each state legislature, which is available from 1834. By measuring the composition of a state legislature the year its constitution went into effect and in each subsequent year, it is possible to capture how much a state’s political elite diverges, in a given year, from the time its constitution was enacted. Party affiliation does not mean the same thing in every state and cannot be directly tied to ideology in any consistent way in the United States across this entire period, but change in the partisan composition of a legislature always reflects some kind of political change and it has the advantage of being recorded for most of US history.

Modeling the relationships between constitutional particularism, partisan change, and the risk of replacement lends support to the argument above. The chances that a state constitution will be replaced in a given year go up by over 40 percentage points for every 10 percent change in party seat shares, and the chances of replacement are between 58 and 79 percentage points greater for every additional 10 percent of policy-specific content. Closer examination shows that partisan change has no effect on the risk of replacement when a state’s constitution has no policy-specific content, and states with lots of policy content are at no greater risk of replacement when the composition of the legislature remains constant. It is only in those states where the constitution includes a large amount of policy content and the composition of the legislature changes that replacement is more likely. The takeaway is that if politicians enshrine their policy preferences in the highest law of the land it may be safer from legislative meddling, but over the long term such action may lead their frustrated successors to totally revise the government.

This article is based on the paper, ‘Why Are Some Institutions Replaced while Others Persist? Evidence from State Constitutions’ in State Politics and Policy Quarterly.

Featured image credit: Basharat Alam Shah (Flickr, CC-BY-2.0)

Please read our comments policy before commenting.      

Note:  This article gives the views of the author, and not the position of USAPP – American Politics and Policy, nor the London School of Economics.

Shortened URL for this post:


About the author

Adam Cayton 80x108Adam Cayton – University of Colorado
In the fall of 2016, Adam Cayton will join the Department of Government at the University of West Florida as an assistant professor. His research focuses on representation, legislative institutions, and institutional change in the United States.