This chapter studies constitutional revision provisions at the theoretical level. Country constitutions systematically involve two categories of items: individual rights and the rules of the political game. The emphasis is on the word “systematically” because they may also include other elements.Footnote 1 Individual rights and the rules of the political game in a democracy must be well-known in advance and respected by all participants in the political game. In other words, they require stability over time, which I will call time consistency. For this reason, constitutions protect their text from change by making modification difficult. In Chapter 1 we saw such mechanisms, in Chapter 2 we studied how they work, and in Chapter 6 we corroborated our expectations: More veto players as well as higher required majorities for each one of them increase constitutional rigidity and make amendments overall more difficult and more rare.
As I discussed in Chapter 3, not all analyses share these arguments. A significant proportion argues that it is not institutions but culture that regulate the frequency of amendments. Others claim that instead of looking at the institutions researchers should base their analyses of constitutional rigidity on the frequency of amendments. For example, Marshfield (Reference Marshfield2018) writes, “A better measure of constitutional flexibility is a constitution’s actual amendment rate because this presumably captures both the formal barriers to amendment contained in the amendment rules as well as cultural attitudes regarding formal amendment” (80). Versteeg and Zackin (Reference Versteeg and Zackin2016) agree, saying, “The measure [of constitutional entrenchment] does not rely on formal amendment rules because these rules are mediated so dramatically by political norms” (661). Versteeg and Zackin confirm that the length of a constitution correlates with the frequency of amendments.
This institution-free analysis leads to a relativistic approach to constitutional length. Versteeg and Zackin (Reference Versteeg and Zackin2016) argue that there is an alternative specific and flexible model of constitutions that had not been recognized by the theoretical literature: “We simply seek to demonstrate that the specific and flexible constitutions currently populating the globe are not simply failures to achieve brevity and entrenchment, but represent a plausible alternative solution to some of the agency problems associated with constitutional design.”
This chapter, consistent with the institutional approach of the book, will be based on the constitutional rigidity approach and will develop a new concept called “time inconsistency,” which combines constitutional rigidity and amendment frequency. The idea is the following: the founders of a constitution design the rules of amendment on the basis of how frequently (or, likely, how rarely) they think the constitution should be changed (as we showed in the previous chapter, the frequency of amendments should be lower if the required majority is three-fourths than if it is two-thirds). However, reality may impose different rules. The quality of the existing constitutional provisions and real-life conditions may lead to more or less frequent amendments than what was initially planned for. It is as if the country as a collective actor “changed its mind” with respect to the initially selected constitutional rigidity. I will call this difference “time inconsistency” and calculate it as the difference between actual amendment frequency and the frequency expected based on constitutional rigidity (as calculated in the previous chapter).
I find that time inconsistency correlates with the length of constitutions in all democratic countries. This is not a neutral feature of constitutional length. It is clearly a negative characteristic, and it brings us back to the traditional analyses of constitutions, according to which the “framework” constitutions (Dixon Reference Dixon2014) are the optimal choice. This discovery opens the door for further investigations of other negative associations of constitutional length – GDP per capita, corruption, inequality, and so on – which suggest that long constitutions are not a “plausible alternative solution” to constitutional design but are instead a suboptimal one.
This chapter is organized into three parts. In the first part, I present the interaction, in game form, between the founders of a constitution and the subsequent generations that may choose to revise it. In the second part, I define the time inconsistency concept and reexamine arguments presented by Tsebelis and Nardi (Reference Tsebelis and Nardi2016) as well as Tsebelis (Reference Tsebelis2017b) who claim that time inconsistency is positively correlated with the length of a constitution. The difference is that these papers did not have the more advanced measures of constitutional rigidity generated in Chapter 2, so the results presented here are more empirically accurate. In addition, they cover more countries and (given that they lead to the same conclusions) provide a robustness check for these arguments. In the third part, I examine the implications of the time inconsistency argument and find that constitutional length is correlated not only with time inconsistency but also with a series of economic indicators (such as GDP per capita, inequalities, or corruption). The presentation is not only based on the 103 democratic countries but also uses the comparison of the fifty US states from the work of Brown (Reference Brown2022), who was able to perform more controlled comparisons and discover Granger causality between constitutional length and economic indicators.
