How Historical Constitutional Data Informs Current Policy Debates
Joe Amick, Terrence Chapman and Zachary Elkins
Introduction
In March 2020, the European Union (EU) suspended their fiscal rules as member countries rolled out responses to the COVID pandemic. These rules, known as the Fiscal Stability Treaty (FST) (an outgrowth of the Stability and Growth Pact (SGP)), would have required member countries to cut debt below 60 percent of GDP and keep annual deficits to below 3 percent of GDP. As expected, with the fiscal rules suspended and national governments taking action to mitigate fallout from the pandemic, deficits rose, reaching 7% in 2020, compared to 0.5% the previous year. By the end of summer 2021, debt to GDP ratios exploded, averaging almost 93 percent across Europe with some countries as high as 209.3% (Greece), 160.0% (Italy), 137.2% (Portugal), 125.7% (Cyprus), 125.2% (Spain), and 118.6% (Belgium). No European country was spared from this trend, as all 27 member states saw increases in their debt to GDP ratios from Q1 2020 to Q2 2021.
At the beginning of 2022, there are signs for optimism. Approximately half the world has received a COVID vaccination and a new highly effective treatment is on the way for those unlucky enough to catch it. Unsurprisingly, policymakers have begun to examine the fiscal outlook beyond the pandemic. In Europe, a key question for policy-makers centers on the reinstatement of SGP requirements and whether the EU should change the rules that have governed member countries for years. Leading economists have already begun to push for more flexibility, but as of now, several European governments remain unconvinced. Regardless, this debate is only beginning and will continue until the old rules are reinstated in 2023 or new rules are agreed upon to take their place.
Balanced Budget Requirements
The fiscal compact in the FST requires EU countries to implement, in domestic law, provisions that would ensure a balanced budget. Also, the compact prefers, but does not require (Fiscal Compact, art. 3, sec. 2), that signatories enact the provision as part of the country’s constitution, which represents an extraordinary attempt to tie governments’ hands. A balanced budget restriction is also what the Venice Commission, the body that advises European governments on all things constitutional, has recommended. According to our research, some 33 countries have included such a restriction at some point in their constitutions (Figure 1). Many of these texts include “escape clauses” for exceptional circumstances, such as war or economic crisis.
Figure 1: Periods in which constitutional budget balance provisions have been in force Sample/Universe: 722 of 854 Constitutional systems in force between 1789-2015
But the question for policy-makers as they consider new rules for Europe is whether such a measure actually works to reduce (and balance) budgets. After all, the commitment to a balanced budget can resemble a New Year’s resolution; creative Presidents and legislators should be able to come up with some rationale to overspend and undertax, even in normal times. And certainly, a global pandemic combined with an economic recession would seem to be two compelling reasons.
Do these kinds of rules actually affect balances? It turns out, on average, they do. In a recent study, we included the rule in a predictive model of a country’s primary balance, which excludes interest payments. We estimated the effect for a wide sample of countries throughout the world from 1950 to 2011. Our analysis shows that having a balanced budget provision is associated with an average increase in a country’s primary balance of between 1.7 and 1.9 percent of GDP, depending upon the statistical model. For some perspective, the budget deficit in the United States typically hovers around 3 percent of GDP, right around the limits imposed by the SGP. A constitutional rule that could more than halve a large country’s budget deficit would be a powerful rule.
The Data
Our source of constitutional data is the dataset from the Comparative Constitutions Project (CCP), whose sample includes roughly 95% of the world’s constitutions written and amended between 1789 and 2020. The variable in our analysis is a binary indicator of whether a given country has a constitutional balanced budget rule in place in a given year. The CCP dataset is the product of a rather involved historical endeavor. CCP authors (Elkins, Ginsburg, and Melton) have identified the chronology of constitutional events for each country since 1789, collected the constitutional text associated with each event, and coded the text for some 650 characteristics. They also maintain an indexed repository (Constitute) of constitutional texts. For example, the text of the balanced budget provision for those constitutions with such a provision is here.
For our analysis of the effects of constitutional provisions on primary fiscal balances, we merge the CCP’s balanced budget rule (BBR) variable with an IMF dataset constructed to analyze the determinants of a country’s primary balance. We also include a measure from the CCP that captures the difficulty in amending a constitution to test the conditional effect that entrenching a provision matters.