7.1 The Intergenerational Constitutional Game
The founders of each constitution ultimately want to generate a document that will regulate the interactions of the political game for generations to come. Whether it is the rights of citizens or the interactions among the political actors, these rules have to be known and respected (and therefore known to be stable) by all political actors. Justice Scalia (Reference Scalia and Gutmann1997) argues that “the whole purpose [of a constitution] is to prevent change – to embed certain rights in such a manner that future generations cannot readily take them away” (40), and Justice Brennan (Reference Brennan1991) argues, “In my view, it is crucial to the durability and efficacy of a charter of personal liberties that it not be subject to easy alteration or suspension … robust entrenchment forbidding compromise or requiring supermajoritarian approval for amendments seems to me best” (4).
On the other hand, if unforeseeable circumstances arise, these constitutional rules have to permit amendment. This is why there are constitutional provisions about the requirements for a constitutional revision.
The theoretical debate in constitutional design is between two major options with regard to the time horizon of constitutions: it can either be one anchored to and shaped by the citizens it represents or be one that stands the test of time. The former perspective represents that of Thomas Jefferson; the latter represents that of James Madison. The two addressed a fundamental question of the role played not only by a nation’s governing document, the constitution, but also by the relation of the governors to the governed: Who decides the rules of the game? Are the living meant to be ruled, as Jefferson argued, by themselves in a revisited document, or should they be ruled by their forbearers through an enduring document?
Jefferson supported constitutional replacement in every generation to allow citizens to revisit institutions and rules, adapting them to changing circumstances. He supported replacing (or at least reevaluating in some form) constitutional bargains every generation, or about every nineteen years – which is, as Elkins et al. (Reference Elkins, Ginsburg and Melton2009: 129) note, the median survival time of constitutions in their sample. Madison, however, took issue with such a suggestion, arguing against instability and in favor of longevity. A government worthy of respect, in Madison’s view, is one that is faithful to its citizens’ wishes while also remaining steadfast in the face of short-lived fads and whimsical ideas. Additionally, long-standing constitutions, according to Madison, are more stable and less susceptible to the “ambition or corruption of one” and the “sagacious, the enterprising, and the moneyed few” (Madison Reference Madison1788).
Figure 7.1 provides the game form of the considerations of founders and future generations. The founders have to decide on three different issues: (1) whether to include a subject matter in the constitution, (2) whether to include many provisions on the subject and make it restrictive, and (3) how much to lock it so it is protected against revisions. Each country gives different answers to these questions. This is why subjects that exist in some constitutions are absent from others, and the locking mechanisms are different not only across countries but even within the same constitution (there may be some eternal clauses while the majority are amendable under specific rules).
Figure 7.1 Writing and revising the constitution game
For future generations, the question of a constitutional revision may arise, and the occurrence will be more frequent as the founders opt to incorporate more subjects and more detailed provisions. The success of such attempts at revisions will be higher the less locked the constitution is (as Chapter 6 demonstrated).
I have indicated with bold letters all the choices that lead to subgame perfect equilibria in this game form. One choice that does not lead to such an equilibrium, however, is the combination of constitutional detail (including a large number of provisions) and a failure to lock them sufficiently, along with the willingness of future generations to modify the constitutional provisions – that is, what Versteeg and Zackin (Reference Versteeg and Zackin2016) call “specific and flexible constitutions … that represent a plausible alternative” (8).
The usual term in the economic literature for the description of such equilibria that are not subgame perfect is “time inconsistent.” Economic theory has long underscored, since Kydland and Prescott’s (Reference Kydland and Prescott1977) Nobel-winning article, “Rules Rather than Discretion,” that time inconsistency ought to be avoided in economic policymaking. This is the standard reason that countries delegate monetary policy to central banks: to take it away from the hands of a government that will change preferences as a function of electoral cycles. This argument has been propagated in the creation of many other independent authorities as well, including environmental protection, mass media, medical regulations, and so on.
If institutions are created in order to avoid time inconsistency in policies, time inconsistency a fortiori should be avoided with respect to the rules of the game – that is, the constitution. In other words, constitutions that change often are subject to discretion rather than rules.Footnote 2
On the basis of Figure 7.1, one can see that long constitutions (involving many detailed provisions) may lead to time-inconsistent outcomes. That is, despite their locking, they may lead future generations to overcome the obstacles and revise the constitution. The same thing is true about locking. If the rules become seriously obsolete, then locking may not be sufficient.
7.2 Implications and Data Analysis
The above analysis confirms two major points that have already been made in this book. First, constitutional amendments are out of (perfect) equilibrium behavior. Second, these amendments are a difficult enterprise that are actually undertaken only when solutions within the constitutional equilibrium (legislation or judicial interpretation) do not work.Footnote 3 Here I will introduce the concept of time inconsistency, measure it in a formal way, and then try to relate it to the length of a constitution. In the previous chapter, we related the constitutional rigidity of a country with its amendment frequency (or rate of amendments).
If we call 
the frequency of amendments, r the rigidity of the constitution, and k and a positive constants, we demonstrate that on average
Let us now call 
′ the actual frequency of amendments in each country. On the basis of the discussion so far, we call the difference between the actual and the expected frequency of amendments time inconsistency (t):
Combining (1) and (2), we conclude that
We will demonstrate that time inconsistency is proportional to length – that is, if we call l the (logarithm of) length of a constitution, then
Before we move on to the empirical analysis, I want to demonstrate that besides the findings of Chapter 6 indicated by equation (1), there is now a significant new contribution to the literature. What was previously known in the literature is that the frequency of amendments is proportional to length. This finding was usually supported in an independent way by showing that longer constitutions will include more provisions and therefore will have a higher need for amendments (Lutz Reference Lutz1994: 357 and 359, Rasch and Congleton Reference Rasch, Congleton, Congleton and Swedenborg2006: 542, Lijphart Reference Lijphart2012: 207). However, more recently some researchers (Versteeg and Zackin Reference Versteeg and Zackin2016, Marshfield Reference Marshfield2018) do not consider constitutional rigidity at all and replace it with amendment frequency. If one eliminates 
from equation (2), then (3) and (4) will lead to
In other words, the actual frequency of amendments is proportional to the length of the constitution, which is what Versteeg and Zackin (Reference Versteeg and Zackin2016) find in their analysis. But what escorts this noninstitutional approach is an association of length with the frequency of amendments instead of the association with time inconsistency that I demonstrate. The result is a relativistic approach to constitutional length (constitutions used to be short; now they are becoming longer, and there is nothing wrong with that). Here is the way they phrase it: “Specificity and flexibility are highly correlated with one another and appeared to have increased together in democratic constitutions… Their flexibility allows them to avoid the ‘dead-hand’ problem, since the living generation clearly acts as the principal in its frequent revision of the constitutional text” (Versteeg and Zackin Reference Versteeg and Zackin2016: 660). On the contrary, my analysis here makes the case that length is an undesirable characteristic associated with time inconsistency (as well as with other undesirable features that we will see). My expectation does not deviate from the assessment of most of the literature.
Let me start by replicating the findings of the literature with my data. Figure 7.2 shows the positive relationship between log length and amendment rate. Figure 7.3 presents the visual representation of my argument, which is that length is associated with time inconsistency. The slope of time inconsistency is less steep than the slope of amendment rate because time inconsistency is, by definition (see equation [3]), the actual rate minus the expected rate (on the basis of constitutional rigidity).
Figure 7.2 Amendment frequency and log length
Figure 7.3 Time inconsistency and log length
As I said in the Introduction, I will start by replicating the analyses of Tsebelis and Nardi (Reference Tsebelis and Nardi2016) and Tsebelis (Reference Tsebelis2017b), but I will use the constitutional rigidity measure introduced in this book instead of the more crude measures used in these articles. This way, we will have a more accurate assessment of the arguments. Tsebelis and Nardi (Reference Tsebelis and Nardi2016) only measure Organization for Economic Co-operation and Development (OECD) countries not only because the data were more easily available but also because they were expecting that the hypothesized relationships (in this case, time inconsistency and length) would be more discernible than in the wider set of democratic countries. Tsebelis (Reference Tsebelis2017b) confirms this intuition but still uses alternative and rough indicators of rigidity. Table 7.1 separates OECD countries from all democracies and uses the more theoretically founded and empirically accurate indicator of time inconsistency.
Table 7.1 Time inconsistency as a function of constitution length
| OECD | DEMOC | |
|---|---|---|
| n | 33 | 100 |
| Log length | 0.541 | 0.391 Footnote * |
| (0.306) | (0.153) | |
| Constant | –2.339 | –1.795 Footnote * |
| (1.318) | (0.667) | |
| R2 | 0.111 | 0.080 |
* p < 0.05.