The replication dataset for our article can be found on the Dataverse. In addition to the CCP variables, we include short-term and medium term real interest rates, gross public debt lagged by one year, public expenditures, and GDP growth, all as a percentage of GDP from an IMF dataset (Mauro et al. 2015). We also include two worldwide commodity price indexes from the MOxLAD data set, one for non-oil commodities and one for oil (Total non-oil and Oil). Finally, we include a measure of democracy from Polity IV, which ranges from 0 to 10, with higher values corresponding to higher levels of democracy. Once all independent variables are onto our CCP data, we have an unbalanced panel with historical coverage from 1950 to 2011 that includes 52 countries.
Built-In Flexibility?
Although many economists and policy-makers see a balanced budget amendment as overly restrictive, especially as the world emerges from the pandemic, our research provides a bit of nuance and a note of caution. As mentioned, we do find that BBRs reduce primary balances, but not equally across the board. In particular, we find that they work best when countries are closest to solvency, that is, when achieving a balanced budget is within reach. In such situations, austerity minded actors are better equipped to use these constitutional obligations to argue for belt-tightening measures. In more extreme debt scenarios, the influence of even a constitutional provision may be overwhelmed by political and economic forces. BBRs also aren’t necessarily binding during a crisis like COVID. In March 2020, the EU took the extraordinary step of suspending their fiscal rules, but our analysis suggests countries find room for fiscal maneuvering even absent a worldwide pandemic.
The case of Peru demonstrates the presence of this budgetary flexibility typical under a constitutional balanced budget requirement. Figure 2 plots the data from Peru, but other countries follow a similar trajectory. In Peru over the time period displayed, primary balances were quite volatile prior to the implementation of the BBR in 1979; they remained so following its implementation. It isn’t until about 1990 - about a decade after its implementation - that Peru consistently runs small surpluses, with the occasional minor deficit.
Figure 2: Primary balance before and after adoption a balanced budget for Peru from 1970 to 2011
Although Peru is but one case, this pattern is quite common in our data for other countries that adopted constitutional budgetary rules. BBRs do not act as a hard constraint on individual national budgets, so in any given year, there may be a surplus or deficit. Recalling this aspect of BBRs is relevant now, as policy-makers in Europe debate changing their fiscal rules again.
Why Flexibility is Inherent to BBRs
These data help put the current debate around the EU’s fiscal rules into perspective. While BBRs tend to work on average over time, they are not strictly binding in any given year. To think that they would be binding during a worldwide pandemic seems implausible. Technocrats do not wield that much power. In our paper, we propose a theory as to why this may be. We propose that constitutions function as much as coordination devices as they do contractual commitments. If the fox (executive) who guards the hen house cannot reliably police him- or herself, it is up to the hens (citizens or opposition elites) to do so. But citizens (and opposition elites) face costs in contesting executive transgressions. Interpreting rules poses high information costs and opposing power involves elements of personal and professional risk. As a result, individuals will not be inclined to publicly oppose a perceived transgression unless they believe that a significant number of others share their interpretation—hence, the need for coordination. Presumably, no individual will have any confidence in such a consensus unless rules are clear and their legitimacy widely shared. BBRs themselves are quite clear, simply stating that government receipts should cover government expenditures. And embedding such rules in a constitution, rather than treating them as statutory law or administrative procedure, likely elevates their legitimacy. Leaders anticipate opposition organized along constitutional lines and, in turn, refrain from testing those limits.
Does reform make sense?
Whatever one thinks about the advisability of balanced-budget rules, they seem to perform their function in balancing receipts and expenses – at least when you put the rule in the constitution. This appears to be true even after accounting for the possibility that only austerity minded governments adopt such measures, and hence are more prone to fiscal restraint even without a constitutional rule.
If the EU does decide to move forward with rules changes, however, they should not overreach. Full scale changes might be analogous to throwing the baby out with the bathwater. Might we suggest simply eliminating arbitrary rules over total debt levels. Recent suggestions of raising total debt to GDP limits from 60 percent to 100 percent seem to make little sense, replacing one arbitrary threshold with another. As noted in the opening paragraphs, many EU countries will still have a lot of work to do to even reach this threshold. Instead, keep things simple. A balanced budget rule that attempts to balance receipts and expenditures provides the long term path toward fiscal sustainability without tying any one country’s hands in the short term, should a new macroeconomic shock hit in the future.