The first assessment is that the slope of OECD countries is steeper (0.45) than all democracies (0.34). However, the fact that there are three times more democratic countries than OECD countries makes the relationship significant for all the democratic countries but not for the OECD ones. This is true for the constants as well. The positive relationship in both cases indicates that longer constitutions have a higher time inconsistency (that is, the combination of rigidity and frequency of amendment). I will now consider this relationship more closely.
7.3 Correlates of Length (Time Inconsistency, GDP per Capita, Corruption, Economic Inequality, Lack of Innovation, etc.)
In order to explain this time inconsistency, one needs to first understand the characteristics of long constitutions and then identify other factors that are associated with them.
7.3.1 What Is Length?
Constitutions can include three different kinds of provisions. First, constitutional provisions can regulate technical or innocuous matters that do not influence political behavior. Second, constitutions can contain aspirational goals, such as the right to work (included in many post–World War II constitutions), which do not impose any specific obligations on the government and are consequently not judicially enforceable (unsurprisingly, none of these countries have completely eradicated unemployment). Thirdly, constitutions contain restrictive or prescriptive statements, such as sections detailing government structure and citizens’ rights. While these three categories might be straightforward at the theoretical level, empirically there is no reliable way of distinguishing between constitutions that contain many substantive restrictions and those that are simply “garrulous” (Voigt Reference Voigt2009). However, the frequency of amendments along with the difficulty of achieving such modifications indicates that long constitutions are restrictive because a country would not undertake the significant or formidable efforts required for amendments if these amendments were not deemed necessary. In other words, constitutional amendments are more likely to be made on restrictive provisions, not on innocuous ones.
Another question regarding length pertains to how words are distributed over topics in the constitution: Are there many topics with little discussion, a very detailed discussion of a few topics, or is it somewhere in between? The Comparative Constitutional Project dataset (Elkins et al. Reference Elkins, Ginsburg and Melton2009) makes the distinction between the “scope” of a constitution (that is, the number of selected subjects included in it) and its “detail” (the number of words used to cover each subject on average). Obviously, the length of every constitution is the product of the two. Given this logical relationship, a regression predicting the length of a constitution (as a function of scope, detail, and their interaction) would provide a coefficient of 1 for the product term and an 
of 1. In other words, both variables cannot be used in the same equation. One could drop the distinction between these two variables and talk about their product (length). However, if we want to investigate along these lines, we can proceed as follows: It is known in the literature that more recent constitutions have a larger scope (i.e., they address more subjects); therefore, I can use the age of the constitution as a proxy for scope, provided the variable “age” is uncorrelated with “detail.” As Figure 7.4 indicates, this is the case in all the countries of the world (regardless of whether they are democracies or not).
Figure 7.4 Scope and details in 187 countries
Now it is possible to identify the characteristics associated with length using the age of the constitution as a proxy for its scope. Table 7.2 examines the variables in the literature associated with the length of constitutions, focusing first on OECD member countries and then on all democracies. The variables I examine are age, detail, federalism, and legal origins. Again, the statistical significance is higher for all democracies than it is for OECD countries, but the coefficients are larger for OECD countries. The important finding is that “detail” has a positive coefficient and “age” (which is a proxy for “scope”) has a small negative one. It is interesting that federalism has a negative but not significant effect, indicating that instead of expanding the constitution to include the interactions between federal and state governments the federal constitutions delegate many issues to the state governments. Also, the legal origins have no effect on the regressions. The conclusion is that across all democratic countries of the world, constitutional length is associated with more restrictions.
Table 7.2 Constitutional length as a function of country characteristics
| OECD | DEMOC | OECD | DEMOC | |
|---|---|---|---|---|
| n | 33 | 99 | 33 | 99 |
| Detail (calc) | 2.401 ** | 1.839 *** | 2.560 *** | 1.869 *** |
| (0.667) | (0.475) | (0.690) | (0.467) | |
| Federalism | –0.108 | –0.014 | –0.088 | –0.013 |
| (0.073) | (0.037) | (0.080) | (0.038) | |
| Age of democracy | –0.003 * | –0.002 *** | –0.003 | –0.002 * |
| (0.001) | (0.001) | (0.002) | (0.001) | |
| Legal origins | 0.091 | 0.031 | 0.082 | 0.027 |
| (0.091) | (0.062) | (0.093) | (0.061) | |
| Num amendments | –0.002 | –0.001 | ||
| (0.002) | (0.002) | |||
| Constant | 4.092 *** | 4.045 *** | 4.065 *** | 4.041 *** |
| (0.150) | (0.084) | (0.160) | (0.083) | |
| R2 | 0.807 | 0.758 | 0.815 | 0.759 |
*** p < 0.001; ** p < 0.01; * p < 0.05.
7.3.2 What Is Associated with Constitutional Length?
For long constitutions to be more time inconsistent – that is, to exhibit a higher number of amendments, despite locking – they must also lead to serious impediments to the political game in their corresponding countries. Tsebelis and Nardi (Reference Tsebelis and Nardi2016) identify two important correlates of constitutional length in OECD countries: GDP per capita and corruption.
Long constitutions are restrictive, and, as such, they prevent the adoption of policies that are desirable to the populations they regulate. This may be a reason for frequent constitutional amendments. One aggregate variable that would cause generalized dissatisfaction and would therefore cause constitutional revisions would be low GDP per capita. Table 7.3 corroborates the inverse relationship between constitutional length and GDP per capita (a relationship depicted graphically in Figure 7.5). In addition, with respect to corruption, Tsebelis and Nardi (Reference Tsebelis and Nardi2016) argue that causal links could be pointing in both directions: It could be that founders were captured by special interests who were asking for additional detailed provisions to be locked so that their privileges would be guaranteed. Alternatively, it may be that virtuous founders tried to include provisions in order to prevent or reduce the influence of organized interests.
Table 7.3 GDP per capita as a function of constitutional length and economic variables
| OECD | DEMOC | OECD | DEMOC | |
|---|---|---|---|---|
| n | 32 | 97 | 30 | 70 |
| Length (log) | –0.299 ** | –0.442 ** | –0.264 * | –0.225 * |
| (0.108) | (0.149) | (0.103) | (0.101) | |
| Education | 0.001 | 0.006 *** | ||
| (0.002) | (0.002) | |||
| Natural resources | 0.002 | –0.004 | ||
| (0.009) | (0.006) | |||
| Trade | 0.001 | 0.000 | ||
| (0.001) | (0.001) | |||
| Investment | –0.015 | –0.016 * | ||
| (0.008) | (0.008) | |||
| Constant | 5.767 *** | 5.919 *** | 5.819 *** | 5.156 *** |
| (0.449) | (0.650) | (0.485) | (0.539) | |
| R2 | 0.271 | 0.071 | 0.418 | 0.314 |
*** p < 0.001; ** p < 0.01; * p < 0.05.
Figure 7.5 Log GDP and log length
Tsebelis and Nardi (Reference Tsebelis and Nardi2016) also anticipated that these relations would be clearer in OECD countries because these countries respect their constitutions, and, consequently, safer inferences can be made from the study of OECD countries. In Table 7.4, I include education and corruption as control variables (on top of the economic ones). This inclusion removes the statistical significance of length on GDP per capita, which occurs because corruption is strongly correlated with constitutional length as Figure 7.6 demonstrates.
Table 7.4 GDP per capita as a function of length, economic variables, education, and corruption
| OECD | DEMOC | OECD | DEMOC | |
|---|---|---|---|---|
| n | 30 | 70 | 30 | 70 |
| Length (log) | –0.264 * | –0.225 * | –0.044 | –0.025 |
| (0.103) | (0.101) | (0.103) | (0.079) | |
| Education | 0.001 | 0.006 *** | –0.000 | 0.003 * |
| (0.002) | (0.002) | (0.001) | (0.001) | |
| Natural resources | 0.002 | –0.004 | –0.001 | 0.001 |
| (0.009) | (0.006) | (0.010) | (0.005) | |
| Trade | 0.001 | 0.000 | 0.001 | 0.001 |
| (0.001) | (0.001) | (0.001) | (0.001) | |
| Investment | –0.015 | –0.016 * | –0.007 | –0.004 |
| (0.008) | (0.008) | (0.006) | (0.005) | |
| Corruption (TPI) | –0.055 *** | –0.111 *** | ||
| (0.011) | (0.011) | |||
| Constant | 5.819 *** | 5.156 *** | 4.390 *** | 3.652 *** |
| (0.485) | (0.539) | (0.565) | (0.433) | |
| R2 | 0.418 | 0.314 | 0.696 | 0.744 |
*** p < 0.001; ** p < 0.01; * p < 0.05.
Figure 7.6 Corruption and log length
However, more recent and controlled analyses give stronger results. Adam Brown (Reference Brown2022) examines the constitutions of the US states not in a cross-sectional analysis as I do but in a time-series comparison, finding that longer constitutions generate more amendments, more frequent judicial adjudications, and more negative economic indicators. I will here discuss his economic findings in more detail because they are indeed more convincing than the cross-country analysis of this book.
Brown does not find a relationship between length and constitutional rigidity and considers the assessment “long constitutions are restrictive” in Tsebelis (Reference Tsebelis2017b) as a consequence of constitutional rigidity, while it is actually the result of more “detail” (see discussion around Figure 7.4 and Table 7.2 in this chapter).Footnote 4 Because of this, he proposes length itself as an independent variable. In the words of this book, he does not differentiate between “scope” (how many different topics are included in a constitution) and “detail” (how many words are used for each topic). His argument is that the inclusion of anything in the constitution (whether it be a different subject or more details in the text) is restrictive.
He also tests for a correlation between length and corruption and does not find any in his data. By contrast, he finds that more length was used in the constitutions adopted between 1870 and 1900 when political elites and authorities were mistrusted, not in relation to corruption.
Brown uses twenty years of constitutional history of the US states (except for Alabama) with biannual evaluation of constitutional length (which sometimes changes by 4,500 words!), as well as other control variables, and focuses on four dependent variables: GDP per capita, unemployment rate, the Gini coefficient of income inequality, and state policy innovativeness (Boehmke et al. Reference Boehmke, Brockway, Desmarais, Harden, LaCombe and Linder2018). He controls for a series of variables, such as the strength of parties, percentage of different minorities, age of the constitution, and number of amendments, and finds that constitutional length has a significant impact on all his dependent variables: a negative impact on GDP per capita, positive on the unemployment rate, positive on the Gini coefficient (inequalities), and negative on policy innovativeness. He goes one step further, testing for Granger causality – that is, he examines whether in his time series the lagged constitutional length can predict the economic indicators and whether the lagged economic indicators can predict constitutional length. He finds that the first but not the second is true, concluding, “We may therefore say that constitution length Granger-causes economic and policy performance, but economic and policy performance do not Granger-cause constitution length – and this is true for all four outcomes considered here” (Brown Reference Brown2022: 136). This empirical evidence of the effect of length is consistent with my arguments but goes much further in terms of empirical evidence because it controls for a series of cultural variables (one can presume that US states are significantly more homogeneous than the countries I am considering) and because time series are much more conclusive causal evidence than cross-sectional analyses.
Conclusions
This chapter demonstrates that long constitutions are restrictive and generate a higher rate of amendments than what they were designed for (time inconsistency). Given the difficulty of performing constitutional revisions, such revisions are not likely to be undertaken without reason. They are likely to affect enforceable provisions that are hindering government majorities from acting the way they judge to be appropriate. In this sense, they are constraining majorities from making decisions according to their wishes, and they are confronting the democratic expression of the representatives of the people. Therefore, long constitutions are not just garrulous (Voigt Reference Voigt2009), but they are also restrictive, as Brown (Reference Brown2022) argues. Constitutional revisions may be due to design if a constitution is not rigid (that is, if amendment provisions are permissive). However, this chapter shows that the length of constitutions across all democratic countries of the world is correlated with time inconsistency – that is, the combination of locking and amendment frequency. This means that long constitutions do not just “represent a plausible alternative solution to some of the agency problems associated with constitutional design” (Versteeg and Zackin Reference Versteeg and Zackin2016) but represent a suboptimal solution as recognized for many years by the literature.
This chapter starts with an equilibrium analysis of the constitutional amendment provisions and shows that if constitutional amendments are to be successful in democracies then they require the support of majorities that exceed the limits specified by the amendment provisions of the constitution. More than that, this chapter demonstrates that the difference between the actual rate of amendments and the rate expected by the amendment provisions depends on the length of the constitution, and it shows that this length has an impact on the time inconsistency of the constitution (that is, the combination of locking and amendment frequency).
The usual means to eliminate time inconsistency in the literature is to delegate to an independent authority. This is not a possible solution in constitutional matters, though, because there is no higher authority than the people. If the people modify a constitution despite the obstacles included by the founders, it means that either there were radically new conditions or there was a design flaw due to potentially objectionable provisions being included and protected in the constitution. These provisions were later considered impediments either because the conditions changed or because large majorities changed their minds. Therefore, the best way to reduce time inconsistency is to avoid locking the constitution and/or avoid lengthy ones. Only rules that are widely accepted and are not likely to be overturned should be locked.
This is not the only argument in the literature. As I mentioned, Ginsburg and Melton (Reference Ginsburg and Melton2015) argue that “the constitutions of India, Mexico, and Brazil, to take three prominent examples, are amended nearly every year. Such constitutions have the virtue of being frequently changed through internal mechanisms, avoiding the costly route of a total replacement. In such countries, we argue that the stakes of amendment are lower, and so cultural resistance to amend is less than in societies where it is infrequent” (689). Indeed, all three of these countries are highly time inconsistent, but the Indian case is due to the extremely lengthy constitution (the amendment rule is similar to legislation). In this respect, there is little formal difference between India and a country without a written constitution like the UK, where parliament can change any law it wants to by a simple majority. As for Brazil (with 68,000 words of constitutional text and a three-fifths majority required in both chambers on two different occasions for amendment), a detailed analysis is required. However, when analyzing the Brazilian Constitution, Couto and Arantes (Reference Couto and Arantes2008) find that “The Brazilian constitution of 1988 presents a high rate of constitutional amending, with 62 amendments in twenty years (3.1 amendments per year); most of them sponsored by the Executive branch, aiming at implementing public policies” (Couto and Arrantes Reference Couto and Arantes2008: 1; emphasis mine). They argue that there is a high percentage of policymaking provisions inside the constitution, and they create a new measure of constitutional provisions, finding that 30 percent of them are policy related.Footnote 5 However, further research is required for these two cases. The Mexican case has already been discussed in Chapter 5 where I argued that the only visible “virtue” was political, being the attempt for wide-ranging coalitions. Further, these coalitions make constitutional amendments as easy as legislation and, consequently, a dominant solution for the parties.
The analysis presented in this chapter focuses on time inconsistency. The general approach in the time inconsistency literature is that, at the beginning of the game, institutional measures (rules) should be taken to prevent time inconsistency from manifesting itself (discretion).Footnote 6 With respect to constitutions, the analogy would be that “a constitution is Peter sober while the electorate is Peter drunk” (Holmes Reference Holmes, Elster and Slagstad1988: 195–196). Both Hayek (Reference Hayek2006: 157) and Elster (Reference Elster2010) raise objections to such an approach. I argued that given the difficulty of constitutional change, adopted constitutional amendments were necessary, and I argued that the restrictions that were included in the original constitution are essentially undermining the essence of a document that sets the rules of the game. Therefore, it would be more reasonable to reduce the restrictive provisions (by reducing the length) instead of locking the constitutions more. Dixon (Reference Dixon2014) has divided constitution writing into “codified-” and “framework-” styled approaches and provides legal arguments in favor of the latter style. I provide a similar empirically generated argument that long constitutions are restrictive. If my analysis is correct, the authors of the first constitution of a country should exercise constraint and not assume that they can lock anything they want in the constitution. Doing so leads to long, time-inconsistent constitutions. However, this is a “retroactive” suggestion with twenty-twenty hindsight. A prospective suggestion would be to have the people who engage in constitutional revisions take the time to prune their constitutions. In other words, if a certain provision is restrictive, it would be more efficient to just drop it instead of replacing it with a different one.
What the correlation of length with time inconsistency indicates is that too many things are locked in constitutions, which results in the undermining of their effectiveness. Therefore, length and locking of constitutions is not a matter of culture but of arrogance and lack of restriction on the part of constitution writers. To use Jeremy Waldron’s terms, “Any alternative conception that might be concocted by elected legislators next year or in ten years’ time is so likely to be wrong-headed or ill motivated that his own formulation is to be elevated immediately beyond the reach of ordinary legislative revision” (Waldron Reference Waldron1999: 222). These arguments iterate on constitutional grounds the wisdom of the ages, such as the biblical verse Matthew 5:37 (English Standard Version): “Let what you say be simply ‘Yes’ or ‘No’; anything more than this comes from evil,” or even older sayings like “λακωνιζειν εστι φιλοσοφειν,” or “brevity is the source of wit.”Footnote 7 These assessments should be considered seriously when studying constitutions and should be abided by when writing constitutions.
Open access